Re: [ECOLOG-L] Hypothesis testing in ecology
Perhaps I missed it in this discussion, but I haven't seen any reference to differences in how easily hypotheses can be formulated in different cases. In studying a series of similar systems, such as a series of glacial lakes, it is easy to formulate hypotheses based on the idea that there are similarities between them. But when sending an ROV into a deep trench what is the basis for formulating a useful hypothesis? When it comes to hypothesis testing in ecology, I don't think that one standard can be applied to all systems. Bill Silvert
Re: [ECOLOG-L] Hypothesis testing in ecology
I've been meaning to comment here too. When I teach statistics, my goal is to give the graduate students a toolbox if you will, of useful ways to test ideas. More complex statistics comes later. In teaching, I use the idea of testing hypotheses, with a very important caveat. Both, null and alternative hypotheses have to be biologically sensible and biologically possible. I know I find many published papers that gloss over the null, but it turns out, on deeper inspection, that it was not a possibility and so refuting it was unavoidable. Apply that idea, that the null also must be reasonable, logical and possible, and you may find that many null hypotheses are none of those. Jim On Thu, Mar 10, 2011 at 13:13, Kevin Mueller kem...@psu.edu wrote: If we iteratively modify our hypotheses through the process of data collection, data analysis, or manuscript preparation, how different is this process from observational or exploratory research? It is, of course, different to some debatable extent. Regardless, I think Paul's comments shed light on the reality that there is a large gray area between the extremes of purely observational studies and purely hypothesis driven studies (which his 2005 paper apparently documents). Given this, I find the explicit or underlying claims of superiority made by proponents of hypothesis driven research to ring false (despite some of the strong benefits of hypothesis testing that Paul and others have made clear). I find such claims ironic since the result of many observational or exploratory studies is, gasp, a hypothesis. Finally, regardless of the language we use to reference hypotheses in our introductions, I ask: Is it always beneficial to cloak studies that are somewhat exploratory behind a veil of singlular hypothesis testing? Or might we also sometimes gain and share insights by making the process of data exploration and hypothesis testing/modification more apparent in our manuscripts? To be clear, my comments are more in response to a general narrow-mindedness that I've observed among some natural scientists, not to any particular post or 'poster' in this recent thread (i.e. I found Paul's post insightful and not especially narrow-minded). Kevin Mueller On Mar 9, 2011, at 11:00 PM, Paul Grogan wrote: Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output – the peer-reviewed publication disseminating the new knowledge – is as accurate and accessible to others as possible. As a result, I usually finish my manuscript Introduction sections with: “We used our data to test the following hypotheses” (rather than “We tested the following hypotheses... which gives the impression of great foresight on the part of the author).
Re: [ECOLOG-L] Hypothesis testing in ecology
Well-put! It would be great if people (particularly reviewers) always kept this in mind. --Ruchira On Sun, Mar 20, 2011 at 7:57 AM, James J. Roper jjro...@gmail.com wrote: I've been meaning to comment here too. When I teach statistics, my goal is to give the graduate students a toolbox if you will, of useful ways to test ideas. More complex statistics comes later. In teaching, I use the idea of testing hypotheses, with a very important caveat. Both, null and alternative hypotheses have to be biologically sensible and biologically possible. I know I find many published papers that gloss over the null, but it turns out, on deeper inspection, that it was not a possibility and so refuting it was unavoidable. Apply that idea, that the null also must be reasonable, logical and possible, and you may find that many null hypotheses are none of those. Jim On Thu, Mar 10, 2011 at 13:13, Kevin Mueller kem...@psu.edu wrote: If we iteratively modify our hypotheses through the process of data collection, data analysis, or manuscript preparation, how different is this process from observational or exploratory research? It is, of course, different to some debatable extent. Regardless, I think Paul's comments shed light on the reality that there is a large gray area between the extremes of purely observational studies and purely hypothesis driven studies (which his 2005 paper apparently documents). Given this, I find the explicit or underlying claims of superiority made by proponents of hypothesis driven research to ring false (despite some of the strong benefits of hypothesis testing that Paul and others have made clear). I find such claims ironic since the result of many observational or exploratory studies is, gasp, a hypothesis. Finally, regardless of the language we use to reference hypotheses in our introductions, I ask: Is it always beneficial to cloak studies that are somewhat exploratory behind a veil of singlular hypothesis testing? Or might we also sometimes gain and share insights by making the process of data exploration and hypothesis testing/modification more apparent in our manuscripts? To be clear, my comments are more in response to a general narrow-mindedness that I've observed among some natural scientists, not to any particular post or 'poster' in this recent thread (i.e. I found Paul's post insightful and not especially narrow-minded). Kevin Mueller On Mar 9, 2011, at 11:00 PM, Paul Grogan wrote: Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output – the peer-reviewed publication disseminating the new knowledge – is as accurate and accessible to others as possible. As a result, I usually finish my manuscript Introduction sections with: “We used our data to test the following hypotheses” (rather than “We tested the following hypotheses... which gives the impression of great foresight on the part of the author).
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear list members, My comments on (scientific) hypothesis testing have been based on the following definition of hypothesis: A hypothesis is a proposed explanation for an observable phenomenon. Clearly, any published article that mention the word hypothesis not necessarily was testing a true scientific hypothesis. Is it common to read: We tested the hypothesis that survival rate of male mountain lions was greater than female mountain lions in Yellowstone National Park. That is not a true hypothesis, there is no explanation involved. My questions now are (in the good spirit of a positive discussion): How much we know about what an hypothesis is? How much training on philosophy of science ecologists have? I am sending the link of an article where the authors propose to abandon the idea of testing hypothesis (http://www.cell.com/abstract/S0092-8674(08)00953-7). One of the author is a professor in the Department of Philosophy at Harvard University. The title of the article: A Brief History of the Hypothesis. They propose to work on questions and models. We propose that building hypotheses should be abandoned in favor of posing a straightforward question of a system and then receiving an answer, using that answer to model reality, and then testing the reproducibility and predictive power of the model, modifying it as necessary. It is better to see science as a quest for good questions to try to answer, rather than a quest for bold hypotheses to try to refute. Interestingly, in the article they mention that even Newton was not a follower of working with hypothesis. Best, Manuel -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology: Precision is what makes it valuable
Hi Kevin, this is great. Here's a link to a PDF of the article I wrote in the British Ecological Society journal http://post.queensu.ca/~groganp/Hypotheses%20in%20Ecology2foradobe.pdf. Figure 1 in particular may help in making more clear what I mean by iterative, and how one cycle feeds into other larger cycles that ultimately yields a product - a published piece of new knowledge (that could I think be derived from 'observational' or 'exploratory' research as you put it)...and ultimately that new knowledge leads to further new hypotheses. Cheers, Paul Paul Grogan Plant and Ecosystem Ecologist Dept. of Biology, Queen's University, Kingston, Ontario K7L 3N6, Canada. Phone: (613) 533 6152.Fax: (613) 533 6617. http://post.queensu.ca/~groganp/
Re: [ECOLOG-L] Hypothesis Testing in Ecology: Precision is what makes it valuable
Honorable Ecolog Forum: At the risk of repeating and repeating myself, I am once again going to cast my good sense and caution to the winds and confess that I have operated most of my life on the proposition that one (I) must go with the roughest guess that gets the job done (is demonstrably relevant) rather than endlessly confess that no (precise) conclusion could be reached, so more research (funding) will be needed before taking any conclusion seriously (if provisionally). I have also relied upon Raymond Gilmore's dictum that The suspension of judgment is the highest exercise in intellectual discipline. I have the gut feeling that intellectual enquiry that generates new understanding (screw knowledge) would not stand a snowball's chance in Hell of getting funding, largely because those with the purse strings are highly unlikely to take chances on something chancy. I would like to be wrong about this, and look forward to clear evidence which refutes this assertion/hypothesis. I strongly suspect that the proliferation of that elephant in the oikos, that dominant nest parasite, the yellow-bellied grantsucker, has just about wiped out that timid, hapless tinkerer, the wide-eyed naivete who keeps on the move, randomly meandering without direction. Still, how large a breeding population of unpopulars is necessary to maintain its viability? Or is it the result of a mutant gene that keeps popping up despite being edged out again and again? Is insecurity a problem? If so, will it be cured with Greek notation and infinite decimal points? How many variables can any research design handle? How many are there? It's relevance all the way down, and something like successive approximation all the way up. Or something like that? To clarify: This is not to deny the utility of hypotheses or statistics with in the realm of their relevance; it is only to suggest that they may not be the be-all and end-all of ecology. I can't prove any of this. I still stand in awe of Nature. Just not of committees. WT PS: I feel sorry for students who expected to succeed in ecology (let them eat MBA's [unfortunately they have to have them too]). But hey, they're getting something that might come in handy--adaptiveness and resilience, and the sexiest intellectual pursuit out there. - Original Message - From: Martin Meiss mme...@gmail.com To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Wednesday, March 09, 2011 3:32 PM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology: Precision is what makes it valuable Paul Grogan has stated very elegantly the case for a well formulated hypothesis, but I wish point out another aspect of the matter. People who are prospecting for iron will pass right over gold without seeing it. This is more than just a metaphor; it reflects how the human mind seems to work. That iterative process of refining the hypothesis can also be seen as selectively excluding opportunities for novel observations and discoveries. In a sense, one becomes progressively less open-minded. I don't mean that in the common pejorative sense, but I think it shows how there is still room for the researcher who naively makes observations and gathers data without specifically looking for anything in particular. Martin Meiss 2011/3/9 Paul Grogan grog...@queensu.ca Hi, I am fascinated by the varying use of hypotheses in ecology, and have been following the recent emails with great interest. All scientific research must presumably share a common goal to reach the highest attainable levels of precision in explicitly articulating the research focus, and the ensuing research results. For me, precise research hypotheses are the most effective means of achieving this goal. The most important components of an hypothesis are that it is novel and contains a testable prediction – An hypothesis is “a supposition made as a starting point for further investigation from known facts”. The process of initial hypothesis generation, literature review, methodological considerations, and further refinement (or even replacement) of the hypothesis is iterative, and may pass through several cycles before a novel, testable and precise hypothesis is reached. The efficiency of the subsequent processes of experimentation (or other approaches to testing such as modelling or surveying), data analyses, and write-up, is markedly enhanced by the a priori development of a clearly stated and focussed research hypothesis. Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output – the peer-reviewed publication disseminating the new knowledge – is as accurate
Re: [ECOLOG-L] Hypothesis testing in ecology
If we iteratively modify our hypotheses through the process of data collection, data analysis, or manuscript preparation, how different is this process from observational or exploratory research? It is, of course, different to some debatable extent. Regardless, I think Paul's comments shed light on the reality that there is a large gray area between the extremes of purely observational studies and purely hypothesis driven studies (which his 2005 paper apparently documents). Given this, I find the explicit or underlying claims of superiority made by proponents of hypothesis driven research to ring false (despite some of the strong benefits of hypothesis testing that Paul and others have made clear). I find such claims ironic since the result of many observational or exploratory studies is, gasp, a hypothesis. Finally, regardless of the language we use to reference hypotheses in our introductions, I ask: Is it always beneficial to cloak studies that are somewhat exploratory behind a veil of singlular hypothesis testing? Or might we also sometimes gain and share insights by making the process of data exploration and hypothesis testing/modification more apparent in our manuscripts? To be clear, my comments are more in response to a general narrow- mindedness that I've observed among some natural scientists, not to any particular post or 'poster' in this recent thread (i.e. I found Paul's post insightful and not especially narrow-minded). Kevin Mueller On Mar 9, 2011, at 11:00 PM, Paul Grogan wrote: Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output – the peer-reviewed publication disseminating the new knowledge – is as accurate and accessible to others as possible. As a result, I usually finish my manuscript Introduction sections with: “We used our data to test the following hypotheses” (rather than “We tested the following hypotheses... which gives the impression of great foresight on the part of the author).
