I'd like to point out that, in addition to the report by ETS, the company that created GRE testing, there have been many data-based GRE validity studies published in peer-review journals. For instance:
--- Edith L., Goldberg and George M. Alliger (1992): Assessing the validity of the GRE for students in psychology: a validity generalization approach. Major findings: "Overall, the results from this meta-analysis do not paint a particularly favorable picture for the validity of the Graduate Records." --- Morrison, T and Morrison, M (1997): A meta - analytic assessment of the predictive-validity of the quantitative and verbal components of the graduate record examination with graduate grade-point average representing the criterion of graduate success. Major findings: "The results of this meta-analysis suggest that the quantitative and verbal components of the GRE possess minimal predictive validity... When this finding is coupled with studies suggesting that performance on the GRE is age-, gender-, and race specific (House, 1989; Kaczmarek & Franco, 1986; Scheuneman, 1987), the use of this test as a determinant of graduate admission becomes even more questionable." -----Original Message----- From: Ecological Society of America: grants, jobs, news [mailto:[email protected]] On Behalf Of Terryhbell Sent: Friday, September 05, 2014 7:45 AM To: [email protected] Subject: Re: [ECOLOG-L] FW: [ECOLOG-L] GRE Scores In Picking a PhD Student. Do they Matter? You can question the source, but here is one study on the subject: https://www.ets.org/research/policy_research_reports/publications/report/2005/hsiu Terry Sent from my iPhone > On Sep 4, 2014, at 8:39 PM, "Ganter, Philip" <[email protected]> wrote: > > Well, lets examine the reasoning in this last post (for convenience -- > parallels can be found in several posts in this thread). GREs don't work for > some. Scrap them. So we establish the precedent that, if a measure does not > always predict the ability of an applicant, it is worthless. > > Lets apply our algorithm to other measures commonly used as factors in > acceptance decisions. GPA? Grade inflation, out with it! Personal > statements? Often written by committee of applicants friends and relatives. > Out with it! Publications? Many undergraduates go to schools without > undergraduate research opportunities and must use what time they have after > going to class and studying working to pay the rent and food bills, so out > with publications. Require an undergraduate degree? It is not impossible > that an applicant, studying on his or her own, could educate his- or herself > adequately enough to do well in graduate school (Recall famous scientists > without terminal degrees), so out with degrees. > > In fact, this razor will cut away any attempt to evaluate applicants and we > have reached the last post's ideal: total equality in education. All > applicants simply assigned numbers and acceptances meted out with a random > number table. > > I was a student from a small school with absolutely no reputation in science > (deservedly so). After being accepted into grad school, I was told that it > was my GRE scores that had been the decisive factor. Odd that. That evil > company actually contributing to a poor student's (I was unemployed and very > poor when the acceptance letter arrived) opportunity at grad school. > Anomaly? Evidence for the utility of GREs? Just another anecdote? > > This thread has gotten to be just grousing. The original post asked an > interesting question. What to do when indicators disagree? No one has > posted a really good answer to that conundrum (guess that's what makes it a > conundrum). Everyone seems to be willing to contribute an anecdote but we > aren't politicians, we're scientists. Anyone got any data? > > Phil Ganter > Biological Sciences > Tennessee State University > Nashville, TN > > > ________________________________ > From: Andrew Wright <[email protected]> > Reply-To: Andrew Wright <[email protected]> > Date: Thu, 4 Sep 2014 12:06:32 -0500 > To: <[email protected]> > Subject: Re: [ECOLOG-L] GRE Scores In Picking a PhD Student. Do they Matter? > > Some people just don't test well, making the GREs totally useless as a > gauge of talent across all. Furthermore, I have been told that their > use is supported mainly by payments to the Universities from the > company that runs the GREs, at the costs to the already poor students. > They seem merely to be a commercial enterprise aimed at exploiting > students, rather than a reliable indicator of ability. I feel they > should be scrapped as another (albeit relatively minor) economic barrier to > equality in education. > > -- > Andrew Wright, Ph.D. > > "We don't have to save the world. The world is big enough to look > after itself. What we have to be concerned about is whether or not the > world