Statistical Software Analyst/Programmer
The Center for Statistical and Mathematical Computing at Indiana University has an open position for a software analyst/ programmer (http://www.indiana.edu/~statmath). Responsibilities: Under minimal supervision, responsible for technical support and consultation with users for statistical software. Support statistical software across all University Information Technology Services (UITS) supported platforms (windows, mac, unix). Responsibilities involve evaluation, testing, and support of statistical software. Prepare documentation and present short classes as needed; attend meetings as required. Qualifications: Masters degree in statistics/quantitative methodology or closely related area; advanced degree preferred. In-depth knowledge of statistical software required; experience with SPSS, SAS, Minitab, and RATS particularly helpful. Experience with multiple computer platforms required; experience in a university environment and experience with UNIX and Windows NT operating systems is preferred. Competitive salary. To apply send cover letter and application with three references to: Thea Brown ([EMAIL PROTECTED]) or University Information Technology Services, 2711 East Tenth Street, Indiana University, Bloomington, Indiana 47408. Application deadline: Friday, January 21, 2000. Indiana University is an equal opporutnity employer.
unsubscribe
RE: Christmas Reading?
Have a go at Prof. Stigler's latest "Statistics on the Table" (1999, Harvard University Press). A very scholarly and often entertaining collection of historical essays about lead characters and ideas in the great story of statistics. It's amazing how deep some apparently modern ideas are actually rooted in time. Eugene *** Date: Tue, 21 Dec 1999 09:43:07 -0500 From: "Tatikola, Kanaka [PRI]" [EMAIL PROTECTED] Subject: RE: I personally like THE HISTORY OF STATISTICS , The measurement of uncertainity before 1900, by Stephen M. Stigler. Kanaka
Re: adjusting marks; W. Edwards Deming
Jim Clark wrote: Artificially giving all students (or almost all) the same grade does not minimize variation in the underlying trait, achievement, in this case. It simply hides the variation so that one does not know to what extent one is minimizing differences in achievement, and rewards students for not trying to achieve more than some minimal level. I don't Deming would have said assignment of Pass/Fail should be "artificial". If the student doesn't perform, then of course they shouldn't Pass. He did say, on the other hand, that grading imposes an artificial scarcity of A's (also of C's and D's). These are again Deming's words, and echo Dennis Robert's comments about the pure subjectivity of the grading process. The motivation for the students should be in Joy of Learning (one of Deming's 14 points) rather than the grade. This I agree with wholeheartedly. How can we achieve this? I think it is our main challenge as educators. Using the grading system as a motivational substitute for Joy of Learning is lazy, inefficient management of our classes. Students who are fairly sure they are not going to get the coveted A, or who only need a "C or better" are going to give less effort. This will increase variation, and operates contrary to the stated goal of the system. My question is again: Is ranking really necessary? Given the goal of reducing variation, what does it help? Students in competition for the scarce A's will withhold information from one another. Does this achieve the stated aim of the system in an optimal way? W. Edwards Deming would have said, most emphatically, no. He spoke quite often of the educational system particularly in his later years; his message was not at all meant to be limited to manufacturing. Grading is not equivalent to ranking, unless one uses a forced distribution. One can grade without any restriction on the number of As or other grades other than the achievement of the students. I would be interested in hearing about any empirical evidence that non-use of grading schemes produces better or even as good learning as the use of grades? I think this is a very important point: what can we do in place of ranking? Now, as much as you say you don't use ranking, I am not sure you can get away without out. What if all of a sudden everyone got A's by your criteria? Wouldn't the administration get on your case? Then, you might say, just make the criteria harder so that we get back to a "normal" proportion of As, Bs etc. Well, aren't you just back to ranking? I don't have any data from the classroom experience, but I do have an observation from business. Texas Instruments had a policy of ranking plants in terms of their performance. The employees at the top plants received bonuses. Great idea, right? Motivates people, makes them perform to the best of their abilities, just like grading. The problem is, the innovations were hoarded by the individual plants to secure the bonuses, to the detriment of the company at large. Optimization of individual processes can be detrimental to the system, if the system at large is not considered in the optimization process. Thanks for the continuing discussion. I have been profoundly influenced by the words of W. Edwards Deming, and hope others will take a look at what he had to say, at least to stimulate discussions such as this. As he himself said, you don't simply "implement" his system, much like you don't learn to play piano by buying one and placing it in your living room. In the same way, you don't simply implement Deming's method as it applies to teaching by implementing P/F and be done with it. I would like to know, are there any others out there who have been influenced by Deming? Has his message lost its force in our current climate of economic prosperity? Peter Westfall