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I contend that the majority of researches are NOT based on hypothesis testing. Every natural resource management agencies (Federal, state, and municipality) spends majority of their budget for data collection and monitoring to ensure that the focused natural resources are properly managed/protected. As such, the main research question is what current state of natural resources (e.g., abundance, distribution, mortality, growth, harvest, etc)? You don't need Null hypothesis testing to answer these questions. Instead, you would be using Bayesian statistics, rather than traditional frequentist null hypothesis testing statistics. And, yes, lots of grants are available, and many phD research projects were generated from those researches. Toshihide Hamachan Hamazaki, 濱崎俊秀PhD Alaska Department of Fish and Game Division of Commercial Fisheries 333 Raspberry Rd. Anchorage, AK 99518 Phone: (907)267-2158 Cell: (907)440-9934
Re: [ECOLOG-L] Hypothesis Testing in Ecology
this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To:ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries Wildlife 1 Rabbit Hill Road Westborough, MA 01581 508-389-6352fax 508-389-7891 http://www.nhesp.org - No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3487 - Release Date: 03/07/11 -- David McNeely William J. Resetarits, Jr. Professor Department of Biological Sciences Texas Tech University Lubbock, Texas 79409-3131 Phone: (806) 742-2710, ext.300 Fax (806) 742-2963 -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237
Re: [ECOLOG-L] Hypothesis Testing in Ecology
If you think Darwin's comment overstates the case, I recommend Michael Ghiselin's book, The Triumph of the Darwinian Method. He tested alternate hypotheses regularly as he gathered his observations. He did gather much information about many things and collected widely on many areas (e.g. coral reef formation), but his successes were conscious of a method that led to new insights. This view resonates with the Resetarits comment about which proposals in areas of biodiversity focused on collecting are most likely to get funded. Anthony Joern Professor of Biology Co-Director, Institute for Grassland Studies Kansas State University -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Jane Shevtsov Sent: Wednesday, March 09, 2011 2:43 AM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Darwin's comment is indeed famous, but let's not forget that it was made in a private letter in the context of defending the theory of evolution by natural selection. For that reason, it may well overstate the case. I'm no expert on Darwin, but I'm willing to guess (hypothesize?) that a good fraction of his observations of worms, barnacles, and South America were not initially made to support or refute any view, although they may well have been used that way later. Can anyone speak to this? Jane Shevtsov On Tue, Mar 8, 2011 at 6:51 AM, Hal Caswell hcasw...@whoi.edu wrote: People seem to be struggling over how to understand the value of observational research in the context of hypothesis-oriented discussions. One missing fact is that hypothesis-oriented research does not have to involve modern statistics, because scientific hypothesis-testing is not the same as statistical null hypothesis testing. Im surprised that no one has quoted Darwins perceptive comment about observational research (an activity in which he was an acknowledged master): How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service! (see http://www.darwinproject.ac.uk/entry-3257 for the entire letter, to H. Fawcett, 18 Sept. 1861) Hal Caswell On Mar 8, 2011, at 8:49 AM, Martin Meiss wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Is it not true that in attempting to say something about environmental influence on barnacle biology, Darwin realized he did not know enough about barnacles to use them as a model for his theories? Thus arose one of the most famous and definitive studies of any time about the morphology and biology of a large taxon. At least an old story makes that claim. mcneely Jane Shevtsov jane@gmail.com wrote: Darwin's comment is indeed famous, but let's not forget that it was made in a private letter in the context of defending the theory of evolution by natural selection. For that reason, it may well overstate the case. I'm no expert on Darwin, but I'm willing to guess (hypothesize?) that a good fraction of his observations of worms, barnacles, and South America were not initially made to support or refute any view, although they may well have been used that way later. Can anyone speak to this? Jane Shevtsov On Tue, Mar 8, 2011 at 6:51 AM, Hal Caswell hcasw...@whoi.edu wrote: People seem to be struggling over how to understand the value of observational research in the context of hypothesis-oriented discussions. One missing fact is that hypothesis-oriented research does not have to involve “modern statistics”, because scientific hypothesis-testing is not the same as statistical null hypothesis testing. I’m surprised that no one has quoted Darwin’s perceptive comment about observational research (an activity in which he was an acknowledged master): How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service!” (see http://www.darwinproject.ac.uk/entry-3257 for the entire letter, to H. Fawcett, 18 Sept. 1861) Hal Caswell On Mar 8, 2011, at 8:49 AM, Martin Meiss wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I
Re: [ECOLOG-L] Hypothesis Testing in Ecology
William and others, Personally, I think that the answer to the question Is all data gathering research? is clearly and unequivocally YES...just as I think this is not really the question you are addressing here. Instead, I think you are more properly asking Is all data gathering fundable research? (or perhaps Is all data gathering research that is useful for professional advancement?). For these latter questions, I think your comments are important and useful to keep in mind, for both students and professionals; however, I think your initial paragraph too broadly dismisses activities that are crucial to our understanding of nature. As but one example: I was recently reading a paper by Jerry Coyne et al (Evolution 2008) examining the origins of sexual dimorphism in birds. As their data, they used information on hybrids gathered from the literature. Now, my guess is that many of us (if we wanted) could use the original hybrid reports as an example of non-research data gathering, since on their own they really have no useful purpose other than just as a bit of information, perhaps only interesting to other ornithologists. But, with enough of these pieces out there, Coyne et al. were able to address an interesting theoretical question. As I learned early on, write down and record everything, as you never know what will be important later on. Chris *** Chris Brown Associate Professor Dept. of Biology, Box 5063 Tennessee Tech University Cookeville, TN 38505 email: cabr...@tntech.edu website: iweb.tntech.edu/cabrown -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Resetarits, William Sent: Tuesday, March 08, 2011 2:34 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology It seems a rather critical issue has raised its head at this juncture in the discussion. Is all data gathering research. I think we risk being disingenuous and misleading the many students on this listserve if we don't clearly and unequivocally answer NO. To suggetst hat the system is somehow faulty and that it is OK for folks, especially students, to follow their hearts and simply gather data on their favorite organisms or systems is doing them a grave disservice. One of the first, and undoubtedly the most important, thing I learned in my PhD. was also the most simple. The key question in any research project, whether empirical, experimental or theoretical, is... What's the question? Or as one of my committee members so eloquently put it, why should I care. The fact that no one knows anything about a particular taxon or a system, or I really like organism X is rarely an adequate answer. No one really doubts the absolute value of pure descriptive natural history, and data is a good thing, but it cannot realistically be an end in itself for a professional scientist in this day and age. Even the most storied present day natural historians, and those of the past as well, bring much more to the table. In any realistic funding climate, question driven science will, and should, take precedence. This does not mean that one can't do pure natural history in the context of question driven science, but it alone is unlikely to be sufficient to drive the research to the top of anyone's funding list, onto the pages of top journals, or to drive a candidate to the top of many job lists, at least at the PhD. level. Similarly, biodiversity discovery is important, ongoing, and it gets funded. Why? NSF's Program in Biotic Surveys and Inventories, recently expired programs in Microbial Observatories, and Microbial Inventories and Processes, and to some extent the ongoing Dimensions of Biodiversity program, among others, target biodiversity discovery. But all of them require well-framed questions that convince the target audience that THIS biodiversity discovery project should be funded over the 90% of those submitted that cannot be funded. The key is what else it brings to the table beyond just documenting what is out there. Most applied funding that allows for simple inventories and surveys is driven by economic and political considerations, not scientific. As valuable as it was for documenting the flora, fauna, ethnography, and geology of the American West, the Corps of Discovery expedition was NOT a scientific expedition but funded solely for economic and political purposes. Onl! y Jefferson's personal missive to gather data on plants, animals, Indian tribes etc., made it something beyond an exploration and mapping expedition. The actual science was done by others long after the Corps had returned. Similarly, naturalists (such as Darwin) were employed on commercial and exploratory voyages largely to bring back interesting, and more importantly, economically valuable plants and animals. Such was the case with the Beagle. We all admire Darwin as a natural historian, but that isn't why we
Re: [ECOLOG-L] Hypothesis Testing in Ecology: Precision is what makes it valuable
Hi, I am fascinated by the varying use of hypotheses in ecology, and have been following the recent emails with great interest. All scientific research must presumably share a common goal to reach the highest attainable levels of precision in explicitly articulating the research focus, and the ensuing research results. For me, precise research hypotheses are the most effective means of achieving this goal. The most important components of an hypothesis are that it is novel and contains a testable prediction An hypothesis is a supposition made as a starting point for further investigation from known facts. The process of initial hypothesis generation, literature review, methodological considerations, and further refinement (or even replacement) of the hypothesis is iterative, and may pass through several cycles before a novel, testable and precise hypothesis is reached. The efficiency of the subsequent processes of experimentation (or other approaches to testing such as modelling or surveying), data analyses, and write-up, is markedly enhanced by the a priori development of a clearly stated and focussed research hypothesis. Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output the peer-reviewed publication disseminating the new knowledge is as accurate and accessible to others as possible. As a result, I usually finish my manuscript Introduction sections with: We used our data to test the following hypotheses (rather than We tested the following hypotheses... which gives the impression of great foresight on the part of the author). I published results of a survey of ecological journals in 2005 which suggested that (in order of decreasing specificity and detail) only ~40% of papers contained explicit hypotheses, ~15% had questions, 25% had objectives, and the remainder had aims. Clearly not all ecologists are in agreement on the effectiveness of hypotheses. As suggested above, I agree with Manuels recent comment that questions, no matter how precise, are not the same as hypotheses (because the predictive element in the latter forces the researcher to APPLY the current knowledge). I also agree with Jane Shetsov in her comments yesterday that hypothesis-oriented research does not have to involve modern statistics, because scientific hypothesis-testing is not the same as statistical null hypothesis testing. The latter didactic approach may be useful to some ecologists, but multiple working hypotheses are more common in ecology. Furthermore, the next higher level putting ones questions and results in a meaningful ecological context is at least as important. This is the level that I try to work at. In any event, at whatever level they are used, what is most important is that the development and use of explicit hypotheses compels the researcher to be PRECISE in thought and language, and to focus on generating NEW knowledge It is the process that is most important. Paul Grogan (Dept. of Biology, Queen's University, Ontario, Canada)
Re: [ECOLOG-L] Hypothesis Testing in Ecology: Precision is what makes it valuable
Paul Grogan has stated very elegantly the case for a well formulated hypothesis, but I wish point out another aspect of the matter. People who are prospecting for iron will pass right over gold without seeing it. This is more than just a metaphor; it reflects how the human mind seems to work. That iterative process of refining the hypothesis can also be seen as selectively excluding opportunities for novel observations and discoveries. In a sense, one becomes progressively less open-minded. I don't mean that in the common pejorative sense, but I think it shows how there is still room for the researcher who naively makes observations and gathers data without specifically looking for anything in particular. Martin Meiss 2011/3/9 Paul Grogan grog...@queensu.ca Hi, I am fascinated by the varying use of hypotheses in ecology, and have been following the recent emails with great interest. All scientific research must presumably share a common goal to reach the highest attainable levels of precision in explicitly articulating the research focus, and the ensuing research results. For me, precise research hypotheses are the most effective means of achieving this goal. The most important components of an hypothesis are that it is novel and contains a testable prediction – An hypothesis is “a supposition made as a starting point for further investigation from known facts”. The process of initial hypothesis generation, literature review, methodological considerations, and further refinement (or even replacement) of the hypothesis is iterative, and may pass through several cycles before a novel, testable and precise hypothesis is reached. The efficiency of the subsequent processes of experimentation (or other approaches to testing such as modelling or surveying), data analyses, and write-up, is markedly enhanced by the a priori development of a clearly stated and focussed research hypothesis. Furthermore, often during the data interpretation or write-up stage, additional reflection on the processes of experimentation and evaluation of the data may indicate to the scientist (or to a manuscript reviewer) that the test did not reflect the hypothesis as well as originally thought. In such cases, further refinement or editing of the hypothesis statement should be made so that the final research output – the peer-reviewed publication disseminating the new knowledge – is as accurate and accessible to others as possible. As a result, I usually finish my manuscript Introduction sections with: “We used our data to test the following hypotheses” (rather than “We tested the following hypotheses... which gives the impression of great foresight on the part of the author). I published results of a survey of ecological journals in 2005 which suggested that (in order of decreasing specificity and detail) only ~40% of papers contained explicit ‘hypotheses’, ~15% had ‘questions’, 25% had ‘objectives’, and the remainder had ‘aims’. Clearly not all ecologists are in agreement on the effectiveness of hypotheses. As suggested above, I agree with Manuel’s recent comment that ‘questions’, no matter how precise, are not the same as hypotheses (because the predictive element in the latter forces the researcher to APPLY the current knowledge). I also agree with Jane Shetsov in her comments yesterday that “hypothesis-oriented research does not have to involve “modern statistics”, because scientific hypothesis-testing is not the same as statistical null hypothesis testing”. The latter didactic approach may be useful to some ecologists, but multiple working hypotheses are more common in ecology. Furthermore, the next higher level – putting one’s questions and results in a meaningful ecological context is at least as important. This is the level that I try to work at. In any event, at whatever level they are used, what is most important is that the development and use of explicit hypotheses compels the researcher to be PRECISE in thought and language, and to focus on generating NEW knowledge – It is the process that is most important. Paul Grogan (Dept. of Biology, Queen's University, Ontario, Canada)
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I have to agree with Christopher B. on his points. Stating unequivocally (if not dogmatically) that work that isn't hypothesis-driven is NOT research simply doesn't correspond to the meaning of research if we look it up in a dictionary. Granted, specialized fields such as ecology may redefine English words to suit their special purpose, but I am certainly not aware that consensus within ecology has emerged to justify such restricted usage. An earlier poster pointed out that if a granting agency only wants to fund hypothesis-driven research, one should heed that when applying for their funds. This is not a comment on the value of different approaches to research; it's just a pragmatic meeting of requirements. This touches on the fact, also addressed by Christopher B's comments, that science, by its nature, is not an individual enterprise. The knowledge base drawn upon, the resources made available, and the consequences of outcomes, all function at the level of large institutions or all of society. We are all drawing from a common pool and are all contributing to the pool. Funding agencies are a mechanism to evaluate and reward certain types of contributions, but people within a particular agency shouldn't imagine that their agency speaks for all of science. Some individuals, because of habit, training, temperament, and intellectual styles may wish to focus on one type of research (say, rigorous hypothesis testing) and others prefer another type, say exploration and data gathering without a-priori expectations. This diversity is good; let each individual function in the niche to which he/she is most suited or most enjoys. It's fine if some of us just publish observations if others of us can use those observations. It is up to higher-level control mechanisms (or an invisible hand, as in economics) to make the most use of these contributions, to bring together people and data that reinforce one another, and to provide nudges in useful directions. As has been pointed out by other posters, what is most valuable may change as a field or sub-field matures, or as society's needs change, but there's still room for everyone. I think this is especially true when we consider how new information technology can get more data before more people, even data that were gathered a hundred years ago. Martin Meiss 2011/3/9 Christopher Brown cabr...@tntech.edu William and others, Personally, I think that the answer to the question Is all data gathering research? is clearly and unequivocally YES...just as I think this is not really the question you are addressing here. Instead, I think you are more properly asking Is all data gathering fundable research? (or perhaps Is all data gathering research that is useful for professional advancement?). For these latter questions, I think your comments are important and useful to keep in mind, for both students and professionals; however, I think your initial paragraph too broadly dismisses activities that are crucial to our understanding of nature. As but one example: I was recently reading a paper by Jerry Coyne et al (Evolution 2008) examining the origins of sexual dimorphism in birds. As their data, they used information on hybrids gathered from the literature. Now, my guess is that many of us (if we wanted) could use the original hybrid reports as an example of non-research data gathering, since on their own they really have no useful purpose other than just as a bit of information, perhaps only interesting to other ornithologists. But, with enough of these pieces out there, Coyne et al. were able to address an interesting theoretical question. As I learned early on, write down and record everything, as you never know what will be important later on. Chris *** Chris Brown Associate Professor Dept. of Biology, Box 5063 Tennessee Tech University Cookeville, TN 38505 email: cabr...@tntech.edu website: iweb.tntech.edu/cabrown -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Resetarits, William Sent: Tuesday, March 08, 2011 2:34 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology It seems a rather critical issue has raised its head at this juncture in the discussion. Is all data gathering research. I think we risk being disingenuous and misleading the many students on this listserve if we don't clearly and unequivocally answer NO. To suggetst hat the system is somehow faulty and that it is OK for folks, especially students, to follow their hearts and simply gather data on their favorite organisms or systems is doing them a grave disservice. One of the first, and undoubtedly the most important, thing I learned in my PhD. was also the most simple. The key question in any research project, whether empirical, experimental or theoretical
