Re: article on Excel errors
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (Gerry Dizinno) wrote: > A while back someone posted a reference to an article (actually a web site > that referenced the article) that purported to identify some serious errors > associated with the statistical functions of Excel. > 'On the accuracy of statistical procedures in Microsoft Excel 97' McCullough, B.D. and Wilson, B. in Computational Statistics and Data Analysis, 28 July 1999, Vol 31 Part 1, Pages 27-37 regards p@c Sent via Deja.com http://www.deja.com/ = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
No Subject
unsubscribe edstat-l end = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: regression to the mean
In article <94pmgo$rn7$[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote: >Avid regression-to-the-mean watchers may be interested to know that, >according to yesterday's summary of the growing rich-poor divide >(on teletext news), the current top 10% of earners have had >a higher percentage increase in income over the past x years >(for some x that I've forgotten) than have the lowest 10% of earners. This has nothing to do with regression to the mean. The people in the top 10% and the bottom 10% have changed. In Galton's data, the top 10% of the sons were just about as far from the mean as the top 10% of the fathers. -- This address is for information only. I do not claim that these views are those of the Statistics Department or of Purdue University. Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399 [EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558 = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
article on Excel errors
A while back someone posted a reference to an article (actually a web site that referenced the article) that purported to identify some serious errors associated with the statistical functions of Excel. I have misplaced both the web site and the article. All I have recorded is that the article appeared, at some time, in the Journal of Computational and Graphical Statistics. Does anyone have the exact reference for this article. I'd like to get a copy. Thanks so much. Gerry Gerry Dizinno, Ph.D. Assistant to the President for Planning & Institutional Research St. Mary's University One Camino Santa Maria San Antonio, TX 78228-8507 Phone: 210-436-3601 Fax: 210-431-4245 email: [EMAIL PROTECTED] = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Communicating stat results that are large or small numbers to lay audience
The lecture that Mike Granaas spoke about was by Arnold Barnett at MIT. It is a great lecture, and is still available in audio. One of my favorite lines is a comment to the effect that "There are only two things that I object to in this ratio--the numerator and the denominator." At 03:20 PM 1/25/01 -0600, Mike Granaas wrote: > >One of the Chance lectures (http://www.dartmouth.edu/~chance) from about 2 >years ago spoke about this. The example I remember had to do with the >safety of airtravel. Specifically the speaker translated the probability >of dying in an airplane accident into "one flight per day for xxx years" >(I think the xxx was in the low thousands). > >A few years back there was a tv news special "are we scaring ourselves to >death", or something like that. The interesting part is the post-show >discussion where they translated various risk factors into "days of life >expectency lost." They took some of the issues of the day (pesticides on >apple trees) and compared them to some everyday activities (driving) and >some known health hazzards (smoking, poverty). > >This is the flavor that I would be considering. > >Michael > >On Thu, 25 Jan 2001, Arthur J Kendall wrote: > >> We are currently looking at how quantitative risk assessment is used. >Doses can be in micrograms, morbidity can be in rates per hundred thousand >exposed, mortality in 10 million exposed, and so forth. In this context, we >are looking for citations, suggestions, URL's, etc. that address the issue >of presenting to a LAY audience how to grasp the meaning of large and small >numbers, e.g., 3E9, 1.4E-9, billions, millions, billionths, and millionths. >The numerate community can be asked to recall how to deal with these >numbers, but we will be addressing a lay audience and want to give them some >help. >> >> >> > >*** >Michael M. Granaas >Associate Professor[EMAIL PROTECTED] >Department of Psychology >University of South Dakota Phone: (605) 677-5295 >Vermillion, SD 57069 FAX: (605) 677-6604 >*** >All views expressed are those of the author and do not necessarily >reflect those of the University of South Dakota, or the South >Dakota Board of Regents. > > > >= >Instructions for joining and leaving this list and remarks about >the problem of INAPPROPRIATE MESSAGES are available at > http://jse.stat.ncsu.edu/ >= David C. Howell Phone: (802) 656-2670 Dept of Psychology Fax: (802) 656-8783 University of Vermont email: [EMAIL PROTECTED] Burlington, VT 05405 http://www.uvm.edu/~dhowell/StatPages/StatHomePage.html http://www.uvm.edu/~dhowell/lies4thedition/index.html http://www.uvm.edu/~dhowell/gradstat/index.html = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Comparison of statistical software