unsubscribe
unsubscribe [EMAIL PROTECTED]
Re: adjusting marks
"David A. Heiser" wrote: - Original Message - From: Peter Westfall [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Tuesday, December 21, 1999 6:45 PM Subject: Re: adjusting marks Bob Hayden wrote: - Forwarded message from Peter Westfall - Deming himself (if I remember correctly) graded everyone as "A" until the administration noticed, and then they made his courses Pass-Fail. Deming was also very much against ranking students in any way, except for the possible exception of identifying an exceptional student that others might emulate (the 3*sigma student) and identifying the exceptionally poor student ( 3*sigma) for remediation. All other students should be be essentially equivalent, in Deming's philosophy. - End of forwarded message from Peter Westfall - Would you recommend this for drivers' license tests? Oh, I get it, that's what we're doing already! No wonder. I have to admit, it would sure simplify quality control if we considered anything within +- 3 s.d. to be OK. Then I guess the motivation would be to throw in a few clunkers now and then to keep the s.d. as large as possible? Bob, Your remarks sound facetious. I was hoping to stimulate some serious discussion. Have you read anything by Deming? Here is Deming's philosophy, as well as I can paraphrase it for the present situation: Students/teachers/administrators form a system. The system has an aim, which is (presumably) to educate everyone as well as possible, for the good of the students, and for the good of society. What good does ranking do? Does it help to achieve the aim of the system? Or rather, is it simply a weeding process? Is ranking necessary? (these are mainly Deming's words, but I must admit I see lots of value there.) Regarding making the standard deviation large, Deming would say that management's (professors, administrators) job entails minimizing variation among students. This can be done in the usual ways - admissions procedures, advising, prerequisites. Individual classes are "processes" within the larger system, and in the process of continual improvement, one seeks ways to minimize variation within the processes. Deming shows a diagram where the knowledge of people before training is scattered and highly variable, and after training the mean level is higher but the variation smaller. The inference is that the more effective the classroom experience, the less variation in the final levels of knowledge and abilities of the students, as they pertain to the subject at hand. My question is again: Is ranking really necessary? Given the goal of reducing variation, what does it help? Students in competition for the scarce A's will withhold information from one another. Does this achieve the stated aim of the system in an optimal way? W. Edwards Deming would have said, most emphatically, no. He spoke quite often of the educational system particularly in his later years; his message was not at all meant to be limited to manufacturing. Peter --- Very Intersting I don't agree with Demming. Life is essentially a matter of diversity, and being able to find one's own "niche". The process of ranking is inherent in life whenever there is stress on a population. Going to college is indeed "stress". If in order to suceed, I need to obtain a PhD from Stanford, then I have to get high grades and attain other acheivments to get in that few percent that gets accepted. If my college grades are all "pass", how am I going to compete with the applicate with A+++ grades from NCU? How are new hires for the expensive New York/Washington law firms hired? Not on pass/fail but on which law school and how the professors rated the student and what were the extra curricular activities? Much of this is subjective, but when you have 300 applicants for one job, you have to do some ranking to pick the top 3 or 5. Demming I think has the quality control mindset of pass/fail in terms of manufactured objects, where everything is acceptable between -3 and +3 sigma (Now it is -6 to +6 sigma.) This may be fine for shop work on the floor. In (I think Deming had some serious problems with 6 sigma QC, but that is besides the point.) this country the only thing we manufacture now is credit and money to buy manufactured goods from other countries. You need a very diverse population now. The process of ranking as flawed as it is, works, because there are so many areas where one can find his own niche, and ranking is one way of finding one's niche. DAH No doubt about it, we can't make everyone the same, nor do we want to. We can, however, make their levels of understanding and logical thought processes similar through proper education. Human diversity is expected. We can't change people's
Re: Correlation conversion
"haytham siala" [EMAIL PROTECTED] writes: Can anyone please tell me how to convert a Kendal-tau correlation to a Pearson correlation. It is easier to get a camel through the eye of a needle than to convert a Tau to a Pearson correlation. -- --(Signature) Robert M. Hamer 732 235 4218 Use my last name @rci.rutgers.edu "Mit der Dummheit kaempfen Goetter selbst vergebens" -- Schiller
Re: adjusting marks; W. Edwards Deming
this shows how naive deming really was ... who says learning "should" be a joy? learning is WORK ... and, work is hard. now, some kids really relish the task and challenges ... but many others do not ... should we blame THEM? but, i don't really see what deming has to do with our discussion of "adjusting" marks ... At 08:33 AM 12/22/99 -0600, Peter Westfall wrote about deming: The motivation for the students should be in Joy of Learning (one of Deming's 14 points) rather than the grade. -- 208 Cedar Bldg., University Park, PA 16802 AC 814-863-2401Email mailto:[EMAIL PROTECTED] WWW: http://roberts.ed.psu.edu/users/droberts/drober~1.htm FAX: AC 814-863-1002