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hola Manuel, Yours is a tough question, looking for specificity that is beyond what I looking for in grant proposals. When I reviewed grad student research proposals , I was looking for a statement that would demonstrate some thought about the proposed project and plans for analysis beyond the immediate results. I was not looking for a statistical hypothesis. I agree that, as you say, statement of an hypothesis can be easy and the difficulty is designing the test. Maybe, because I was dealing with student proposals, I was trying to teach the applicants that they needed to state what they thought was obvious and think about predictions to test what they were proposing to do. In English we also say, do as I say, not as I do. I hope that in my example, we weren't 100% guilty of the 'not as I do' part since I differentiate between grants intended to support research, and what I do at work, which isn't usually research to my way of thinking because we don't do much analysis of the results of any given inventory (from the office perspective, the main point is to know where rare species are in order to protect them). So, what do I consider a scientific hypothesis to be? For practical purposes I've looked for a statement of a question to be investigated and a discussion of how it is to be tested. You may well be right, that there isn't much carefully defined hypothesis testing in Ecology, but I think that it is useful to encourage attempts to approach that goal, and to try to do it oneself, if only to keep in mind that (inverting things) what we observe may have more than one cause or the cause that seems obvious may not be the operative one (using the multiple working hypotheses ideas). Saludos, Pat -- Pat Swain NHESP Community Ecologist From: Manuel Spínola [mailto:mspinol...@gmail.com] Sent: Monday, March 07, 2011 12:02 PM To: Swain, Pat (FWE) Cc: ECOLOG-L@listserv.umd.edu Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Dear Pat, But what do you consider a scientific hypothesis? Because the statement of an hypothesis could be easy, the difficult task is the logic of the study to test the hypothesis, something that you have to do with the predictions because you cannot test an hypothesis itself, but throught its predictions. My believe is that there is an illusion about hypothesis testing in Ecology. In spanish we say: Haz lo que yo digo pero no lo que yo hago (do what I say but not what I do). Most published articles on ecological journals are not about truly hypothesis testing. Best, Manuel 2011/3/7 Swain, Pat (FWE) pat.sw...@state.ma.usmailto:pat.sw...@state.ma.us Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic
Re: [ECOLOG-L] Hypothesis Testing in Ecology
People seem to be struggling over how to understand the value of observational research in the context of hypothesis-oriented discussions. One missing fact is that hypothesis-oriented research does not have to involve “modern statistics”, because scientific hypothesis-testing is not the same as statistical null hypothesis testing. I’m surprised that no one has quoted Darwin’s perceptive comment about observational research (an activity in which he was an acknowledged master): How odd it is that anyone should not see that all observation must be for or against some view if it is to be of any service!” (see http://www.darwinproject.ac.uk/entry-3257 for the entire letter, to H. Fawcett, 18 Sept. 1861) Hal Caswell On Mar 8, 2011, at 8:49 AM, Martin Meiss wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Ecology suffers from a surfeit or people who feel that if you don't do things their way it isn't right. One of the greatest events in marine ecology in my opinion was the discovery of abyssal vent communities fuelled by chemosynthesis. I have no idea what the funding proposal for this research was, but the key factor was that an ROV went to a new kind of location and just looked around. Some of the greatest discoveries in all fields af science involved stumbling across something totally unexpected, and certainly not hypothesized. Bill Silvert - Original Message - From: Martin Meiss mme...@gmail.com To: ECOLOG-L@LISTSERV.UMD.EDU Sent: terça-feira, 8 de Março de 2011 13:49 Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Actually, there is no reason why library research shouldn't be rigorously hypothesis driven: I wish to test the hypothesis that there is no article in the ecological literature on the incidence of frogs in lily ponds. Uf we find some articles that are almost there, say on the incidence of toads near water hazards, we can put wide confidence intervals on our graphs. Martin 2011/3/8 mcnee...@cox.net Martin Meiss mme...@gmail.com wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin Martin, I had the same response. I suppose that folks like John Wesley Powell could have cast hypotheses to cover their appeals for funding. Maybe T. Jefferson, M. Lewis, and W. Clark could have jointly written a grant proposal, stating as hypotheses that the Missouri River reached to the Rocky Mountains, that the Rocky Mountains were only as tall as the Appalachians, that there were rivers in the west that reached the Pacific Ocean, that there was an extant elephant species in the interior of North America, that Native Americans would be friendly and trade with the expedition, .. . Again, why? that Some things we just don't know, and collecting information toward finding out is a good thing. In some cases, the only legitimate question to ask is, What is there? Once we know that, then we can craft hypotheses about the what and the where. Now, so far as library work is concerned, surely you realize that one can craft excellent hypotheses that can be very effectively tested by examining data that have already been collected. Meta analysis has become an extremely important way to get answers in a wide range of fields. But you are right, exploration is research, hypothesis or no. Darwin did not set out around the world to test the hypothesis of common descent, or that of natural selection. He set out to see what was there (and to have an adventure rather than a pulpit). mcneely 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Martin Meiss mme...@gmail.com wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin Martin, I had the same response. I suppose that folks like John Wesley Powell could have cast hypotheses to cover their appeals for funding. Maybe T. Jefferson, M. Lewis, and W. Clark could have jointly written a grant proposal, stating as hypotheses that the Missouri River reached to the Rocky Mountains, that the Rocky Mountains were only as tall as the Appalachians, that there were rivers in the west that reached the Pacific Ocean, that there was an extant elephant species in the interior of North America, that Native Americans would be friendly and trade with the expedition, .. . Again, why? that Some things we just don't know, and collecting information toward finding out is a good thing. In some cases, the only legitimate question to ask is, What is there? Once we know that, then we can craft hypotheses about the what and the where. Now, so far as library work is concerned, surely you realize that one can craft excellent hypotheses that can be ver! y effectively tested by examining data that have already been collected. Meta analysis has become an extremely important way to get answers in a wide range of fields. But you are right, exploration is research, hypothesis or no. Darwin did not set out around the world to test the hypothesis of common descent, or that of natural selection. He set out to see what was there (and to have an adventure rather than a pulpit). mcneely 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I think Martin identifies one of the problems with a very restrictive definition of what science is - it excludes a bunch of stuff that most of us would think of as research. In fact, I would say that sequencing the human genome did not involve hypothesis testing - it was natural history at the molecular level - and most people would consider it one of the greatest scientific achievements of the last decade. However, I also have some sympathy with Pat's take that simply telling us what is there often has limited value. And if we think back to how this is 'supposed' to work (based on textbook science), descriptions are often the source of hypotheses that we should tyhen test. I would say at this point we have many, many untested or poorly tested hypotheses, which may explain why many scientists are not very supportive of work that will just provide more hypotheses to test. Best. Jeff Houlahan I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's not even research. (...I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research..., ...some of the projects that I rejected as not being research might well have been fundable ...)This appears to be defining the word research in a way I have never seen or heard before. Does this mean that none of the scientific work that was done before the rise of modern statistics was not research? Where the people doing that work also not really scientists? And whatever happened to library research? Martin 2011/3/7 Wayne Tyson landr...@cox.net Honorable Forum: Re: I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). --Pat Swain (Monday, March 07, 2011 6:03 AM) I don't want to appear to jump to conclusions, so I would be interested in Swain's expansions upon this issue. I wonder if Pat would have funded a survey which was based upon random sampling/mapping that would provide a baseline dataset and provide another level of scrutiny of the different/interesting as well as an opportunity to discover that which one's present state of knowledge might otherwise overlook. Please describe the theoretical foundation for walking the site rather than randomly sampling it, and how one approaches gaining knowledge of a site without a (statistically) valid inventory. WT - Original Message - From: Swain, Pat (FWE) pat.sw...@state.ma.us To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Monday, March 07, 2011 6:03 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Having reviewed dozens of research proposals for NSF, USGS, Delta Bay Authority, and Faculty Grant Programs at Universities I have rejected research w/o hypotheses stated when the program required hypotheses be stated. In two cases, the proposals set up LTREM sites in which data collection and hypothesis generation was the goal. Again, its much harder to get a grant through for exploratory research because confirmatory research has very obvious end-points demonstrating a product evolved from the funding. Exploratory research can go on for a long time w/o anything but data collection. On Mon, Mar 7, 2011 at 8:03 AM, Swain, Pat (FWE) pat.sw...@state.ma.uswrote: Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries Wildlife 1 Rabbit Hill Road Westborough, MA 01581 508-389-6352fax 508-389-7891 http://www.nhesp.org -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for
Re: [ECOLOG-L] Hypothesis Testing in Ecology
William and Honorable Ecolog Forum: The benefit of addressing those questions, or gathering data in the context of those questions, rather than simply plunging ahead with gathering more data, is that the answers to those questions can guide us to be more efficient in prioritizing what data we still need to gather with our limited time and resources. --William Resetarits (Ecolog, Tuesday, March 08, 2011 12:34 PM) What ARE those questions, and what ARE their priorities (and through what disciplined, universally supported process are they established)? WT - Original Message - From: Resetarits, William william.resetar...@ttu.edu To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Tuesday, March 08, 2011 12:34 PM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology It seems a rather critical issue has raised its head at this juncture in the discussion. Is all data gathering research. I think we risk being disingenuous and misleading the many students on this listserve if we don't clearly and unequivocally answer NO. To suggetst hat the system is somehow faulty and that it is OK for folks, especially students, to follow their hearts and simply gather data on their favorite organisms or systems is doing them a grave disservice. One of the first, and undoubtedly the most important, thing I learned in my PhD. was also the most simple. The key question in any research project, whether empirical, experimental or theoretical, is... What's the question? Or as one of my committee members so eloquently put it, why should I care. The fact that no one knows anything about a particular taxon or a system, or I really like organism X is rarely an adequate answer. No one really doubts the absolute value of pure descriptive natural history, and data is a good thing, but it cannot realistically be an end in itself for a professional scientist in this day and age. Even the most storied present day natural historians, and those of the past as well, bring much more to the table. In any realistic funding climate, question driven science will, and should, take precedence. This does not mean that one can't do pure natural history in the context of question driven science, but it alone is unlikely to be sufficient to drive the research to the top of anyone's funding list, onto the pages of top journals, or to drive a candidate to the top of many job lists, at least at the PhD. level. Similarly, biodiversity discovery is important, ongoing, and it gets funded. Why? NSF's Program in Biotic Surveys and Inventories, recently expired programs in Microbial Observatories, and Microbial Inventories and Processes, and to some extent the ongoing Dimensions of Biodiversity program, among others, target biodiversity discovery. But all of them require well-framed questions that convince the target audience that THIS biodiversity discovery project should be funded over the 90% of those submitted that cannot be funded. The key is what else it brings to the table beyond just documenting what is out there. Most applied funding that allows for simple inventories and surveys is driven by economic and political considerations, not scientific. As valuable as it was for documenting the flora, fauna, ethnography, and geology of the American West, the Corps of Discovery expedition was NOT a scientific expedition but funded solely for economic and political purposes. Onl! y Jefferson's personal missive to gather data on plants, animals, Indian tribes etc., made it something beyond an exploration and mapping expedition. The actual science was done by others long after the Corps had returned. Similarly, naturalists (such as Darwin) were employed on commercial and exploratory voyages largely to bring back interesting, and more importantly, economically valuable plants and animals. Such was the case with the Beagle. We all admire Darwin as a natural historian, but that isn't why we remember him and why he is on the British ten-pound note and voted the second most admired Brit in history (behind only Churchill - for very pragmatic reasons). Why the situation now is different is that he lived in a time when you had to expand the realm of natural history and systematic data both to generate and shed light on important questions. I agree with Jeff that we have a backlog of questions. The benefit of addressing those questions, or gathering data in the context of those questions, rather than simply plunging ahead with gathering more data, is that the answers to those questions can guide us to be more efficient in prioritizing what data we still need to gather with our limited time and resources. On 3/8/11 8:51 AM, David L. McNeely mcnee...@cox.net wrote: Martin Meiss mme...@gmail.com wrote: I am amazed by Pat Swain's statements implying that unless a program of work includes formal hypothesis testing, it's
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries Wildlife 1 Rabbit Hill Road Westborough, MA 01581 508-389-6352fax 508-389-7891 http://www.nhesp.org
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear Pat, But what do you consider a scientific hypothesis? Because the statement of an hypothesis could be easy, the difficult task is the logic of the study to test the hypothesis, something that you have to do with the predictions because you cannot test an hypothesis itself, but throught its predictions. My believe is that there is an illusion about hypothesis testing in Ecology. In spanish we say: Haz lo que yo digo pero no lo que yo hago (do what I say but not what I do). Most published articles on ecological journals are not about truly hypothesis testing. Best, Manuel 2011/3/7 Swain, Pat (FWE) pat.sw...@state.ma.us Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries Wildlife 1 Rabbit Hill Road Westborough, MA 01581 508-389-6352fax 508-389-7891 http://www.nhesp.org -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology
If a funding agency wants hypothesis driven research, then you give them hypothesis driven research! This varies form state to state, agency to agency. Getting exploratory research funded is not easy, its much easier to get funding for confirmatory research. Malcolm On Mon, Mar 7, 2011 at 8:03 AM, Swain, Pat (FWE) pat.sw...@state.ma.uswrote: Ecolog-L, Way back when the question about hypothesis testing in ecology was first posed to the group, one of the questions was whether anyone had rejected projects or grant proposals for lack of hypotheses. The discussion has gone on while I thought about posting a response to that, but with Jane Shevtsov's prodding, I offer the following thoughts on hypothesis testing and research. For some years I was on a committee to review and select graduate student research proposals for grant support for a regional botanical organization at the same time that I was involved in evaluating proposals for small contracts from my office which is focused on rare species and uncommon natural communities in the state. (I stress the research grants vs. contracts; and I am no longer on the committee which no doubt has different biases from mine, and my office doesn't have money for small contracts like we used to). On the grad research committee, I was far more likely to approve proposals for consideration if a hypothesis was stated, and I tended to veto projects that didn't do that. For example, I think that pure survey of a property for species (making a list of all the species of some taxonomic group) encountered isn't research, but such a project can be developed and proposed in ways that has research in it (effects of land use history, recreation, management...). If a student wanted to inventory a property as a research project, as someone funding grants I wanted the reasons given for why that property is worth the effort and what will be done with the results. I recall one otherwise quite good proposal I didn't consider because it just said that the property was interesting and the nonprofit owning it should know what was on it. I wanted to be shown what assumptions are being made (those should be stated as hypotheses to be tested in a proposal for a research grant), predictions! of where differences might be and why and expectations that post inventory analyses would be undertaken. However, some of the projects that I rejected as not being research might well have been fundable (I think some were) by my office where we want to know what rare species are in particular places, and what is rare. We have funded contracts for surveys for particular taxonomic groups in general as well others focused on rare species/natural communities along rivers, on particular properties, and so on. I think these general surveys are valuable, but they don't overtly involve hypotheses and testing. However, it can and does include assumptions/hypotheses; as one of the posters on the topic pointed out there are always assumptions made. One doesn't walk every square inch of a site, rather picks areas (from aerials, maps, knowledge, observations when out there) places that are most likely to be different/interesting (have rare things). So my thinking back when I was on the grad research committee was that for an inventory to be research and worth funding with a grant, the proposal had to clearly state hypotheses to be tested, and better, to discuss (yes, in only 2 pages) underlying assumptions going into the project. Maybe some of what I was after was an overt awareness of the questions and assumptions involved in setting up the project. And some idea of expected analysis of the results. My convoluted discussion summarizes to 'yes, I rejected proposals that didn't have hypotheses stated'. Pat -- Patricia Swain, Ph.D. Community Ecologist Natural Heritage Endangered Species Program Massachusetts Division of Fisheries Wildlife 1 Rabbit Hill Road Westborough, MA 01581 508-389-6352fax 508-389-7891 http://www.nhesp.org -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