"Ken K." <[EMAIL PROTECTED]> wrote in message 94i9b3$2j2$[EMAIL PROTECTED]">news:94i9b3$2j2$[EMAIL PROTECTED]... > JMP - powerful, claims to be easy to use, but I always found it quirky. For > example, ALL of the modeling is done via its "Fit Model" tool. To do > anything you'd better know this well I think MINITAB is a better choice. JMP > 3's help and documentation was pretty mediocre. Also it is really a > Macintosh program sitting in Windows. It has kept the Mac quirkiness. I think if you give JMP version 4 a look, you'll find most of these criticisms have been addressed. Modeling is still through Fit Model, but it is no longer a Mac progrram running in windows. In version 4, we completely re-wrote the code to use totally windows-native code. The help system is also quite extensive, and reflects everything that is seen in the printed manuals. The manuals are also included on the CD. Claimer: I work on the JMP team and wrote the online help. -- ` ___ ' - (O o) - --ooO--(_)--Ooo--- _ __ __ _ | | \/ | __ \ ® Lee Creighton | | \ / | |__) | SAS Statistical Instruments _ | | |\/| | ___/ | |__| | | | | |[EMAIL PROTECTED] \/|_| |_|_| 5275R SAS Campus Drive (919) 531-3755 Statistical Discovery Software = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: change scores
dennis roberts wrote: > At 05:19 PM 1/25/01 +, Gene Gallagher wrote: > >> They conclude that the measurement of change has been and >> continues to be a highly controversial subject. >> -- >> Eugene D. Gallagher >> ECOS, UMASS/Boston > > > > [big snip] > > so, when someone asks how much has my child "learned" in vocabulary or > math or CHANGED ... from grade 1 to say ... grade 3 ... or grade 9 ... > > we haven't a clue ... and no way to communicate answers to these > questions to parents, that's for sure (nor maybe even amongst the > professionals) > > sort of sad, isn't it? Extremely so. (and my 'kids' are busy having their own kids today, so my interest in the outcome is fairly long term.) Which comes back to, we need a measure that many/most parents can understand, and relate to some vision of 'progress.' the Massachusetts schema doesn't look to me to meet this objective very well. NOw, this group understands that measuring change is technically difficult. Either we can find a method that is reasonably clear, or we let the Mass. schema type thing do its thing. Not a happy prospect. Jay -- Jay Warner Principal Scientist Warner Consulting, Inc. North Green Bay Road Racine, WI 53404-1216 USA Ph: (262) 634-9100 FAX: (262) 681-1133 email: [EMAIL PROTECTED] web: http://www.a2q.com The A2Q Method (tm) -- What do you want to improve today? = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Communicating stat results that are large or small numbers tolay audience
One of the Chance lectures (http://www.dartmouth.edu/~chance) from about 2 years ago spoke about this. The example I remember had to do with the safety of airtravel. Specifically the speaker translated the probability of dying in an airplane accident into "one flight per day for xxx years" (I think the xxx was in the low thousands). A few years back there was a tv news special "are we scaring ourselves to death", or something like that. The interesting part is the post-show discussion where they translated various risk factors into "days of life expectency lost." They took some of the issues of the day (pesticides on apple trees) and compared them to some everyday activities (driving) and some known health hazzards (smoking, poverty). This is the flavor that I would be considering. Michael On Thu, 25 Jan 2001, Arthur J Kendall wrote: > We are currently looking at how quantitative risk assessment is used. Doses can be >in micrograms, morbidity can be in rates per hundred thousand exposed, mortality in >10 million exposed, and so forth. In this context, we are looking for citations, >suggestions, URL's, etc. that address the issue of presenting to a LAY audience how >to grasp the meaning of large and small numbers, e.g., 3E9, 1.4E-9, billions, >millions, billionths, and millionths. The numerate community can be asked to recall >how to deal with these numbers, but we will be addressing a lay audience and want to >give them some help. > > > *** Michael M. Granaas Associate Professor[EMAIL PROTECTED] Department of Psychology University of South Dakota Phone: (605) 677-5295 Vermillion, SD 57069 FAX: (605) 677-6604 *** All views expressed are those of the author and do not necessarily reflect those of the University of South Dakota, or the South Dakota Board of Regents. = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: change scores