Re: adjusting marks
Michael Granaas ([EMAIL PROTECTED]) wrote: : While more careful admissions processes would certainly limit the : variability in students, and therefor grading, how is it any different : from grading? If you are going to be more careful with admissions you : need a ranking system of some sort to determine who will succeed and who : will fail. This is just puts the Social Darwinism issue at a different : stage of the process. There's a fundamental difference between admissions decisions and grading decisions: the former involve allocating an inherently scarce resource. There's a limit to the total number of students a school or program can admit, regardless of how certain qualities are distributed among the applicants. However imperfect the available criteria for selecting a subset of applicants are, you're going to *have* to use *some* criteria. All you can do is try to make them as "fair" as possible. There's a genuine cost associated with admitting another applicant. But evaluating performances within a class doesn't involve any inherently scarce resource. There's no particular cost that increases with the grade a student gets.
Re: adjusting marks
Not all grading practices "on a curve" are performed as described by Eric Bohlman. OK maybe I am clueless about all of this but I often saw grading on a curve being implemented when lots of students performed poorly on a test. Thus test scores were adjusted (usually in the upward direction) to make up for the poor performance that might be attributed to poor teaching, poor test construction, bad items or whatever. I never, as a teacher, used any curving procedure to lower students grades! But obviously the students scores for those performing poorest on the test had the highest increases when the curve was applied whereas those performing well saw little if any increase in their scores. Perhaps that is the unfairness you and others are referring to. Or are you referring to the decision to rescale test scores so they fit a more normal distribution? In which case, I agree that there are problems with that approach and see no reason for why anyone should assume that test scores should conform to a normal distribution or force them to do so. In fact, most teacher-made tests (and here I really want to say all) are criterion-referenced tests so why can't all students meet the criterion? There is no reason at all for why that cannot not be done except that some might think that one instructor grades more leniently than another and at the university level students will sign up in droves for the class taught by the instructor ho is the easier grader. So am I off my rocker or what? (After developing tests as a teacher, I now develop tests for states and local school districts so if I am missing a big point here, please let me know. I would hate to think I was causing harm to students.) Deanna === Deanna M. De'Liberto, President/Director of Assessment D Squared Assessments, Inc. (Specialists in Test Development/Validation and Test Administration) 9 Bedle Road, Suite 250 Hazlet, NJ 07730-1209 Phone: (732) 888-9339 Email:[EMAIL PROTECTED] Web: http://www.quikpage.com/D/dsquared Member of the Association of Test Publishers === Confidentiality Notice: This e-mail transmission may contain confidential or legally privileged information that is intended only for the individual or entity named in the e-mail address. If you are not the intended recipient, you are hereby notified that any disclosure, copying, distribution, or reliance upon the contents of this e-mail is strictly prohibited. If you have received this e-mail transmission in error, please reply to the sender, so that DSA can arrange for proper delivery, and then please delete the message from your inbox. Thank you. In a message dated 12/22/1999 2:16:36 PM Eastern Standard Time, [EMAIL PROTECTED] writes: EAKIN MARK E ([EMAIL PROTECTED]) wrote: : While I do not grade on a curve, I feel that if reasons exist,it is more : valid to adjust atypical grades distributions than not to adjust them. : My reason for not grading on a curve is more for class harmony. Grading on : a curve often means taking points away from some students while adding to : others. I noticed that a class can suddenly become hostile if some : students are treated better than others. This hostile environment can be : detrimental to a class's performance also. To put it even more bluntly, grading "on a curve" really means establishing a budget of grade points and then distributing that budget among the students, which means that the grade a particular student gets depends not only on the distribution decisions but on the size of the budget. Where on earth does this concept of a budget come from? It implies at least two questionable, to say the least, underlying assumptions: 1) That the "total" of whatever it is that grades are supposed to measure is a constant depending only on class size. 2) That it's possible to evaluate the collective performance of a group on a task *before* they've performed that task. The purpose of a budget is to make it possible to allocate limited resources. Since when is academic performance a limited resource, or even any sort of resource subject to allocation? What on earth does it mean to say to a student "your performance would be an A, but that would put me over budget so I can only give you a B" or "your performance would be a D, but I've got some extra grade points left over so I can give you a C"? The disharmony you talk about is really the result of pitting students against each other in such a way that each student's success depends on other students' failure. Why would someone want to do this? If we're not talking about allocating an inherently scarce resource, the only reason I can think of is a deliberate desire to create disharmony in order to use "divide and conquer" to prevent collective action. If