An interesting thread. I read a book recently (at the age of 78 I do that more frequently), and I wondered if I had just forgotten what Iearned in school (Fresno State and UC Berkeley Mus Vert Zool) around 50-60 years ago. I'm pretty sure, though that I never heard much about Stephen Forbes, who wrote such papers as The ornithological balance wheel (1882, Trans Ill. St. Hort.Soc 15:120-31). I only heard his name in connection with the classic paper The Lake as a microcosm. (Bull of the Scientific Association (Peoria, I), 1887:77-87), but reading the biography by Robert Croker (Smithsonian Institution 2001) was a revelation to me. This man, raised in a pioneer family in the 1840s and who never received a college degree until he was awarded a PhD from U of Indiana on the basis of his research in several fields, did practical ecological work in terrestrial and aquatic ecology, and was a founding member and the second president of the Ecological Society of America. Of particular interest in relation to this thread is that he performed and directed studies of the Illinois River from the days when it was virtually undisturbed (1875) through the era of increased enclosure of the floodplain of the river by farmers who diked it as cropland, and of increasing pollution from Chicago before and after 1900, when the Chicago sanitary and ship canal was opened joining a tributary of the Illinois River to Lake Michigan, and on through the era of partial treatment of sewage after protests were raised by members of a productive commercial fishery which had existed on the river. He charted quantitatively the changes in numbers and species of fish populations, identified their food at different staqes of their life cycles, and measured populations of thee food items including macro-invertebrates, zooplankton, algae and benthic vegetation. He tackled the same issues with an ecosystem approach in bird and insect populations which were problems for farmers of tree and row crops. So anyhow, I was broadened, at this late stage, in my knowledge of when modern quantitative ecological work began to be done quantitative, and in particular, that there is a record of what happens to a river as we mess with it. The book has a complete bibliography. Mike On 11:59 AM, Marcus Ricci wrote: pre wrap I'll buy that, David. I do think that folks on the list are sometimes a little more concerned about definitions and differences, lumpers and dividers, than I am typically am. Folks may have noticed that, about 1/2-way into my post, it started petering out as I realized that there *was* a lot of similarity. Still, there is enough difference to warrant a different term, for me. However, I definitely agree with David's point about the evolution of our science. I agree that the development of technology and knowledge allow us to study things in different ways or more closely than we could have studied them tens, if not hundreds, of years ago. If folks agreed to amend the natural history definition to include and their interactions with the environment, I'd buy that. However, it sounds like many folks already include that, implicitly. Cheers, Marcus Marcus Ricci, M.S., CPESC 1301 Monroe Avenue Charleston, IL 61920 email: spotted_bluelt;atgt;hotmail.com A thing is right when it tends to preserve the integrity, stability and beauty of the biotic community. It is wrong when it tends otherwise. -- Aldo Leopold -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of David L. McNeely Sent: Thursday, March 03, 2011 10:57 AM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Marcus, with due respect, and I do respect your opinion and contributions: You are simply pointing out the evolution of our science, which now probes more deeply into the nature of nature than did those who did its work in earlier centuries. We evolved from describing the objects in nature to investigating how those objects interact with other parts of nature. It is still the study of nature and natural objects -- just additional things about them. A turtle's life history IS a part of how it interacts with environment. Ecology (or the less fancy name natural history) studies that. Maybe a different way of looking at it than yours, but still legitimate. I'm also not trying to say we should abandon the term ecology in favor of the older term natural history, though that would be intellectually defensible. It would also be nice if the general public could understand what our science is about, rather than confusing it with environmental activism (a legitimate endeavor in its own right). But enough of all this. The important thing is to know about turtles, including how turtles live and function, how other things relate to them, and how
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
I'll buy that, David. I do think that folks on the list are sometimes a little more concerned about definitions and differences, lumpers and dividers, than I am typically am. Folks may have noticed that, about 1/2-way into my post, it started petering out as I realized that there *was* a lot of similarity. Still, there is enough difference to warrant a different term, for me. However, I definitely agree with David's point about the evolution of our science. I agree that the development of technology and knowledge allow us to study things in different ways or more closely than we could have studied them tens, if not hundreds, of years ago. If folks agreed to amend the natural history definition to include and their interactions with the environment, I'd buy that. However, it sounds like many folks already include that, implicitly. Cheers, Marcus Marcus Ricci, M.S., CPESC 1301 Monroe Avenue Charleston, IL 61920 email: spotted_blueathotmail.com A thing is right when it tends to preserve the integrity, stability and beauty of the biotic community. It is wrong when it tends otherwise. -- Aldo Leopold -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of David L. McNeely Sent: Thursday, March 03, 2011 10:57 AM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Marcus, with due respect, and I do respect your opinion and contributions: You are simply pointing out the evolution of our science, which now probes more deeply into the nature of nature than did those who did its work in earlier centuries. We evolved from describing the objects in nature to investigating how those objects interact with other parts of nature. It is still the study of nature and natural objects -- just additional things about them. A turtle's life history IS a part of how it interacts with environment. Ecology (or the less fancy name natural history) studies that. Maybe a different way of looking at it than yours, but still legitimate. I'm also not trying to say we should abandon the term ecology in favor of the older term natural history, though that would be intellectually defensible. It would also be nice if the general public could understand what our science is about, rather than confusing it with environmental activism (a legitimate endeavor in its own right). But enough of all this. The important thing is to know about turtles, including how turtles live and function, how other things relate to them, and how they contribute to the overall state of nature. Too many people don't care. mcneely Marcus Ricci spotted_b...@hotmail.com wrote: I'd like to add my $0.02 because I disagree that ecology is simply a dressing up of natural history. Although I value natural history and historians, they are not studying the same things as ecologists. According to my Dictionary of Ecology, Evolution and Systematics, natural history is the study of nature, natural objects and natural phenomena. Ecology is the study of the *interrelationships* between living organisms and their environment (my emphasis). So, the former is the study of a subject or phenomena, the latter is the study of *how the subject interacts and relates to its environment.* Some may consider this the same definition, some may consider it parsing essentially the same definition. I consider them different definitions: one *focuses* on the turtle itself, what it eats, where it lives, how it reproduces. The other *focuses* on the place in the web that the turtle occupies, how its consumption of food or production of offspring effects the other occupants of its food web - either predators or competitors - and how the web would respond if a turtle population exploded or disappeared. Perhaps a little simplistic, but analogies work for me when definitions get too stickily close, which I will be the first to agree that these 2 do, when you start looking at them closely. Cheers, Marcus Marcus Ricci, M.S., CPESC Lake Decatur Watershed Specialist, Macon County SWCD 1301 Monroe Avenue Charleston, IL 61920 email: spotted_blueathotmail.com A thing is right when it tends to preserve the integrity, stability and beauty of the biotic community. It is wrong when it tends otherwise. -- Aldo Leopold -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Wayne Tyson Sent: Wednesday, March 02, 2011 3:21 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Thanks, David. Now I don't have to toss all my Darwin stuff into the dustbin. WT PS: David or others: Can you suggest any shortcuts to the best possible understanding of the pre-contact state
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Jeff's comments are good ones. I don't know why all the apostrophes came through as question marks, but maybe that's appropriate -- these are difficult issues and I, for one, have more questions than answers. On one hand, there are certainly examples that demonstrate that understanding is neither necessary nor sufficient for prediction. On the other hand, the two are certainly connected. It's been pointed out that causal knowledge, unlike statistical knowledge, enables us to predict how a system will behave under interventions. Maybe that helps -- I don't think you can understand a phenomenon without causal knowledge. Also, let's look at pedagogical questions. How do we ask students to demonstrate understanding of concepts? BTW, I want to clarify a remark I made earlier about chaos. While the long-term behavior of a system exhibiting chaotic behavior cannot be predicted in the sense that the time series can't be predicted, we CAN predict other aspects of its dynamics, such as the parameter values resulting in different modes of behavior. So maybe before we can productively discuss the relationship between prediction and understanding, we ought to clarify what we mean by prediction. How broadly or narrowly do we want to construe the term? Best, Jane On Thu, Mar 3, 2011 at 7:19 AM, Jeff Houlahan jeffh...@unb.ca wrote: In response to Jane?s comments ? I admit that understanding and prediction are not the same thing but they are much more closely related than most people appreciate, in my opinion. I would go so far as to say that prediction is a necessary if not sufficient condition of understanding. So while it is possible to predict without understanding (as in Jane?s Babylonian?s example ? although I knew nothing about the Babylonians and their ability to predict, I have no doubt that?s true) I think it is impossible to demonstrate understanding without prediction. In fact, I realized that I can?t come up with a definition of understanding that satisfies me without talking about prediction (none of the on-line definitions that I found worked very well for me). My definition of understanding would be ?The ability to make specific predictions based on a general description of how the world works.? Now, I guess it?s possible that somebody could understand how the world works but not be able to make any predictions but that means that they can?t demonstrate their understanding. In my opinion, understanding that can?t be demonstrated has little(no?) value because I can?t distinguish that person from all the people who claim they have understanding but have none. My above definition leaves room for ?thinking? you understand when you don?t, in situations where you make good predictions for the wrong reasons. But, even here prediction is critical because we will only detect our mistake when we try to make a new prediction and our ?understanding? leads us astray. That is, the only evidence of our mistake will be poor prediction. So, my original claim was not that understanding and prediction are the same thing but that understanding cannot be demonstrated without prediction. And predictions have to better than we would make by chance. And the only way to evaluate that is through some measure of probability/likelihood. Best. Jeff Houlahan PS I would be interested to hear any examples where understanding can be demonstrated without prediction. -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
One more thing: what predictions can you make if you understand what caused the extinction of the (non-avian) dinosaurs? Jane On Thu, Mar 3, 2011 at 7:19 AM, Jeff Houlahan jeffh...@unb.ca wrote: In response to Jane?s comments ? I admit that understanding and prediction are not the same thing but they are much more closely related than most people appreciate, in my opinion. I would go so far as to say that prediction is a necessary if not sufficient condition of understanding. So while it is possible to predict without understanding (as in Jane?s Babylonian?s example ? although I knew nothing about the Babylonians and their ability to predict, I have no doubt that?s true) I think it is impossible to demonstrate understanding without prediction. In fact, I realized that I can?t come up with a definition of understanding that satisfies me without talking about prediction (none of the on-line definitions that I found worked very well for me). My definition of understanding would be ?The ability to make specific predictions based on a general description of how the world works.? Now, I guess it?s possible that somebody could understand how the world works but not be able to make any predictions but that means that they can?t demonstrate their understanding. In my opinion, understanding that can?t be demonstrated has little(no?) value because I can?t distinguish that person from all the people who claim they have understanding but have none. My above definition leaves room for ?thinking? you understand when you don?t, in situations where you make good predictions for the wrong reasons. But, even here prediction is critical because we will only detect our mistake when we try to make a new prediction and our ?understanding? leads us astray. That is, the only evidence of our mistake will be poor prediction. So, my original claim was not that understanding and prediction are the same thing but that understanding cannot be demonstrated without prediction. And predictions have to better than we would make by chance. And the only way to evaluate that is through some measure of probability/likelihood. Best. Jeff Houlahan PS I would be interested to hear any examples where understanding can be demonstrated without prediction. -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hi Jane, great question about how to get students to demonstrate understanding. One I hadn't thought about. Off the top of my head I would say that they would demonstrate understanding if they knew what predictions a certain concept would make, which of those predictions had been tested, and how much evidence there was for the concept (that evidence would be in the form of accurate/precise predictions). As for causal knowledge that is a tricky one and one I have wrestled with because in some ways causes are like those Russian babushka dolls - causes ended up nested in causes. For example, if we're talking about causes for amphibian decline and I say low pH is one cause of declines, have I identified a cause? Somebody could say low pH is not the cause - it is the form of aluminum that is present at low pH. Somebody else could say it's not the amount of aluminum of that form that's in the water that is the cause, it is the aluminum binding to some receptor inside the frog that is the cause. So, in theory, pH could perfectly predict amphibian trends and I would then conclude we have perfect understanding but in fact there could be an explanation beneath pH where we know almost nothing. So, in my opinion, prediction is they only way to demonstrate understanding but understanding is often layered and perfect prediction at one level wouldn't necessarily imply understanding at another level. These are really difficult issues and I think it's easy to see them as esoteric and of not much practical concern but I think ecology has actually done a pretty poor job of quantifying our understanding and without that how do we know where to focus our resources? How well do we understand the effects of invasive species on the distribution and abundance of native organisms? I actually have no idea. Have we barely scratched the surface and we should direct more resources at that question? Do we almost have that question completely sorted out and so should move on to the next important question? Should we throw up our arms and say that this is an unanswerable question? Best. Jeff PS Sorry for straying so far from the original question Jeff's comments are good ones. I don't know why all the apostrophes came through as question marks, but maybe that's appropriate -- these are difficult issues and I, for one, have more questions than answers. On one hand, there are certainly examples that demonstrate that understanding is neither necessary nor sufficient for prediction. On the other hand, the two are certainly connected. It's been pointed out that causal knowledge, unlike statistical knowledge, enables us to predict how a system will behave under interventions. Maybe that helps -- I don't think you can understand a phenomenon without causal knowledge. Also, let's look at pedagogical questions. How do we ask students to demonstrate understanding of concepts? BTW, I want to clarify a remark I made earlier about chaos. While the long-term behavior of a system exhibiting chaotic behavior cannot be predicted in the sense that the time series can't be predicted, we CAN predict other aspects of its dynamics, such as the parameter values resulting in different modes of behavior. So maybe before we can productively discuss the relationship between prediction and understanding, we ought to clarify what we mean by prediction. How broadly or narrowly do we want to construe the term? Best, Jane On Thu, Mar 3, 2011 at 7:19 AM, Jeff Houlahan jeffh...@unb.ca wrote: In response to Jane?s comments ? I admit that understanding and prediction are not the same thing but they are much more closely related than most people appreciate, in my opinion. I would go so far as to say that prediction is a necessary if not sufficient condition of understanding. So while it is possible to predict without understanding (as in Jane?s Babylonian?s example ? although I knew nothing about the Babylonians and their ability to predict, I have no doubt that?s true) I think it is impossible to demonstrate understanding without prediction. In fact, I realized that I can?t come up with a definition of understanding that satisfies me without talking about prediction (none of the on-line definitions that I found worked very well for me). My definition of understanding would be ?The ability to make specific predictions based on a general description of how the world works.? Now, I guess it?s possible that somebody could understand how the world works but not be able to make any predictions but that means that they can?t demonstrate their understanding. In my opinion, understanding that can?t be demonstrated has little(no?) value because I can?t distinguish that person from all the people who claim they have understanding but have none. My above definition leaves room for ?thinking? you understand when you don?t, in situations where you make good predictions for the wrong reasons. But, even here