On 24 Jan 2001 20:11:25 -0800, [EMAIL PROTECTED] (Jay Warner) wrote: > > First off, can we get an operational definition of 'self esteem'? ... Oh! I hoped he was using 'self esteem' as a place holder for a hypothetical study. Sort of like, Professional baseball teams sometimes have trades featuring 'a minor leaguer to be named later' -- here was, 'some useful concept to be named later.' If you just take 'self esteem' in a generic way, by itself, it was never more than a joke... despite what the El-Hi school system of California might think. (oooh, perhaps I slander them - what I know is mainly from comedians on TV.) My homepages (not my stats-FAQ) has a URL, - "Self-efficacy" information "self-esteem" (repairing that term) with sufficient detail and complexity. re: "Arthur Bandura" By the way, if you have Pre-Post on one measure, you almost need to plot the points on a well-labeled graph (what is max, what is min?) before you BEGIN to draw conclusions. - Then, a *disordinal* interaction is just about the only EFFECT that I can think of, which cannot be discounted as artifactual or pretty trivial. If you don't have a control group on hand, you need to have information about what a control group SHOULD look like. For instance, in Education: If you group the highest IQ versus lowest, the "regression" for a year or two will be opposite: the highest will learn more new stuff, faster, and get further ahead. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: change scores
At 05:19 PM 1/25/01 +, Gene Gallagher wrote: > They conclude that the measurement of change has been and >continues to be a highly controversial subject. >-- >Eugene D. Gallagher >ECOS, UMASS/Boston this is so true and, if we think about it ... this is rather amazing (and depressing at the same time ... amazingly depressing ... has a certain ring to it) ... given, that so much of measurement ... and assessment ... of people and kids in schools ... and health regimens and how the economy goes ... is based on a fundamental notion of "change" from one point in time to another ... to another ... etc. when we are parents ... with new kids ... we usually have some kind of a baby book ... and, probably, are "prompted" to create a chart ... from 0 on the baseline to years ... and on the Y we have "height" or "weight" etc. ... so, the book encourages you to make a chart ... see the trend ... that is, make a visual of the "change" what if we wanted to do that in school, with students ... say, pick "vocabulary" ... or "facility in math" ... and start the baseline at 1st grade ... on up to 12th grade ... with tick marks at 2, 3 . 10, 11 and 12 ... and the Y being "amount or scope of vocabulary" or "speed in doing multiplication" or whatever could we do it? DO we do it? the answer to the first one is ... yes (at least rudimentarily) ... but the answer to the second is NO!!! so, when someone asks how much has my child "learned" in vocabulary or math ... or CHANGED ... from grade 1 to say ... grade 3 ... or grade 9 ... we haven't a clue ... and no way to communicate answers to these questions to parents, that's for sure (nor maybe even amongst the professionals) sort of sad, isn't it? = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Communicating stat results that are large or small numbers tolay audience
We are currently looking at how quantitative risk assessment is used. Doses can be in micrograms, morbidity can be in rates per hundred thousand exposed, mortality in 10 million exposed, and so forth. In this context, we are looking for citations, suggestions, URL's, etc. that address the issue of presenting to a LAY audience how to grasp the meaning of large and small numbers, e.g., 3E9, 1.4E-9, billions, millions, billionths, and millionths. The numerate community can be asked to recall how to deal with these numbers, but we will be addressing a lay audience and want to give them some help. ** Message from InterScan E-Mail VirusWall NT ** ** No virus found in attached file noname.htm * All attachmes were scanned by Interscan 3.5, and no virusus were found. * End of message ***
Re: regression to the mean