grading on the curve
this discussion is interesting ... there seems to be TWO general kinds of "grading" on the curve ... it would be interesting to try to "estimate" how frequently each happens ... 1. LOWERing cutoffs ... thus, INcreasing the #s of those getting various higher grades 2. making cutoffs such that the distribution of GRADES resembles a normal distribution i assume that #1 occurs much more frequently and, from my perspective, there is NO good rationale for doing #2 ... unless one assumes that ability within a class is normally distributed AND ... and far more crucial ... that achievement SHOULD resemble the distribution of ability ... in any case ... instructors are suppose to give students some reasonable description of the grading system used ... at the BEginning of a course ... which i assume would include some facimile of a grading scale ... or what one has to do to earn certain grades ... and in this context, i would think that anyone who might 'consider" RAISING cutoffs so that FEWER students get higher grades ... would be challenged from students .. as this appears to border on unethical practice ... At 02:32 PM 12/22/99 -0500, [EMAIL PROTECTED] wrote: I never, as a teacher, used any curving procedure to lower students grades! == dennis roberts, penn state university educational psychology, 8148632401 http://roberts.ed.psu.edu/users/droberts/droberts.htm
Re: Prediction Model Question
There were several earlier messages, and then I thought Don Burrill said most of what needed to be said -- On 20 Dec 1999 22:43:52 -0800, [EMAIL PROTECTED] (Donald F. Burrill) wrote: For openers, I quote from Pedhazur (2nd edition), p 329 (summary for Chapter 9), so that we're all on the same wavelength, more or less: "... Regardless of the coding method used, the results of the overall analysis are the same. ..." (This is the point that other respondents and I had in mind when we were questioning your interpretation of Pedhazur.) Continuing a few sentences later: "... The coding methods do differ in the properties of their regression equations. A brief summary ... follows. ..." After the summaries of each method, the final paragraph: "Which method of coding one chooses depends on one's purpose and interest. [For one purpose], dummy coding is the preferred method. Orthogonal coding is most efficient [for another purpose]. It was shown, however, that the different types of multiple comparisons ... can be easily performed [with] effect coding. Consequently, effect coding is generally the preferred method of coding categorical variables." Burke Johnson had written: 1. I agree with Joe that the term "dummy" in dummy coding is a rather [ snip, various details.] Later Don recommended constructing dummies for the Interactions in such a manner so that they would be orthogonal to the main effects, in order to reduce confusion of confounded interpretations; that bit of advice received a minor criticism from someone else who pointed out that you should never be trying to interpret *those* coefficients in the first place. Well, I agree with both Don and the critic. I create my interactions as orthogonal, or approximately orthogonal -- in the old days, your program was too likely to blow up if you did not get rid of all the numerical problems you could, whenever you could. Further, if I happen to look at the wrong listing, it will still have numbers that are in the right range, and PROBABLY right. Finally, it may be a cheap piece of consistency, but it gives me one less item that I have to explain to the non-statisticians who look at various results. Like the critic, though, I never want to interpret the coded main effects in any regression that has also included the interactions. - I would not mind receiving guidance on this final point. It is *conceivable* to use codings so that the coefficients and tests for main effects do have meaning when the interaction is included in a regression. If it is what I remember seeing in an ANOVA text many years ago, the weights and coefficients can be constructed to take into account the Ns of the cells (more complicated than -1,0,1). I believe: The test that this gives you for main effects is either exactly the same as some other way of constructing the problem, or it is considered obsolete. The construction that I like is Searle's partitioning of sums of squares, usually in a hierarchy: (A), (B|A), (AB|A,B) for instance. Today, Burke Johnson sent an SPSS-worked example to Don B., with a CC: to me, since I had posted earlier. The example was supposed to show that different codings give different results. The example shows that the total SS and test is always the same. And the example shows that different codings can give you different results for coefficients and tests when you look at Main effects when Interactions are already in the equation -- which is entirely consistent with what Don and I (I think) have both said, i.e., those effects should be presumed to be uninterpretable -- so the illustration just heightens the question of whether those effects are *ever* interpretable, since the inconsistency proves that they are not strictly interpretable for most sets of coefficients. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html
Re: teaching statistical methods by rules?