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear list members, As someone who has -been teaching model based stats to natural scientists for decades -has mastered the logic and arcana of Neyman-Pearson Decision Theoretic Hypothesis Testing (p-values) -routinely uses model based statistics and parameter estimation with conf intervals whenever possible -exposes students to the Nester collection of quotes I offer the following: -The Anderson book is well recommended. -At the same time, it is important that students understand NPDTHT, in order to understand and evaluate the great bulk of published work in ecology. -Teaching model based stats to students puts considerable demand on the student and it puts many of them between a rock (supervisor who adheres to NPDTHT) and a hard place (course in model based stats). -Rational treatment of uncertainty is a must in ecology. -NPDTHT proves nothing. It merely excludes chance (at some stated level of uncertainty) as an explanation for some observed result. David Schneider c/o Biology, Memorial University, St. John's NL http://www.elsevierdirect.com/ISBN/9780126278651/Quantitative-Ecology Quoting Manuel Spínola mspinol...@gmail.com: Dear list members, For those interested on statistical hypothesis testing, null hypothesis significant testing and p-values I would like to suggest the following web site with many quotes from many well known statisticians. http://warnercnr.colostate.edu/~anderson/nester.html and for new approaches on statistics (moving away from hypothesis testing and p-values) applied to ecology: Anderson, D. R. 2008. Model based inference in the life sciences. Springer, NY. Best, Manuel On 01/03/2011 12:46 p.m., Ruchira Datta wrote: To calculate p-values properly requires paying a lot of attention to how you choose the null hypothesis and whether it is really appropriate for your problem and the state of the art. I do not have a lot of experience in ecology, but in bioinformatics people often choose null hypotheses because they make the p-values easy to compute, or because everyone does it that way, or (more cynically) because they make their results appear significant. One can get a good p-value by choosing a null hypothesis that is almost certain to be wrong, regardless of the fact that the consensus was already that this null hypothesis was almost certain to be wrong before any of the reported experiments were undertaken. That doesn't mean the reported experiments advanced scientific understanding. --Ruchira On Tue, Mar 1, 2011 at 6:24 AM, Jeff Houlahanjeffh...@unb.ca wrote: Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Within this thread I was reminded of another situation related to publishing individual observations and the increasing use/creation of shared databases. Why are there so many requests to create shared databases? Because people have delegated individual observations and observations with small sample sizes to the trash. If the investigator is unable to devote 100% time to a project, or the project gets cut off due to funding or other circumstances, just toss it! It will reflect bad on you. The purpose of publication is not self glory, it is to communicate information so that it can be used by others to report/communicate their findings and allow us to further science. If every person who observed something odd published it, there would be a database exactly where it belongs, in publications. So, there really is a value to descriptive research and the years of ignoring and denigrating this important field has now come to an impasse as people needing the information are now calling out to others to dump said data into databases for others to use. No need. Publish the stuff in appropriate outlets and then it will again be available, put data in an appendix. Tadah! How completely obvious. Herpetological Review is a good example of one of these databases in herpetology, and the author gets credit for contributing. A note in HR is what it is, no one brags about it, but if you drag through there you can find piles in each issue and each one can be developed into a masters thesis or a major work. However, no one has time to work up every observation they make in the field into a major work. So, the authors of such notes are sharing information with the world. No one even claims that what is written there is representative of anything, it just happened. Some notes are single observations, some only comprise a small number of animals. But the info is there so others can investigate or use it. Its funny that this same system is used in clinical research listed under case accounts in medicine. No one seems to look down on those though. It really amazes me how we can look down our noses at the Chimney Sweep, and yet the day will come when our Chimney must be cleaned. How many realize that one of the most important papers, the description of DNA by Watson and Crick, was barely one column long with a pencil sketch and was essentially natural history. IT was definitely descriptive, and you could argue whether a hypothesis was really involved. As for statistical analysis? lets not go there. Malcolm On Sun, Feb 27, 2011 at 11:44 PM, Jane Shevtsov jane@gmail.com wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
Re: [ECOLOG-L] Hypothesis Testing in Ecology
On Wed, Mar 2, 2011 at 5:41 AM, Manuel Spínola mspinol...@gmail.com wrote: I think there is a confusion about hypothesis testing that Jane was referring to in the original post. We are moving away from her questions. Well, I was asking about both types of hypothesis testing. They're different things but strongly reinforce each other. Best, Jane On 01/03/2011 10:50 a.m., Matt Chew wrote: Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/ -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hi Matt, Since this conversation has gotten off to a solid start (but where's Wirt Atmar? I expected to hear more from our resident ex-physicist), I can now reveal more of my thoughts. Specifically, you've come near a very important point. Even natural history requires what may be called hypotheses or assumptions, but these are even more crucial in hypothesis testing. We have to make all kinds of auxiliary hypotheses (things like I identified these plants correctly or these animals move randomly over the landscape) in the course of testing a focal hypothesis. If the prediction derived from this hypothesis fails to come about, we have to figure out which hypothesis to blame. And that's absolutely deadly for falsificationism. I recommend an excellent essay called The 'Corroboration' of Theories by the philosopher Hilary Putnam. (Don't worry about the fact that it's philosophy -- it's actually far more readable than the average ecology paper.) It's not available online, but I'll be happy to send a PDF to anyone who asks. Jane Shevtsov On Tue, Mar 1, 2011 at 8:50 AM, Matt Chew anek...@gmail.com wrote: Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires 0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hi Jeff, Prediction and understanding are actually very different things and being good at one doesn't necessarily imply being good at the other. An example from the book _Foresight and Understanding_ by Stephen Toulmin: the Babylonians had no concept of the heliocentric solar system but they were quite good at predicting the movements of planets in the night sky. In fact, even after Newton, it took quite a while for astronomical tables based on a real understanding of the solar system to catch up to the accuracy of those made by the old method, which took no understanding at all. On the other hand, if a system exhibits chaotic behavior, long-term prediction is impossible -- but we can certainly understand the dynamics. Best, Jane Shevtsov On Tue, Mar 1, 2011 at 6:24 AM, Jeff Houlahan jeffh...@unb.ca wrote: Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Rudhira, I would definitely include the studies you describe as natural history and as ecology. I was somewhat tongue in cheek, and a little bit serious in my earlier post. To me, bottom line, ecology is an attempt to understand the nature (or history) of nature. In recent years we have refined the work to be more experimental, more quantitative, and to use hypothesis testing, but ultimately we are just trying to find out how nature works. That's what Charles Darwin was doing, when he observed, questioned, hypothesized, and yes, experimented. To me, the only significant distinction between modern science and its progenitors is the theoretical nature of recent times. Of course, our father Charles Darwin firmly established the theoretical nature of our science, and that was before the term ecology was created. More important than these disputations is the effort to understand nature, whatever name we give to that effort. But for those who would denigrate natural history, respect of the effort is important, also. For my part, I am proud to be a natural historian, naturalist, ecologist, whatever. I just wish that over the years I had been better at asking questions and deciphering nature to get answers. Darwin felt that observations have to be for or against something (his words)-- that hypothesis creation and testing, and ultimately theory development were the essence of our science. But he was called a naturalist, or natural historian. mcneely Ruchira Datta ruch...@berkeley.edu wrote: I think there might be a useful distinction between natural history and ecology, namely, the degree to which observations are replicated. With the phrase natural history there is no connotation or expectation that observations can be strictly replicated (this does not mean patterns cannot be found). In ecology, one might be able to replicate observations to a greater or lesser degree, but when aiming to do so one may have to regulate the environment in a somewhat unnatural way. E.g., one may take isolated soil samples or plants to the lab, strictly regulate the flow of nutrients and so forth, and observe what happens to the microbial communities. I feel it would be stretching the definition to call this natural history. On the other hand, one can also do observational studies in, e.g., wildlife ecology that are clearly part of natural history. To me, natural history would also include observations of abiotic geological processes that are not in themselves the subject of study of ecology, except insofar as they impact life. So it appears to me that while natural history and ecology certainly intersect, it may be useful to maintain some distinction between the terms. --Ruchira On Wed, Mar 2, 2011 at 10:27 AM, David L. McNeely mcnee...@cox.net wrote: Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian) -- David McNeely
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear list members, For those interested on statistical hypothesis testing, null hypothesis significant testing and p-values I would like to suggest the following web site with many quotes from many well known statisticians. http://warnercnr.colostate.edu/~anderson/nester.html and for new approaches on statistics (moving away from hypothesis testing and p-values) applied to ecology: Anderson, D. R. 2008. Model based inference in the life sciences. Springer, NY. Best, Manuel On 01/03/2011 12:46 p.m., Ruchira Datta wrote: To calculate p-values properly requires paying a lot of attention to how you choose the null hypothesis and whether it is really appropriate for your problem and the state of the art. I do not have a lot of experience in ecology, but in bioinformatics people often choose null hypotheses because they make the p-values easy to compute, or because everyone does it that way, or (more cynically) because they make their results appear significant. One can get a good p-value by choosing a null hypothesis that is almost certain to be wrong, regardless of the fact that the consensus was already that this null hypothesis was almost certain to be wrong before any of the reported experiments were undertaken. That doesn't mean the reported experiments advanced scientific understanding. --Ruchira On Tue, Mar 1, 2011 at 6:24 AM, Jeff Houlahanjeffh...@unb.ca wrote: Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear Malcom, I am not aware of that procedure. Can you explain a little bit more how this work? I would like to hear more on that because I teach statistics and I could consider teaching that procedure in my courses. What do you mean with complex regression analysis and a suite of statistics to evaluate the p-values? Best, Manuel On 02/03/2011 11:26 a.m., malcolm McCallum wrote: Much better result would be to find the probability for each of several statistical hypothesis given my data, but that is not possible through a p value. Yes it is possible. You use multiple hypotheses in complex regression analyses, and then you use a suite of statistics to evaluate the p-values. On Wed, Mar 2, 2011 at 7:41 AM, Manuel Spínolamspinol...@gmail.com wrote: Dear list members, I think there is a confusion about hypothesis testing that Jane was referring to in the original post. We are moving away from her questions. Hypothesis testing as some are considering in some posts is an aim of inferential statistics, is not the same as testing scientific hypothesis. Scientific hypothesis and statistical hypothesis are 2 different things. You can test a scientific hypothesis without the use of statistics. Regarding to the use and usefulness of p values you can find the following article interesting: Douglas J. Johnson. 1999. The insignificance of statistical significance testing. Journal of Wildlife Management, 63(3):763--772. As far as I know the p values is the probability of obtaining the observed results or more extreme results given that the specific hypothesis (usually a null a hypothesis) is true. Not a very interesting result in most observational studies. Best, Manuel On 01/03/2011 10:50 a.m., Matt Chew wrote: Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for BiologySociety PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de ríohttps://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIShttp://www.icomvis.una.ac.cr/ -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
I'd like to add my $0.02 because I disagree that ecology is simply a dressing up of natural history. Although I value natural history and historians, they are not studying the same things as ecologists. According to my Dictionary of Ecology, Evolution and Systematics, natural history is the study of nature, natural objects and natural phenomena. Ecology is the study of the *interrelationships* between living organisms and their environment (my emphasis). So, the former is the study of a subject or phenomena, the latter is the study of *how the subject interacts and relates to its environment.* Some may consider this the same definition, some may consider it parsing essentially the same definition. I consider them different definitions: one *focuses* on the turtle itself, what it eats, where it lives, how it reproduces. The other *focuses* on the place in the web that the turtle occupies, how its consumption of food or production of offspring effects the other occupants of its food web - either predators or competitors - and how the web would respond if a turtle population exploded or disappeared. Perhaps a little simplistic, but analogies work for me when definitions get too stickily close, which I will be the first to agree that these 2 do, when you start looking at them closely. Cheers, Marcus Marcus Ricci, M.S., CPESC Lake Decatur Watershed Specialist, Macon County SWCD 1301 Monroe Avenue Charleston, IL 61920 email: spotted_blueathotmail.com A thing is right when it tends to preserve the integrity, stability and beauty of the biotic community. It is wrong when it tends otherwise. -- Aldo Leopold -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Wayne Tyson Sent: Wednesday, March 02, 2011 3:21 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Thanks, David. Now I don't have to toss all my Darwin stuff into the dustbin. WT PS: David or others: Can you suggest any shortcuts to the best possible understanding of the pre-contact state of fishes and other aquatic/marine organisms/ecosystems in the New World (although I'm really interested in California, specifically coastal southern California streams and rivers)? I'm also interested in the best possible estimates of watersheds and stream hydrology for that period/region. Works that contrast the pre- and post-contact states and trends would do most of my work for me, which, given my increasing level of laziness, would be most welcome. For example, I am positing that some streams that are today intermittent or dependent upon urban runoff are quite different from their pre-contact states--some flowed all year, and hosted salmonid runs. (Ethnographic and historical [anecdotal] information [observations] references would be interesting, if not provable. A somewhat aside: Given the popularity of computer models, I'm wondering if any reconstruction of pre-contact climate and hydrology might have been done or in the works . . . It would seem that a program that could do this might be applicable anywhere. - Original Message - From: mcnee...@cox.net To: ECOLOG-L@LISTSERV.UMD.EDU; Wayne Tyson landr...@cox.net Sent: Wednesday, March 02, 2011 10:27 AM Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian) - No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3477 - Release Date: 03/02/11
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
I think there might be a useful distinction between natural history and ecology, namely, the degree to which observations are replicated. With the phrase natural history there is no connotation or expectation that observations can be strictly replicated (this does not mean patterns cannot be found). In ecology, one might be able to replicate observations to a greater or lesser degree, but when aiming to do so one may have to regulate the environment in a somewhat unnatural way. E.g., one may take isolated soil samples or plants to the lab, strictly regulate the flow of nutrients and so forth, and observe what happens to the microbial communities. I feel it would be stretching the definition to call this natural history. On the other hand, one can also do observational studies in, e.g., wildlife ecology that are clearly part of natural history. To me, natural history would also include observations of abiotic geological processes that are not in themselves the subject of study of ecology, except insofar as they impact life. So it appears to me that while natural history and ecology certainly intersect, it may be useful to maintain some distinction between the terms. --Ruchira On Wed, Mar 2, 2011 at 10:27 AM, David L. McNeely mcnee...@cox.net wrote: Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian)