but this fails to take into account (amongst other things) ... that the way rich and poor tend to get increments to their salaries/earnings (as a general rule) ... is due to some kind of a multiplicative constant ... that is ... some % value is that amount that increments their salaries ... over the years, under this arrangement ... the gaps between the top and bottom will necessarily get larger and larger and larger ... also, it is true that the highest paid will tend to get a larger % than the lower paid ... though even if that were the same % ... the principle above would still make the gaps larger and larger over time but, even with that ... if you take two years ... and line up this year's salaries of say 1000 including those from the top down to the bottom ... and these same folks salaries next year ... there will not be a perfect correlation (though this will be quite high) ... thus, there still will be SOME regression to the mean that is ... if we isolate the top 50 ... and the bottom 50 ... and look at their percentile ranks (or mean z scores) from this year to next ... you will still see that the lower 50 have relatively higher percentile ranks than they did the first year ... while the top 50 will tend to have LOWER percentile ranks (or mean z scores) ... so, RTM still happens ... EVEN if the uppers GAIN in salary over the ones the bottom At 05:07 PM 1/25/01 +, wrote: >Avid regression-to-the-mean watchers may be interested to know that, >according to yesterday's summary of the growing rich-poor divide >(on teletext news), the current top 10% of earners have had >a higher percentage increase in income over the past x years >(for some x that I've forgotten) than have the lowest 10% of earners. > > -- Ewart Shaw >-- >J.E.H.Shaw [Ewart Shaw][EMAIL PROTECTED] TEL: +44 2476 523069 > Department of Statistics, University of Warwick, Coventry CV4 7AL, U.K. > http://www.warwick.ac.uk/statsdept/Staff/JEHS/ >yacc - the piece of code that understandeth all parsing > > >= >Instructions for joining and leaving this list and remarks about >the problem of INAPPROPRIATE MESSAGES are available at > http://jse.stat.ncsu.edu/ >= _ dennis roberts, educational psychology, penn state university 208 cedar, AC 8148632401, mailto:[EMAIL PROTECTED] http://roberts.ed.psu.edu/users/droberts/drober~1.htm = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Regression vs ANOVA
Alexander Tsyplakov <[EMAIL PROTECTED]> wrote: : No, regression models can have stochastic and/or discrete : regressors. I can only agree that regression models have There are no constraints whatsoever on the x variables for the significance tests and estimates to be valid. Power is another matter. Wheeler is right; both are special cases of the general linear model. = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: change scores
In article <94o81d$9o4$[EMAIL PROTECTED]>, Elliot Cramer <[EMAIL PROTECTED]> wrote: > Dale Berger <[EMAIL PROTECTED]> wrote: > : Dear Colleagues, > > : A student is evaluating a summer program for junior high students. One of > : the goals was to raise 'self esteem.' Measures were taken before the > > there is no good answer. You might look at "problems in measuring > change" edited by chester Harris. He should have done a randomized > experiment and he could then have gotten a definite answer. > > You might look at the deviations from the regression of post and pre and > see if this relates to anything > Someone pointed out the book by DT Campbell and DA Kenny (1999) A primer on regression artifacts. The Guilford Press. I ordered it, and it has been a pleasure to read. Chapter 6 in Campbell & Kenny (1999) discuss the various ways of correcting for regression to the mean in pre- and post-test comparisons. They present 3 different ways for correcting for the regression to the mean effect: residualized change, "backward" residualized change, and estimated true change. They provide an example which produces qualitatively different answers for each method. Each method can be justified based on the meaning of "change." I can't reproduce their arguments here. Campbell & Kenny provide detailed equations with extensive citations of the primary literature and reviews. They conclude that the measurement of change has been and continues to be a highly controversial subject. -- Eugene D. Gallagher ECOS, UMASS/Boston Sent via Deja.com http://www.deja.com/ = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
regression to the mean
Avid regression-to-the-mean watchers may be interested to know that, according to yesterday's summary of the growing rich-poor divide (on teletext news), the current top 10% of earners have had a higher percentage increase in income over the past x years (for some x that I've forgotten) than have the lowest 10% of earners. -- Ewart Shaw -- J.E.H.Shaw [Ewart Shaw][EMAIL PROTECTED] TEL: +44 2476 523069 Department of Statistics, University of Warwick, Coventry CV4 7AL, U.K. http://www.warwick.ac.uk/statsdept/Staff/JEHS/ yacc - the piece of code that understandeth all parsing = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
RE: Q-techniques