Alan McLean wrote, among other things: On the other hand, a body of knowledge can be thought of as a set of 'rules'. I think you are concentrating on the information in what is learned and ignoring the format. This works for computers, which learn in only one format (memory), but not for people, for which memory is just one format. My argument: For the sake of example, suppose I want to teach students how to tie their shoes. I could observe what I do and create a verbal description. I could teach students this verbal description, and they could memorize it. I could test them on their ability to remember this information. A student who could remember it probably could tie their shoes. My students might end up knowledge roughly the same information as me, but their knowledge wouldnt be stored in their brains the same way it is stored in mine. I have a connected series of motor movements built into my brain as a habit. And these different storage formats have different implications. My students would be good at verbal descriptions, but probably not so fast at actually tying their shoes. Now to reality. Research on implicit learning has suggested that people can learn something without being able to report what they have learned. Presumably, they have no conscious knowledge of what they have learned. In my published opinion, there are three types of implicit knowledge, with habits being just one. Combined with conscious knowledge, that makes four different types of learning. The format in which something is learned has implications. One is for memory. Research suggests that implicit learning is retained much longer than explicit learning. Another is for usage. Obviously, for verbal report, conscious knowledge is far superior than any other type of knowledge. But the other types of learning probably are probably better for other types of performance. For example, in one study, we either gave subjects implicit knowledge of a rule or explicitly taught them a collection of rules. The subjects with implicit knowledge could use the information in an identification task better than they could report it. The subjects with conscious knowledge could report the rules better than they could use them. The hardest type of learning to describe or define is what I call mental models, and what often corresponds to what people call understanding. For example, you have a mental model of your spouse (or friend). You can use this mental model to predict what your spouse or friend will do. You can also try to use this mental model to verbally describe your spouse or friend, but that isn't a natural use of the mental model and that format of learning isn't that good for verbal report. Someone adept at statistics would have a mental model of standard deviation, the t-test, statistical testing, etc. Teaching students rules or formulas does not develop mental models. Bob F.
Re: Correlation conversion
On 22 Dec 1999 12:07:58 -0800, [EMAIL PROTECTED] (W. Keith Moser) wrote: Actually, the original quote referred to it being easier to get a camel through the eye of a needle than to get a rich man into heaven. It's from the Bible, Matthew 9:24. Not trying to open a debate, just properly attributing a quote. Well, along with clearing up the attribution, you also may have disabused those readers who could have walked away thinking that the Bible had some lines about Pearson correlations. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html
anti-SPAM? was: [Re: Difference of means]
On Wed, 22 Dec 1999 16:18:26 +0800, "DIAMOND Mark" [EMAIL PROTECTED] wrote: ... University policy is to avoid putting email addresses that can be extracted by spammers in the body of newsgroup postings. That would seem to be a misguided and unnecessary policy. And, having just decoded your address to type it in (since Forte Agent sends an Email copy), I can say, "I will never do *that* again" unless there is some good reason. I give my address as REPLY address and in the body of my text and I have posted 15 or 20 times a week, for three years. I hardly see any SPAM at all -- a median of one or items per week, and some of those were sent to a different email address (so I know I can't blame the newsgroup for them). I hear that some bulk-mailers avoid ".edu" addresses. But I don't think all of them do. Also, my ISP weeds out the stuff that is detectable as a bulk mailing -- addressed to everyone at the ISP, or with too many dollar signs in the subject line. And if my University/ ISP can weed-out so successfully, maybe you should ask yours to ask for advice? No, by my own experience, and from what I have read elsewhere, I think that your name is far more likely to be gleaned from websites that you visit.Though, it is possible that some *other* particular newsgroups happen to be a place where Spammers harvest names. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html