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Marcus, with due respect, and I do respect your opinion and contributions: You are simply pointing out the evolution of our science, which now probes more deeply into the nature of nature than did those who did its work in earlier centuries. We evolved from describing the objects in nature to investigating how those objects interact with other parts of nature. It is still the study of nature and natural objects -- just additional things about them. A turtle's life history IS a part of how it interacts with environment. Ecology (or the less fancy name natural history) studies that. Maybe a different way of looking at it than yours, but still legitimate. I'm also not trying to say we should abandon the term ecology in favor of the older term natural history, though that would be intellectually defensible. It would also be nice if the general public could understand what our science is about, rather than confusing it with environmental activism (a legitimate endeavor in its own right). But enough of all this. The important thing is to know about turtles, including how turtles live and function, how other things relate to them, and how they contribute to the overall state of nature. Too many people don't care. mcneely Marcus Ricci spotted_b...@hotmail.com wrote: I'd like to add my $0.02 because I disagree that ecology is simply a dressing up of natural history. Although I value natural history and historians, they are not studying the same things as ecologists. According to my Dictionary of Ecology, Evolution and Systematics, natural history is the study of nature, natural objects and natural phenomena. Ecology is the study of the *interrelationships* between living organisms and their environment (my emphasis). So, the former is the study of a subject or phenomena, the latter is the study of *how the subject interacts and relates to its environment.* Some may consider this the same definition, some may consider it parsing essentially the same definition. I consider them different definitions: one *focuses* on the turtle itself, what it eats, where it lives, how it reproduces. The other *focuses* on the place in the web that the turtle occupies, how its consumption of food or production of offspring effects the other occupants of its food web - either predators or competitors - and how the web would respond if a turtle population exploded or disappeared. Perhaps a little simplistic, but analogies work for me when definitions get too stickily close, which I will be the first to agree that these 2 do, when you start looking at them closely. Cheers, Marcus Marcus Ricci, M.S., CPESC Lake Decatur Watershed Specialist, Macon County SWCD 1301 Monroe Avenue Charleston, IL 61920 email: spotted_blueathotmail.com A thing is right when it tends to preserve the integrity, stability and beauty of the biotic community. It is wrong when it tends otherwise. -- Aldo Leopold -Original Message- From: Ecological Society of America: grants, jobs, news [mailto:ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Wayne Tyson Sent: Wednesday, March 02, 2011 3:21 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Thanks, David. Now I don't have to toss all my Darwin stuff into the dustbin. WT PS: David or others: Can you suggest any shortcuts to the best possible understanding of the pre-contact state of fishes and other aquatic/marine organisms/ecosystems in the New World (although I'm really interested in California, specifically coastal southern California streams and rivers)? I'm also interested in the best possible estimates of watersheds and stream hydrology for that period/region. Works that contrast the pre- and post-contact states and trends would do most of my work for me, which, given my increasing level of laziness, would be most welcome. For example, I am positing that some streams that are today intermittent or dependent upon urban runoff are quite different from their pre-contact states--some flowed all year, and hosted salmonid runs. (Ethnographic and historical [anecdotal] information [observations] references would be interesting, if not provable. A somewhat aside: Given the popularity of computer models, I'm wondering if any reconstruction of pre-contact climate and hydrology might have been done or in the works . . . It would seem that a program that could do this might be applicable anywhere. - Original Message - From: mcnee...@cox.net To: ECOLOG-L@LISTSERV.UMD.EDU; Wayne Tyson landr...@cox.net Sent: Wednesday, March 02, 2011 10:27 AM Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
I think the common interpretation of natural history among ecologists could be called descriptive ecology. It has the tacit hypotheses Matt Chew listed, but I don't think people associate natural history with explicit hypothesis-testing. It's about collecting and describing observations that seem meaningful, and the observations are not made in order to test a clear, explicit model. While natural history is not explicitly hypothesis-driven, the observations collected in natural history are one basis for the formation of new hypotheses. Darwin didn't tromp around collecting barnacles to test the hypothesis of evolution by natural selection. He made and recorded careful observations, considered the patterns in those observations, and proposed his hypothesis to explain those patterns. Anyway, what distinguishes natural history from the rest of ecology is the lack of explicit hypotheses that the collected data are intended to address. Also, arguably, natural history extends to all fields of science; I would call a descriptive study of a nebula natural history, and Robert Hooke's study of cork cells was definitely natural history, but these studies would be in the fields of astronomy and plant anatomy, respectively. Jim Crants On Wed, Mar 2, 2011 at 12:27 PM, David L. McNeely mcnee...@cox.net wrote: Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian)
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Maybe this was true in the 19th century, and there are still some minor outlets where observational notes with limited or no replication is accepted so you can publish the kind of observations you suggest, but modern and mainstream natural history studies require huge sample sizes and extensive replication, often over long periods of time. Your suggestion that it is not the case is simply a misunderstanding that is popularly promoted. On Wed, Mar 2, 2011 at 2:09 PM, Ruchira Datta ruch...@berkeley.edu wrote: I think there might be a useful distinction between natural history and ecology, namely, the degree to which observations are replicated. With the phrase natural history there is no connotation or expectation that observations can be strictly replicated (this does not mean patterns cannot be found). In ecology, one might be able to replicate observations to a greater or lesser degree, but when aiming to do so one may have to regulate the environment in a somewhat unnatural way. E.g., one may take isolated soil samples or plants to the lab, strictly regulate the flow of nutrients and so forth, and observe what happens to the microbial communities. I feel it would be stretching the definition to call this natural history. On the other hand, one can also do observational studies in, e.g., wildlife ecology that are clearly part of natural history. To me, natural history would also include observations of abiotic geological processes that are not in themselves the subject of study of ecology, except insofar as they impact life. So it appears to me that while natural history and ecology certainly intersect, it may be useful to maintain some distinction between the terms. --Ruchira On Wed, Mar 2, 2011 at 10:27 AM, David L. McNeely mcnee...@cox.net wrote: Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian) -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
Re: [ECOLOG-L] Hypothesis Testing in Ecology
There is a paper by Halliday and Jaeger published in Herpetologica some years ago that is central to this discussion. It discusses the differences between explorative and confirmative research. Most scientists spend their time doing confirmative research, which is hypothesis testing. However, explorative research ends with the development of new hypotheses. When we examine much descriptive research it ends with proposed hypotheses in need of testing. Jaeger, R.G., and T.R. Halliday. 1998. On confirmatory versus exploratory research. Herpetologica 54(Suppl): S64-S66. It discusses the question very thoroughly and effectively. Its a must read for new graduate students!!! Malcolm On Wed, Mar 2, 2011 at 11:39 AM, Jane Shevtsov jane@gmail.com wrote: Hi Matt, Since this conversation has gotten off to a solid start (but where's Wirt Atmar? I expected to hear more from our resident ex-physicist), I can now reveal more of my thoughts. Specifically, you've come near a very important point. Even natural history requires what may be called hypotheses or assumptions, but these are even more crucial in hypothesis testing. We have to make all kinds of auxiliary hypotheses (things like I identified these plants correctly or these animals move randomly over the landscape) in the course of testing a focal hypothesis. If the prediction derived from this hypothesis fails to come about, we have to figure out which hypothesis to blame. And that's absolutely deadly for falsificationism. I recommend an excellent essay called The 'Corroboration' of Theories by the philosopher Hilary Putnam. (Don't worry about the fact that it's philosophy -- it's actually far more readable than the average ecology paper.) It's not available online, but I'll be happy to send a PDF to anyone who asks. Jane Shevtsov On Tue, Mar 1, 2011 at 8:50 AM, Matt Chew anek...@gmail.com wrote: Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires 0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
[ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Ecolog: What specifically distinguishes natural history from ecology? WT - Original Message - From: Matt Chew anek...@gmail.com To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Tuesday, March 01, 2011 8:50 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires 0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew - No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3475 - Release Date: 03/01/11
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Thanks, David. Now I don't have to toss all my Darwin stuff into the dustbin. WT PS: David or others: Can you suggest any shortcuts to the best possible understanding of the pre-contact state of fishes and other aquatic/marine organisms/ecosystems in the New World (although I'm really interested in California, specifically coastal southern California streams and rivers)? I'm also interested in the best possible estimates of watersheds and stream hydrology for that period/region. Works that contrast the pre- and post-contact states and trends would do most of my work for me, which, given my increasing level of laziness, would be most welcome. For example, I am positing that some streams that are today intermittent or dependent upon urban runoff are quite different from their pre-contact states--some flowed all year, and hosted salmonid runs. (Ethnographic and historical [anecdotal] information [observations] references would be interesting, if not provable. A somewhat aside: Given the popularity of computer models, I'm wondering if any reconstruction of pre-contact climate and hydrology might have been done or in the works . . . It would seem that a program that could do this might be applicable anywhere. - Original Message - From: mcnee...@cox.net To: ECOLOG-L@LISTSERV.UMD.EDU; Wayne Tyson landr...@cox.net Sent: Wednesday, March 02, 2011 10:27 AM Subject: Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? Wayne, Ernst Haeckel coined the term which became our modern term ecology. You probably knew this. Haeckel mistook the root of biological science, natural history, for one of its branches, ecology. Ever since, we have had this conundrum. Ecology is natural history dressed up to look better for those who have difficulty accepting that science is old and was effective in the old days. For those who have some sniffing hang-up about being natural historians, there is no more honorable, nor more interesting, endeavor than trying to figure out how nature works. And one doesn't have to be arrogant, or attempt to dismiss other's efforts, to do it effectively. David McNeely, fish ecologist (ie., natural historian) - No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3477 - Release Date: 03/02/11
Re: [ECOLOG-L] Question Ecology Natural History etc Re: [ECOLOG-L] Hypothesis Testing in Ecology
Nothing in modern studies; however, many contend that natural history is still done like back in the 19th century. If you look at those ancient papers you will find all kinds of pure speculation and things that simply could never be published today. Modern natural history, or life history studies involve tedious work and very large sample sizes of organisms. Old fashioned pre-modern studies might see a bird fly over and write about how the bird was angry or gleefully flying about. It is important to distinguish between what modern organismal biologists label as natural history and what natural history was in the day. Natural history tends to be organism focused instead of ecosystem focused. Some people will lump these studies under wildlife ecology, animal ecology, or even zoology. In any case, I hope that helps explain it. You will very often hear people discuss natural history in a negative light as if it is some kind of ancient field no longer practiced. In fact, they are not talking about modern life history studies but those approaches from early times. It is pretty interesting to note that life history ecologists are a dying breed in that many of these involving vertebrates require a life time to complete and that isn't well-suited to modern tenure guidelines! Also, they are difficult. I recall Hobart M. Smith once saying he didn't do life history because it was too hard! Natural history/life history is focused on the organism's ecological responses to the environment Its just a different approach to the same questions. On Tue, Mar 1, 2011 at 8:47 PM, Wayne Tyson landr...@cox.net wrote: Ecolog: What specifically distinguishes natural history from ecology? WT - Original Message - From: Matt Chew anek...@gmail.com To: ECOLOG-L@LISTSERV.UMD.EDU Sent: Tuesday, March 01, 2011 8:50 AM Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires 0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew - No virus found in this message. Checked by AVG - www.avg.com Version: 10.0.1204 / Virus Database: 1435/3475 - Release Date: 03/01/11 -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
Re: [ECOLOG-L] Hypothesis Testing in Ecology
To calculate p-values properly requires paying a lot of attention to how you choose the null hypothesis and whether it is really appropriate for your problem and the state of the art. I do not have a lot of experience in ecology, but in bioinformatics people often choose null hypotheses because they make the p-values easy to compute, or because everyone does it that way, or (more cynically) because they make their results appear significant. One can get a good p-value by choosing a null hypothesis that is almost certain to be wrong, regardless of the fact that the consensus was already that this null hypothesis was almost certain to be wrong before any of the reported experiments were undertaken. That doesn't mean the reported experiments advanced scientific understanding. --Ruchira On Tue, Mar 1, 2011 at 6:24 AM, Jeff Houlahan jeffh...@unb.ca wrote: Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Most of studies in ecology are observational not experimental. Perhaps, perhaps not. Depends on who you ask. The point is really moot - if your system is amenable to manipulation, a proper experiment is invaluable. Most field ecologists are dealing with systems that are not easily manupulated. Does this mean they can't do good science, because they are dealing with many intercorrelated gradients that complicate matters? I think not. As Nick Gotelli once told me when I was an impressionable PhD student, Galileo never moved a star. Kind of sums it up. Imagine testing the hypothesis of past competition on Darwin finches under the H-D method. Not sure what H-D is, may have missed it, but regarding Darwin's finches, I think what you are saying is that an experiment can't be conducted that will answer the question of competition past. True, but an accumulation of evidence from a variety of studies spanning behavioral ecology to phylogenetics gives us a pretty strong theory concerning their evolutionary ecology. Of course, something to the contrary could conceivably come up that might make biologists reconsider things. But it ain't happened yet. Pretty strong science I think. However, you might say, but the competition past question still stands... True, science does not provide time machines. I think you might be getting a little hung up on Popper and the role of statistics. Conceptually, I have no problem with Popper, but it's a very narrow view. Stats are useful, but they are just a means to an end and can be easily misused. Likwise, it's important to not confuse the stats with the underlying question! , which is often much more complex than a p-value. Chris ** Dr. Christopher Taylor Professor, Aquatic Ecology Department of Natural Resources Management Texas Tech University Lubbock, TX 79409 From: Ecological Society of America: grants, jobs, news [ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Manuel Spínola [mspinol...@gmail.com] Sent: Monday, February 28, 2011 8:49 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Hi Martin, If you state a scientific hypothesis you need to derive predictions from it, and sometimes you can state the predictions as statistical hypotheses, but not always, in fact, Karl Popper was not thinking on statistics or statistical hypotheses. As Malcolm McCallum said if you use statistics to test a scientific hypothesis I think you are in a more shaking ground, statistics has its own problems. By the way, statistics is inductive. Some people are using information theoretic approaches like AIC (Akaike Information Criterion) to work with what they believe are scientific hypothesis but I don't think so. They have a set of models but not necessarily a set of scientific hypotheses. Other problems of working with hypothesis in ecology are the multicausality of ecological phenomena and the limitation of conducting experiments at some spatial and time scales. Most of studies in ecology are observational not experimental. More on Popper. Karl Popper did not believe that the theory of evolution by natural selection was a scientific theory. He argument that you cannot falsify an hypothesis derived from that theory. How do you falsify that hypothesis using a critical experiment as the H-D followers call it? Best, Manuel On 28/02/2011 05:16 p.m., Martin Meiss wrote: I'm not sure I understand Manuel's distinction between statistical hypootheses and scientific hypothesis. Is not the former supposed in some way to mathematically embody/parameterize the latter? But in any case, it seems to me that it is often hard to rigorously formulate a null hypothesis and a corresponding working hypothesis. Suppose you hear an account where someone had a feeling of foreboding about his mother, only to discover later that just when he was having that feeling, his mother, thousands of miles away, had suddenly died. When people tell stories like this, it's often followed with a challenge, like you can't tell me that's just a coincidence! Well, I'd like to say it is a coincidence, but how could you test it? What is the expected number of times you should have a feeling of foreboding about your mother and she DOESN'T die? What is the expected number of times mothers should die without their sons/daughters having feelings of foreboding? How close to the actual time of death does the feeling of foreboding have to be before we can count it? How creepy does a feeling have to be before it reaches the threshold of genuine foreboding? Now, this doesn't sound very ecological, but I'll bet readers of this listserv can come up with examples from biology that approach this level of nebulosity. Here's my stab at it: How K-selected must an organism be before we say it is K-selected (or r-selected). How