Also of interest are the writings of R.B. Cattell from the '50s and '60s. There is a nice discussion of his "data box" in the first Handbook of Multivariate Statistics, published in the mid '60s. Bill __ William B. Ware, Professor and Chair Educational Psychology, CB# 3500 Measurement, and Evaluation University of North Carolina PHONE (919)-962-7848 Chapel Hill, NC 27599-3500 FAX: (919)-962-1533 http://www.unc.edu/~wbware/ EMAIL: [EMAIL PROTECTED] __ On Wed, 24 Jan 2001, Dale Glaser wrote: > There is also a nice chapter on Q-techniques by Grimm & Yarnold (2000). > Reading and understanding more multivariate statistics. DC: APA. ...it is > authored by Bruce Thompson (ch. 6) and titled: Q-technique factor analysis: > One variation on the two-mode factor analysis of variables. (pp. 207-226) > > > .dale glaser > > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > On Behalf Of Rich Ulrich > Sent: Wednesday, January 24, 2001 7:51 AM > To: [EMAIL PROTECTED] > Subject: Re: Q-techniques > > On Wed, 24 Jan 2001 13:53:25 GMT, "Paolo Covelli" <[EMAIL PROTECTED]> > wrote: > > > I know that the techniques as principal component analysis, factor > analysis > > or canonical correlation analysis are called R-techniques, because the > > correlation matrix R plays an important role in this approach. > > Instead techniques such as discriminant analysis, cluster analysis or > > multidimensional scaling are known as Q-techniques. > > Why this name? > > That's not exactly the way I have seen the distinction. > > I am browsing Jum Nunnally, "Psychometric Theory," 1967, page 361. > He describes the usual R technique with Subjects x Variables (rows and > columns, respectively). Correlations are between columns > (variables). P technique has the same person on every row, using > different Periods or points in time. > > Q technique is described as concerning correlations among those rows, > and, he says, is sometimes referred to as 'transposed' or 'inverted' > analyses. > > The name? Partly because the text goes on to mention O technique, > I suspect that the "Q" name came from splitting the alphabetical > difference between P (with its mnemonic excuse) and R. > > -- > Rich Ulrich, [EMAIL PROTECTED] > http://www.pitt.edu/~wpilib/index.html > > > = > Instructions for joining and leaving this list and remarks about > the problem of INAPPROPRIATE MESSAGES are available at > http://jse.stat.ncsu.edu/ > = > > > > = > Instructions for joining and leaving this list and remarks about > the problem of INAPPROPRIATE MESSAGES are available at > http://jse.stat.ncsu.edu/ > = > = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Help Log-Probit in Excel
This is a multi-part message in MIME format. --46D0737844C9CF0BEDE1D80F Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I don't know where that came from. I studied from Finney as well. I guess I'm suffering from old-timer's disease. :-) --46D0737844C9CF0BEDE1D80F Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --46D0737844C9CF0BEDE1D80F-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: WTB: HP3497A voltmeter
Check the capacitors in the power supply. Just fixed one last month. Take Care, Kevin [EMAIL PROTECTED] wrote: > > Hello, > > The voltmeter of one of our HP3497A died today. If you have one in > good condition, please make an offer, we are interested! > > best regards > > Dr Ir Philippe Boeraeve > > > Dr Ir Philippe Boeraeve > Universite de Liege > Departement "Mecanique des materiaux" > Labo MSM > Chemin des Chevreuils, 1 (Bat. B52) > B-4000 Liege 1 > BELGIUM > > E-mail: [EMAIL PROTECTED] > phone: +32-4-3669348 > fax:+32-4-3669342 -- Spam Filter = Change .combo to .com ** "We ought always to know precisely why a given job is done in a particular way, and why it is done at all, and why it can't be done more efficiently, if it must be done at all."-- T.J.Watson *** = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: A much more basic MCAS fallacy?
dennis roberts wrote: > well, if george bush pushes for more testing for accountability ... across > the nation like in texas, then i suspect more WILL get into this debate ... > either voluntarily or, being forced into it for some reason or another > > the major problem is not that the community of assessment folks ... can't > agree on some reasonable plan ... they ACTUALLY can (even while debating it > very strongly) ... the difficulty is that this is totally a POLITICAL ISSUE > ... that is spawn in the wells of state legislatures ... and so too if > bush's plan moves forward, the wells of the us of a congress > > it is very hard for those in some professional arena ... to make the kind > of technical dents in the pragmatics of the legislatures' thinking caps ... > > so, while i sympathize with chris' plea ... this will be a very difficult > nut to crack ... particularly at the national level I sort of agree. There are good statistical models for school effectiveness (e.g., the work by Harvey Goldstein's group at the IOE in London), but even in the UK the government has been very slow to shift to "value added" models from flawed league tables of raw data. This may change, but probably only because value added tables are (I think) going to be published in Northern Ireland. Thom = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