we live in will be capable of sustaining us in it." Douglas > Adams > > > On 4 September 2014 07:55, Judith S. Weis > <[email protected]> > wrote: > >> Yes, but.... >> I have had a number of foreign students who could not write English >> very well and I had to do a lot of re-writing on their dissertations >> - but the research itself was excellent and we produced many >> publications. Just more work on the major professor's part. >> >> >> >> >>> I agree with this assessment - especially since some small liberal >>> arts colleges engage in grade inflation - GPA's are not always >>> reliable. I think there is considerable value to the GRE scores and >>> having a minimum is useful. Above that, scores vary widely and are >>> not always predictive of ultimate success. The most important thing >>> that should be assessed - and the GREs do not do an adequate job >>> here - is writing ability. Even mediocre students can complete a >>> research project and muddle through the data analysis, but when it >>> comes to writing, the grain and chafe fall into two distinct piles. >>> The worse thing you can do for your career is to take on mediocre >>> students with poor writing skills. If a project is never published >>> then it will count for zero to your CV and career development. I >>> suggest getting the student to send you a writing sample, or >>> evaluate their writing skills based on the materials they have submitted. >>> Mitch Cruzan >>> >>> >>>> On 9/3/2014 6:07 AM, Gary Grossman wrote: >>>> I think that we all look at this issue from a personal perspective, >>>> especially those that did well on standardized tests, and I've had >>>> this same argument with colleagues for 30 years, including the >>>> exact same situation where the student was up for a competitive >>>> assistantship with a mediocre GRE score and a senior-authored >>>> publication in an international journal. You don't tell us how low >>>> the score was and I'd be concerned if it was a low quantitative >>>> score, because grad students need to have a good quantitative >>>> background. But for researchers, publications are the sine quo non >>>> and render a low GRE score moot, provided the student actually >>>> earned the senior authorship (we don't have that info either and I >>>> view senior authorship differently than junior authorship, >>>> especially if there are more than two authors). The one valid >>>> argument that the "keepers of the gates" regarding the GRE is that >>>> it is the one evaluator that is equivalent across all >>>> applications,i.e., as faculty we don't have the time to evaluate if >>>> an A at Furman University is the equivalent of an A at Chapel Hill. >>>> But in the end I've found that the GRE isn't very indicative of >>>> performance by a researcher (I mean really, how could it be, it >>>> contains no information on motivation, persistence, intuition or >>>> many other characteristics that great researchers have). In fact, >>>> I've seen some of the biggest flops as graduate students come from >>>> students with very high GRE scores --- they just happen to be good >>>> at taking standardized tests but not necessarily at research. My >>>> own story -- I took the GRE in 1975 and earned somewhere between >>>> 1150 and 1190 can't remember exactly, but I do remember it was a >>>> mediocre score. I have 110+ journal articles, including multiple >>>> papers in Am. Nat, Ecology, Ecol. Monogr, Oecologia, Freshwater >>>> Biol. etc. The math is pretty easy to do <g>. cheers, g2 >>>> >>>> >>>> On Tue, Sep 2, 2014 at 2:03 PM, Alex M. L >>>> <[email protected] >>> >>>> wrote: >>>> >>>>> Last weekend I got into a long discussion on the value of GRE >>>>> score in a PhD student. As the 2015 applicants start, I open up >>>>> the discussion to the >>>>> community: >>>>> >>>>> I have a female student that has both a Masters (thesis) and >>>>> publication with several years research experience. However, her >>>>> GRE score are quite poor. >>>>> Should I really pass up a seemingly great applicant because of low >>>>> scores? >>>>> >>>>> If a student has a biology Masters or a publication... do GRE >>>>> scores matter? >>>>> Have we not moved past GRE scores when picking the next round of >>>>> PhD researchers for our lab(s)? >>>>> >>>>> If you have a personal story of low scores and still attaining >>>>> your PhD or accepting a similar student... I would love to hear >>>>> from you! >>>>> >>>>> Cheers! >>>>> Alex M.L >>> >>> -- >>> -------------------------------------------------------------------- >>> ---- >>> Mitch Cruzan >>> Professor of Biology >>> Portland State University >>> Department of Biology, SRTC rm 246, PO Box 751 Portland, OR 97207 >>> USA http://web.pdx.edu/~cruzan/ >>> -------------------------------------------------------------------- >>> ---- > > > ------ End of Forwarded Message