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Popper changed his view about evolutionary theory a little later in his long career. He also noted that the course of scientific theory is like Darwinian evolution: mutations (theories) and natural selection. Patrick Foley bees, fleas, flowers, disease patfo...@csus.edu From: Ecological Society of America: grants, jobs, news [ECOLOG-L@LISTSERV.UMD.EDU] On Behalf Of Manuel Spínola [mspinol...@gmail.com] Sent: Monday, February 28, 2011 6:49 PM To: ECOLOG-L@LISTSERV.UMD.EDU Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology Hi Martin, If you state a scientific hypothesis you need to derive predictions from it, and sometimes you can state the predictions as statistical hypotheses, but not always, in fact, Karl Popper was not thinking on statistics or statistical hypotheses. As Malcolm McCallum said if you use statistics to test a scientific hypothesis I think you are in a more shaking ground, statistics has its own problems. By the way, statistics is inductive. Some people are using information theoretic approaches like AIC (Akaike Information Criterion) to work with what they believe are scientific hypothesis but I don't think so. They have a set of models but not necessarily a set of scientific hypotheses. Other problems of working with hypothesis in ecology are the multicausality of ecological phenomena and the limitation of conducting experiments at some spatial and time scales. Most of studies in ecology are observational not experimental. More on Popper. Karl Popper did not believe that the theory of evolution by natural selection was a scientific theory. He argument that you cannot falsify an hypothesis derived from that theory. Imagine testing the hypothesis of past competition on Darwin finches under the H-D method. How do you falsify that hypothesis using a critical experiment as the H-D followers call it? Best, Manuel On 28/02/2011 05:16 p.m., Martin Meiss wrote: I'm not sure I understand Manuel's distinction between statistical hypootheses and scientific hypothesis. Is not the former supposed in some way to mathematically embody/parameterize the latter? But in any case, it seems to me that it is often hard to rigorously formulate a null hypothesis and a corresponding working hypothesis. Suppose you hear an account where someone had a feeling of foreboding about his mother, only to discover later that just when he was having that feeling, his mother, thousands of miles away, had suddenly died. When people tell stories like this, it's often followed with a challenge, like you can't tell me that's just a coincidence! Well, I'd like to say it is a coincidence, but how could you test it? What is the expected number of times you should have a feeling of foreboding about your mother and she DOESN'T die? What is the expected number of times mothers should die without their sons/daughters having feelings of foreboding? How close to the actual time of death does the feeling of foreboding have to be before we can count it? How creepy does a feeling have to be before it reaches the threshold of genuine foreboding? Now, this doesn't sound very ecological, but I'll bet readers of this listserv can come up with examples from biology that approach this level of nebulosity. Here's my stab at it: How K-selected must an organism be before we say it is K-selected (or r-selected). How many factors in an environment must conduce to K-selection before we say it is a K-selecting environment? How many species in that environment must bear the earmarks of K-selection before we accept the hypothesis that it truly is a K-selecting environment? What about all the species in that environment that don't appear to be K-selected? I realize, of course, that different organism may be responding to different factors in the environment, and that we can get around some of these problems by defining a hypotheses sufficiently narrowly. However, the more narrowly we define the hypothesis, the less it tells us about nature because it is less generalizable, and I suppose that most researchers would like to come up with insights that are generalizable. I don't know if this relates to some of the problems that prompted Jane's query, but I'd love to see your thoughts on the matter. Martin M. Meiss 2011/2/28 Shermin dsshermi...@gmail.com I like Manuel's response. To answer Jane's other questions: 1. Does it help you do better science? It can, but not necessarily. See below. Is it crowding out other approaches? I'd like to hear more about this - what other systematic approaches are there? For example, anecdotal observations are generally discouraged, but sometimes anecdotal observations are valuable and should be a) reported and b) inspire further observation and/or experiment. E.g. observations of tool-use in animals in the wild are great example
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I think that while learning to study science, the hypothesis testing format provides a framework in which a student can easily think more deeply about the topic at hand and avoid presumptions of what will happen. I agree with your assessmen tof AIC and of ecology in general. We are all just describing what we see, but learning to do that systematically and to do it accurately and precisely requires good training, and hypothesis testing can do this very effectively. I guess what I'm saying is there is a time and a place for everything. Malcolm On Mon, Feb 28, 2011 at 8:49 PM, Manuel Spínola mspinol...@gmail.com wrote: Hi Martin, If you state a scientific hypothesis you need to derive predictions from it, and sometimes you can state the predictions as statistical hypotheses, but not always, in fact, Karl Popper was not thinking on statistics or statistical hypotheses. As Malcolm McCallum said if you use statistics to test a scientific hypothesis I think you are in a more shaking ground, statistics has its own problems. By the way, statistics is inductive. Some people are using information theoretic approaches like AIC (Akaike Information Criterion) to work with what they believe are scientific hypothesis but I don't think so. They have a set of models but not necessarily a set of scientific hypotheses. Other problems of working with hypothesis in ecology are the multicausality of ecological phenomena and the limitation of conducting experiments at some spatial and time scales. Most of studies in ecology are observational not experimental. More on Popper. Karl Popper did not believe that the theory of evolution by natural selection was a scientific theory. He argument that you cannot falsify an hypothesis derived from that theory. Imagine testing the hypothesis of past competition on Darwin finches under the H-D method. How do you falsify that hypothesis using a critical experiment as the H-D followers call it? Best, Manuel On 28/02/2011 05:16 p.m., Martin Meiss wrote: I'm not sure I understand Manuel's distinction between statistical hypootheses and scientific hypothesis. Is not the former supposed in some way to mathematically embody/parameterize the latter? But in any case, it seems to me that it is often hard to rigorously formulate a null hypothesis and a corresponding working hypothesis. Suppose you hear an account where someone had a feeling of foreboding about his mother, only to discover later that just when he was having that feeling, his mother, thousands of miles away, had suddenly died. When people tell stories like this, it's often followed with a challenge, like you can't tell me that's just a coincidence! Well, I'd like to say it is a coincidence, but how could you test it? What is the expected number of times you should have a feeling of foreboding about your mother and she DOESN'T die? What is the expected number of times mothers should die without their sons/daughters having feelings of foreboding? How close to the actual time of death does the feeling of foreboding have to be before we can count it? How creepy does a feeling have to be before it reaches the threshold of genuine foreboding? Now, this doesn't sound very ecological, but I'll bet readers of this listserv can come up with examples from biology that approach this level of nebulosity. Here's my stab at it: How K-selected must an organism be before we say it is K-selected (or r-selected). How many factors in an environment must conduce to K-selection before we say it is a K-selecting environment? How many species in that environment must bear the earmarks of K-selection before we accept the hypothesis that it truly is a K-selecting environment? What about all the species in that environment that don't appear to be K-selected? I realize, of course, that different organism may be responding to different factors in the environment, and that we can get around some of these problems by defining a hypotheses sufficiently narrowly. However, the more narrowly we define the hypothesis, the less it tells us about nature because it is less generalizable, and I suppose that most researchers would like to come up with insights that are generalizable. I don't know if this relates to some of the problems that prompted Jane's query, but I'd love to see your thoughts on the matter. Martin M. Meiss 2011/2/28 Shermin dsshermi...@gmail.com I like Manuel's response. To answer Jane's other questions: 1. Does it help you do better science? It can, but not necessarily. See below. Is it crowding out other approaches? I'd like to hear more about this - what other systematic approaches are there? For example, anecdotal observations are generally discouraged, but sometimes anecdotal observations are valuable and should be a) reported and b) inspire further observation and/or
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Ecology without hypotheses has been dismissed (sometimes derided) as natural history, but even natural history requires one hypothesis. Reporting an observation requires 0 confidence that an observation is meaningful, can be communicated, and can be interpreted. There are also tacit hypotheses inherent in scale, including the duration, extent and complexity of natural history observations. Hypothesis testing is a particularly relevant topic in US ecology at the moment because choices made in establishing the NEON program involve numerous hypotheses about ecosystem identity, composition, extent and location, the relevance of potential instrumentation and particular scales. However, the term 'hypothesis' is absent from NEON's website ( http://www.neoninc.org ). Explicit hypothesis testing done under NEON auspices will be subject to an array of tacit hypotheses, none of which have been articulated (or, it seems, even considered) by NEON's creators and promoters. Any supposedly non-hypothetical work conducted under NEON will face the same challenge. Matthew K Chew Assistant Research Professor Arizona State University School of Life Sciences ASU Center for Biology Society PO Box 873301 Tempe, AZ 85287-3301 USA Tel 480.965.8422 Fax 480.965.8330 mc...@asu.edu or anek...@gmail.com http://cbs.asu.edu/people/profiles/chew.php http://asu.academia.edu/MattChew
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I wonder if you mean Alpha values? p-values are what they are and only tell you how likely the relationship expressed in your statistical test is to be repeated. The alpha value is an apriori set value used in decision theory. If you set alpha to 0.05, then you state that if the p /= 0.05bla bla. Is this what you were meaning? I didn't so far notice anyone arguing p-values. But, maybe I missed an email. On Tue, Mar 1, 2011 at 8:24 AM, Jeff Houlahan jeffh...@unb.ca wrote: Hi Chris and all, I actually think that it's a mistake to diminish the role of p-values. My opinion on this (stongly influenced by the writings of Rob Peters) is that there is only one way to demonstrate understanding and that is through prediction. And predictions only demonstrate understanding if you make better predictions than you would make strictly by chance. The only way to tell if you've done better than chance is through p-values. So, while there is a great deal more to science than p-values, the ultimate tests of whether science has led to increased understanding are p-values. Best. Jeff Houlahan Dept of Biology 100 Tucker Park Road UNB Saint John -- Malcolm L. McCallum Managing Editor, Herpetological Conservation and Biology Peer pressure is designed to contain anyone with a sense of drive - Allan Nation 1880's: There's lots of good fish in the sea W.S. Gilbert 1990's: Many fish stocks depleted due to overfishing, habitat loss, and pollution. 2000: Marine reserves, ecosystem restoration, and pollution reduction MAY help restore populations. 2022: Soylent Green is People! Confidentiality Notice: This e-mail message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized review, use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message.
[ECOLOG-L] Hypothesis Testing in Ecology
Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear Jane, That is a topic that have interested me for a long time. I teach something of this in my classes to master students in wildlife management and conservation here in Costa Rica. I know this is a controversial issue. First I recommend these 3 books: Scientific Method for Ecological Research. E. David Ford. Method in Ecology: Strategies for Conservation. Kristin S. Shrader-Frechette and Earl D. McCoy A Primer on Natural Resource Science. Fred S. Guthery Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. Like you probably experienced, reviewers like to see hypothesis driven research on the proposal that you submit but most of the time they do not know what a true scientific hypothesis is. Most research in ecology is not hypothesis driven, even when would like to see that. Read any paper in ecological journals and see how many of them are truly hypothesis driven. Hypothesis driven research are not always possible and in many instances is not necessary to have scientific hypothesis, all depend on the context. Most of the time we are interested in parameter estimation on how much a factor or covariable influence a parameter of interest. Besides, If you are going to do hypothesis driven research you need to work with multiple hypothesis (Chamberlin). Falsification is the contribution of Karl Popper to the Hypothetic-Deductive method. It has nothing to do with statistics or statistical hypothesis. The hypothetic-deductive method has been considered as the scientific method, however not many people know how it works. The hypothetic-deductive method is inductive and not deductive like the namesuggest. There is no a superior approach to obtain scientific knowledge. There are much more on this topic but I would like to see other opinions. Best, Manuel Spínola On 27/02/2011 11:44 p.m., Jane Shevtsov wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Manuel Spínola mspinol...@gmail.com wrote: There is no a superior approach to obtain scientific knowledge. My dissertation advisor said on more than one occasion that, The scientific method is doing whatever is necessary to get good answers to questions worth asking. I don't think that was original with him. mcneely
Re: [ECOLOG-L] Hypothesis Testing in Ecology
My experience: During my dissertation proposal defense, I was surprised by this debate coming up. I had listed my hypotheses within a powerpoint presentation as more or less statistical hypotheses rather than explanatory ecological hypotheses. As a 'green' PhD student, I was somewhat aware of the difference, but not aware of the debate. One of my committee members stopped me, declaring that is not a hypothesis. Another committee member replied with yes, it is. I sat back and let them debate it for a while, getting more and more freaked out that I wasn't going to pass my defense. In the end, I got a 'provisional pass' with the expectation that I convert the framework of my proposal in to more expanatory, hypothesis-testing style. I re-wrote - not changing my project at all, just how I presented it - and then I was good to go. I gave this a lot of thought as I moved forward and believe that part of this debate is really about descriptive, natural history work vs hypothesis-testing, rather than statistical vs ecological hypotheses (but I do agree with Sr Spinola that both are worthwhile discussions). In most cases, the natural history has to be done first in order to produce viable hypotheses. Unfortunately, this is not seen as 'dissertation-worthy' by many institutions and limits what studies can get funded and finished. In my case, I simply did the natural history stuff simultaneously and it worked out well. So, in my teaching I discuss the two different types of hypotheses and the importance of natural history to inform viable hypothesis development. On Mon, Feb 28, 2011 at 10:49 AM, Manuel Spínola mspinol...@gmail.comwrote: Dear Jane, That is a topic that have interested me for a long time. I teach something of this in my classes to master students in wildlife management and conservation here in Costa Rica. I know this is a controversial issue. First I recommend these 3 books: Scientific Method for Ecological Research. E. David Ford. Method in Ecology: Strategies for Conservation. Kristin S. Shrader-Frechette and Earl D. McCoy A Primer on Natural Resource Science. Fred S. Guthery Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. Like you probably experienced, reviewers like to see hypothesis driven research on the proposal that you submit but most of the time they do not know what a true scientific hypothesis is. Most research in ecology is not hypothesis driven, even when would like to see that. Read any paper in ecological journals and see how many of them are truly hypothesis driven. Hypothesis driven research are not always possible and in many instances is not necessary to have scientific hypothesis, all depend on the context. Most of the time we are interested in parameter estimation on how much a factor or covariable influence a parameter of interest. Besides, If you are going to do hypothesis driven research you need to work with multiple hypothesis (Chamberlin). Falsification is the contribution of Karl Popper to the Hypothetic-Deductive method. It has nothing to do with statistics or statistical hypothesis. The hypothetic-deductive method has been considered as the scientific method, however not many people know how it works. The hypothetic-deductive method is inductive and not deductive like the namesuggest. There is no a superior approach to obtain scientific knowledge. There are much more on this topic but I would like to see other opinions. Best, Manuel Spínola On 27/02/2011 11:44 p.m., Jane Shevtsov wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/ -- Dr. Elizabeth Congdon Biology Department Georgia
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I was delighted to read this. My background is physics, and I always interpreted Einstein's comment that Nature is subtle but not malicious to mean that we might have to be very devious in discovering her secrets. Sticking to a rigorous scientific method and fussing about hypothesis testing will not get us very far. Bill Silvert PS - I will get some flack from those who recall that Einstein actually said Raffiniert ist der Herrgott, aber boshaft ist er nicht. If you prefer The Lord is subtle but not malicious that is fine with me, I don't confound science with religion. - Original Message - From: David L. McNeely mcnee...@cox.net To: ECOLOG-L@LISTSERV.UMD.EDU Sent: segunda-feira, 28 de Fevereiro de 2011 16:39 Subject: Re: [ECOLOG-L] Hypothesis Testing in Ecology My dissertation advisor said on more than one occasion that, The scientific method is doing whatever is necessary to get good answers to questions worth asking. I don't think that was original with him. mcneely
Re: [ECOLOG-L] Hypothesis Testing in Ecology
This may seem like blatant self-promotion, but I think that the best synthesis of the interplay between natural history, experimentation, and theory is Earl Werner's chapter in our Experimental Ecology book from 1998 (Werner, E.E. 1998. Ecological experiments and a research program in community ecology. Pages 3-26 in W. J. Resetarits Jr. and J. Bernardo, eds. Experimental Ecology: Issues and Perspectives. Oxford, New York). I don't have a pdf, but there are a fair number of books floating around so it shouldn't be too hard to round up a copy of the paper. The plan was to write an opening chapter for the book, but once I received Earl's contribution I knew that anything we might say by way of introduction would be anticlimactic. So, my main contribution was recognizing a good thing when I read it. I feel confident in recommending this as many of my younger colleagues have told me that they used it as a guide while working on their dissertations (along with other chapters) and continue to use is as a framework for guiding their research programs. While it doesn't deal directly with the philosophical issues underlying the original subject of the post, it relates to the general issue raised by Elizabeth and the more pragmatic concerns, and is certainly germane to anyone planning their research and hoping to maximize the long-term impact of that research, its acceptance in high impact journals, and the short and long term probabilities of having that research funded. The one caveat I will add is that the enterprise as laid out can appear quite daunting - there are few individuals who can excel at all three aspects of a such a program - Earl is undoubtedly unique in that regard. We all have our relative strengths and weaknesses, so the point to keep in mind is that one doesn't necessarily have to DO everything themselves - natural history, experimentation and theory - but MUST bring an awareness of all three to bear in generating and attempting to answer ecological questions. I would venture to say that ANY ecological question (writ large), whether basic or applied, can inform or be informed by ecological theory and the broader ecological questions at the heart of our science. On 2/28/11 10:41 AM, Elizabeth Congdon congdo...@gmail.com wrote: My experience: During my dissertation proposal defense, I was surprised by this debate coming up. I had listed my hypotheses within a powerpoint presentation as more or less statistical hypotheses rather than explanatory ecological hypotheses. As a 'green' PhD student, I was somewhat aware of the difference, but not aware of the debate. One of my committee members stopped me, declaring that is not a hypothesis. Another committee member replied with yes, it is. I sat back and let them debate it for a while, getting more and more freaked out that I wasn't going to pass my defense. In the end, I got a 'provisional pass' with the expectation that I convert the framework of my proposal in to more expanatory, hypothesis-testing style. I re-wrote - not changing my project at all, just how I presented it - and then I was good to go. I gave this a lot of thought as I moved forward and believe that part of this debate is really about descriptive, natural history work vs hypothesis-testing, rather than statistical vs ecological hypotheses (but I do agree with Sr Spinola that both are worthwhile discussions). In most cases, the natural history has to be done first in order to produce viable hypotheses. Unfortunately, this is not seen as 'dissertation-worthy' by many institutions and limits what studies can get funded and finished. In my case, I simply did the natural history stuff simultaneously and it worked out well. So, in my teaching I discuss the two different types of hypotheses and the importance of natural history to inform viable hypothesis development. On Mon, Feb 28, 2011 at 10:49 AM, Manuel Spínola mspinol...@gmail.comwrote: Dear Jane, That is a topic that have interested me for a long time. I teach something of this in my classes to master students in wildlife management and conservation here in Costa Rica. I know this is a controversial issue. First I recommend these 3 books: Scientific Method for Ecological Research. E. David Ford. Method in Ecology: Strategies for Conservation. Kristin S. Shrader-Frechette and Earl D. McCoy A Primer on Natural Resource Science. Fred S. Guthery Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. Like you probably experienced, reviewers like to see hypothesis driven research on the proposal that you submit but most of the time they do not know what a true scientific hypothesis is. Most research in ecology is
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I like Manuel's response. To answer Jane's other questions: 1. Does it help you do better science? It can, but not necessarily. See below. Is it crowding out other approaches? I'd like to hear more about this - what other systematic approaches are there? For example, anecdotal observations are generally discouraged, but sometimes anecdotal observations are valuable and should be a) reported and b) inspire further observation and/or experiment. E.g. observations of tool-use in animals in the wild are great example of spontaneous events that one can never set out to observe systematically (except in controlled lab settings) but are nonetheless highly informative. I also wonder about replication - the larger or longer the scale (e.g. ecosystem, biome/longitudinal studies) the harder it is to replicate. This gets at Manuel's distinction about statistical vs. scientific hypotheses. You might have a hypothesis about a process but observe outcomes that are inherently difficult to attach a p-value to or find multiple examples of. Thoughts on that? Finally there's the issue of taxonomic poverty. Hypotheses about clades with few species are more difficult to test than those with a greater number of species. A problem if you're interested in the species-poor clade for other reasons. I.e. there is a trend towards choosing your species/system of study based on your questions of interest, and lately I've heard many talks that begin We chose to study species X because it is an excellent model for testing Y... What if you simply want to know about species X for no other reason than that you want to know about it? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? Yes, and I'm wondering about this trend in the stated aims of some journals as well. -- Shermin de Silva, Ph.D http://elephantresearch.net/fieldnotes http://www.sas.upenn.edu/~sdesilva On Mon, Feb 28, 2011 at 10:49 AM, Manuel Spínola mspinol...@gmail.comwrote: Dear Jane, That is a topic that have interested me for a long time. I teach something of this in my classes to master students in wildlife management and conservation here in Costa Rica. I know this is a controversial issue. First I recommend these 3 books: Scientific Method for Ecological Research. E. David Ford. Method in Ecology: Strategies for Conservation. Kristin S. Shrader-Frechette and Earl D. McCoy A Primer on Natural Resource Science. Fred S. Guthery Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. Like you probably experienced, reviewers like to see hypothesis driven research on the proposal that you submit but most of the time they do not know what a true scientific hypothesis is. Most research in ecology is not hypothesis driven, even when would like to see that. Read any paper in ecological journals and see how many of them are truly hypothesis driven. Hypothesis driven research are not always possible and in many instances is not necessary to have scientific hypothesis, all depend on the context. Most of the time we are interested in parameter estimation on how much a factor or covariable influence a parameter of interest. Besides, If you are going to do hypothesis driven research you need to work with multiple hypothesis (Chamberlin). Falsification is the contribution of Karl Popper to the Hypothetic-Deductive method. It has nothing to do with statistics or statistical hypothesis. The hypothetic-deductive method has been considered as the scientific method, however not many people know how it works. The hypothetic-deductive method is inductive and not deductive like the namesuggest. There is no a superior approach to obtain scientific knowledge. There are much more on this topic but I would like to see other opinions. Best, Manuel Spínola On 27/02/2011 11:44 p.m., Jane Shevtsov wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre
Re: [ECOLOG-L] Hypothesis Testing in Ecology
I'm not sure I understand Manuel's distinction between statistical hypootheses and scientific hypothesis. Is not the former supposed in some way to mathematically embody/parameterize the latter? But in any case, it seems to me that it is often hard to rigorously formulate a null hypothesis and a corresponding working hypothesis. Suppose you hear an account where someone had a feeling of foreboding about his mother, only to discover later that just when he was having that feeling, his mother, thousands of miles away, had suddenly died. When people tell stories like this, it's often followed with a challenge, like you can't tell me that's just a coincidence! Well, I'd like to say it is a coincidence, but how could you test it? What is the expected number of times you should have a feeling of foreboding about your mother and she DOESN'T die? What is the expected number of times mothers should die without their sons/daughters having feelings of foreboding? How close to the actual time of death does the feeling of foreboding have to be before we can count it? How creepy does a feeling have to be before it reaches the threshold of genuine foreboding? Now, this doesn't sound very ecological, but I'll bet readers of this listserv can come up with examples from biology that approach this level of nebulosity. Here's my stab at it: How K-selected must an organism be before we say it is K-selected (or r-selected). How many factors in an environment must conduce to K-selection before we say it is a K-selecting environment? How many species in that environment must bear the earmarks of K-selection before we accept the hypothesis that it truly is a K-selecting environment? What about all the species in that environment that don't appear to be K-selected? I realize, of course, that different organism may be responding to different factors in the environment, and that we can get around some of these problems by defining a hypotheses sufficiently narrowly. However, the more narrowly we define the hypothesis, the less it tells us about nature because it is less generalizable, and I suppose that most researchers would like to come up with insights that are generalizable. I don't know if this relates to some of the problems that prompted Jane's query, but I'd love to see your thoughts on the matter. Martin M. Meiss 2011/2/28 Shermin ds shermi...@gmail.com I like Manuel's response. To answer Jane's other questions: 1. Does it help you do better science? It can, but not necessarily. See below. Is it crowding out other approaches? I'd like to hear more about this - what other systematic approaches are there? For example, anecdotal observations are generally discouraged, but sometimes anecdotal observations are valuable and should be a) reported and b) inspire further observation and/or experiment. E.g. observations of tool-use in animals in the wild are great example of spontaneous events that one can never set out to observe systematically (except in controlled lab settings) but are nonetheless highly informative. I also wonder about replication - the larger or longer the scale (e.g. ecosystem, biome/longitudinal studies) the harder it is to replicate. This gets at Manuel's distinction about statistical vs. scientific hypotheses. You might have a hypothesis about a process but observe outcomes that are inherently difficult to attach a p-value to or find multiple examples of. Thoughts on that? Finally there's the issue of taxonomic poverty. Hypotheses about clades with few species are more difficult to test than those with a greater number of species. A problem if you're interested in the species-poor clade for other reasons. I.e. there is a trend towards choosing your species/system of study based on your questions of interest, and lately I've heard many talks that begin We chose to study species X because it is an excellent model for testing Y... What if you simply want to know about species X for no other reason than that you want to know about it? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? Yes, and I'm wondering about this trend in the stated aims of some journals as well. -- Shermin de Silva, Ph.D http://elephantresearch.net/fieldnotes http://www.sas.upenn.edu/~sdesilva On Mon, Feb 28, 2011 at 10:49 AM, Manuel Spínola mspinol...@gmail.com wrote: Dear Jane, That is a topic that have interested me for a long time. I teach something of this in my classes to master students in wildlife management and conservation here in Costa Rica. I know this is a controversial issue. First I recommend these 3 books: Scientific Method for Ecological Research. E. David Ford. Method in Ecology: Strategies for Conservation. Kristin S. Shrader-Frechette and Earl D. McCoy A Primer on Natural Resource Science. Fred S.
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear Manuel, Thanks for your reply! I'll have to look up the books you recommended. On Mon, Feb 28, 2011 at 7:49 AM, Manuel Spínola mspinol...@gmail.com wrote: Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. I agree. The fact that the two are conflated so often is why I decided to ask about them together. Falsification is the contribution of Karl Popper to the Hypothetic-Deductive method. It has nothing to do with statistics or statistical hypothesis. The hypothetic-deductive method has been considered as the scientific method, however not many people know how it works. The hypothetic-deductive method is inductive and not deductive like the namesuggest. Now that's an interesting comment. Popper went out of his way to avoid induction. In fact, he actually claimed that it doesn't exist in science! Why do you say that the hypothetico-deductive method is actually inductive? Best, Jane On 27/02/2011 11:44 p.m., Jane Shevtsov wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- Manuel Spínola, Ph.D. Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río Institutional website: ICOMVIS -- - Jane Shevtsov Ecology Ph.D. candidate, University of Georgia co-founder, www.worldbeyondborders.org Check out my blog, http://perceivingwholes.blogspot.comPerceiving Wholes In the long run, education intended to produce a molecular geneticist, a systems ecologist, or an immunologist is inferior, both for the individual and for society, than that intended to produce a broadly educated person who has also written a dissertation. --John Janovy, Jr., On Becoming a Biologist
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Dear Jane, That is true (and very interesting), Popper didn't believe in inductive reasoning as part of the scientific process, however, when you apply the H-D method you can only corroborate the hypothesis, you cannot confirm or prove logically an hypothesis, but you can logically reject the hypothesis. Popper proposed to work on falsification instead of corroboration, because he recognized the asymmetry. The following is not about a true hypothesis but help to make the case: If you continue seeing white swan you cannot logically confirm that all the swans in the planet are white (in fact they are not, there is a species of black swan in Australia). You only confirm the hypothesis until more data is collected, but you never are 100% sure that all the swan are white. However, if you see a black swan you can logically reject the hypothesis, and you are 100% sure that not all the swan in the planet are white. Many scientisits will search for more white swans, but Popper will say, don't worry for more white swans, search for a black swan. Hope this help, Sorry about any grammatical error but English is not my first language. Manuel On 28/02/2011 04:51 p.m., Jane Shevtsov wrote: Dear Manuel, Thanks for your reply! I'll have to look up the books you recommended. On Mon, Feb 28, 2011 at 7:49 AM, Manuel Spínolamspinol...@gmail.com wrote: Is necessary to distinguish between statistical and scientific hypothesis. Statistical hypotheses is about patterns, scientific hypotheses are about process (they are based on why or how). My experience on this topic tells me that most ecologists do not know the difference between the 2 kind of hypothesis. I agree. The fact that the two are conflated so often is why I decided to ask about them together. Falsification is the contribution of Karl Popper to the Hypothetic-Deductive method. It has nothing to do with statistics or statistical hypothesis. The hypothetic-deductive method has been considered as the scientific method, however not many people know how it works. The hypothetic-deductive method is inductive and not deductive like the namesuggest. Now that's an interesting comment. Popper went out of his way to avoid induction. In fact, he actually claimed that it doesn't exist in science! Why do you say that the hypothetico-deductive method is actually inductive? Best, Jane On 27/02/2011 11:44 p.m., Jane Shevtsov wrote: Fellow Ecologgers, Lately, I've been thinking a lot about the role of hypothesis testing (both the statistical and falsificationist varieties) in biology in general and ecology in particular. Before saying anything, I want to ask the forum a few questions. 1. What do you think of the current emphasis on hypothesis-driven research? Does it help you do better science? Is it crowding out other approaches? 2. Have you ever had a grant proposal or publication declined because of an absent or unclear hypothesis? 3. Have you ever recommended that someone else's grant proposal or publication be declined for that reason? Was it the main reason? I look forward to hearing what people have to say. Jane Shevtsov -- Manuel Spínola, Ph.D. Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río Institutional website: ICOMVIS -- *Manuel Spínola, Ph.D.* Instituto Internacional en Conservación y Manejo de Vida Silvestre Universidad Nacional Apartado 1350-3000 Heredia COSTA RICA mspin...@una.ac.cr mspinol...@gmail.com Teléfono: (506) 2277-3598 Fax: (506) 2237-7036 Personal website: Lobito de río https://sites.google.com/site/lobitoderio/ Institutional website: ICOMVIS http://www.icomvis.una.ac.cr/
Re: [ECOLOG-L] Hypothesis Testing in Ecology
Hi Martin, If you state a scientific hypothesis you need to derive predictions from it, and sometimes you can state the predictions as statistical hypotheses, but not always, in fact, Karl Popper was not thinking on statistics or statistical hypotheses. As Malcolm McCallum said if you use statistics to test a scientific hypothesis I think you are in a more shaking ground, statistics has its own problems. By the way, statistics is inductive. Some people are using information theoretic approaches like AIC (Akaike Information Criterion) to work with what they believe are scientific hypothesis but I don't think so. They have a set of models but not necessarily a set of scientific hypotheses. Other problems of working with hypothesis in ecology are the multicausality of ecological phenomena and the limitation of conducting experiments at some spatial and time scales. Most of studies in ecology are observational not experimental. More on Popper. Karl Popper did not believe that the theory of evolution by natural selection was a scientific theory. He argument that you cannot falsify an hypothesis derived from that theory. Imagine testing the hypothesis of past competition on Darwin finches under the H-D method. How do you falsify that hypothesis using a critical experiment as the H-D followers call it? Best, Manuel On 28/02/2011 05:16 p.m., Martin Meiss wrote: I'm not sure I understand Manuel's distinction between statistical hypootheses and scientific hypothesis. Is not the former supposed in some way to mathematically embody/parameterize the latter? But in any case, it seems to me that it is often hard to rigorously formulate a null hypothesis and a corresponding working hypothesis. Suppose you hear an account where someone had a feeling of foreboding about his mother, only to discover later that just when he was having that feeling, his mother, thousands of miles away, had suddenly died. When people tell stories like this, it's often followed with a challenge, like you can't tell me that's just a coincidence! Well, I'd like to say it is a coincidence, but how could you test it? What is the expected number of times you should have a feeling of foreboding about your mother and she DOESN'T die? What is the expected number of times mothers should die without their sons/daughters having feelings of foreboding? How close to the actual time of death does the feeling of foreboding have to be before we can count it? How creepy does a feeling have to be before it reaches the threshold of genuine foreboding? Now, this doesn't sound very ecological, but I'll bet readers of this listserv can come up with examples from biology that approach this level of nebulosity. Here's my stab at it: How K-selected must an organism be before we say it is K-selected (or r-selected). How many factors in an environment must conduce to K-selection before we say it is a K-selecting environment? How many species in that environment must bear the earmarks of K-selection before we accept the hypothesis that it truly is a K-selecting environment? What about all the species in that environment that don't appear to be K-selected? I realize, of course, that different organism may be responding to different factors in the environment, and that we can get around some of these problems by defining a hypotheses sufficiently narrowly. However, the more narrowly we define the hypothesis, the less it tells us about nature because it is less generalizable, and I suppose that most researchers would like to come up with insights that are generalizable. I don't know if this relates to some of the problems that prompted Jane's query, but I'd love to see your thoughts on the matter. Martin M. Meiss 2011/2/28 Shermin dsshermi...@gmail.com I like Manuel's response. To answer Jane's other questions: 1. Does it help you do better science? It can, but not necessarily. See below. Is it crowding out other approaches? I'd like to hear more about this - what other systematic approaches are there? For example, anecdotal observations are generally discouraged, but sometimes anecdotal observations are valuable and should be a) reported and b) inspire further observation and/or experiment. E.g. observations of tool-use in animals in the wild are great example of spontaneous events that one can never set out to observe systematically (except in controlled lab settings) but are nonetheless highly informative. I also wonder about replication - the larger or longer the scale (e.g. ecosystem, biome/longitudinal studies) the harder it is to replicate. This gets at Manuel's distinction about statistical vs. scientific hypotheses. You might have a hypothesis about a process but observe outcomes that are inherently difficult to attach a p-value to or find multiple examples of. Thoughts on that? Finally there's the issue of taxonomic poverty.