'Stats Sample' During Interview Day

[lovepsych]

Hello,

I received an interview day itinerary from a School Psychology Ph.D. program and it states that there will be a "short stats sample in the morning to supplement the materials you submitted with your application." Does anyone know what this means or experienced something similar during their interview? I'm slightly freaking out and would appreciate any advice! Thank you.

[cheesecafe]

I am also applying to graduate programs for the first time, so I don't know exactly what you will encounter. However, I did once have an interview for a school psychology research position that was similar to this, where I had a short practical component. It consisted of reading a report--similar to the results section of any academic paper--and answering questions based on the findings. If you took the psych GRE, it felt like some of the questions there (see questions 13-15 here for an example). Because I had specified on my CV that I knew how to use SPSS and Stata, the practical component also included generating basic descriptive analyses in both programs based on a sample dataset.  

If multiple people are being interviewed, I can't imagine your stats sample would be anything that would take too long to grade afterwards. Good luck!

Edited December 28, 2016 by pbjcafe

[fuzzylogician]

Sounds to me like they'll ask some more questions about your background so they can do better statistical analysis of their applicant pool once the cycle is over. I don't think this means some kind of test. However, the best thing you can do is reply to the email you got and ask if they can clarify what they mean by that sentence. (Or call them, if you would rather ask the question anonymously, though I don't think there is any reason to.) 

[spunky]

We have something similar in our department. We use it mostly to try and gauge how many introductory statistics courses (and of which type) will be needed once the new student cohort begins its semester.

Because people come to grad school with somewhat different backgrounds in terms of their methodological training (I’ve seen everything from people having introductory grad-level courses in their transcripts as undergrads to the ones who barely passed intro to research methods) something like a quick practice test/interview helps us know the overall level from which people are starting from to plan ahead in terms of the courses they’ll need.

I really don’t think this is anything to worry about as far as whether or not they’ll like rescind your invitation to their program or anything.

[AAdAAm]

I have never heard of something like this before. Is it a common thing to ask new potentials students to do? Or is this specific to School Psychology?

[spunky]

On 12/29/2016 at 8:51 AM, AAdAAm said:

I have never heard of something like this before. Is it a common thing to ask new potentials students to do? Or is this specific to School Psychology?

I don’t think it’s very common (yet) but I wouldn’t be surprised if something like this starts becoming more and more popular as time goes by.

I mean, there’s already research out there claiming that the level of statistical/methodological training one gets in Psychology graduate programs is… well… let’s just say not the greatest. Dr Leona Aiken’s famous survey comes to mind, although that’s a tad bit dated now. It’s also not a surprise that the overall misuse of statistics has, in part, contributed to the whole brouhaha of the Crisis of Replicability which for better or worse has been scaring people left and right (methodological terrorism, anyone? ) . So I sense that the discipline is (slowly) trying to make methodological/statistical training a little more central to the formation of psychologists as academics. 

Last year in the American Psychological Society, APS, annual meeting I remember a lot of post-workshop talks focused on how we can improve statistical and methodological training in graduate programs and a suggestion that kept coming up was identifying, from very early on, the different groups of students with different levels of expertise as far as statistics and methodology goes. My guess is that maybe programs can tailor their training to the different abilities that students have in order to maximize their time in graduate school? Who knows, a lot of things are up in the air right now. The consensus, however, does seem to be that the way in which things have been done in the past is no longer enough and something needs to change.  

[Piagetsky]

8 hours ago, spunky said:

I don’t think it’s very common (yet) but I wouldn’t be surprised if something like this starts becoming more and more popular as time goes by.

I mean, there’s already research out there claiming that the level of statistical/methodological training one gets in Psychology graduate programs is… well… let’s just say not the greatest. Dr Leona Aiken’s famous survey comes to mind, although that’s a tad bit dated now. It’s also not a surprise that the overall misuse of statistics has, in part, contributed to the whole brouhaha of the Crisis of Replicability which for better or worse has been scaring people left and right (methodological terrorism, anyone? ) . So I sense that the discipline is (slowly) trying to make methodological/statistical training a little more central to the formation of psychologists as academics. 

Last year in the American Psychological Society, APS, annual meeting I remember a lot of post-workshop talks focused on how we can improve statistical and methodological training in graduate programs and a suggestion that kept coming up was identifying, from very early on, the different groups of students with different levels of expertise as far as statistics and methodology goes. My guess is that maybe programs can tailor their training to the different abilities that students have in order to maximize their time in graduate school? Who knows, a lot of things are up in the air right now. The consensus, however, does seem to be that the way in which things have been done in the past is no longer enough and something needs to change.  

"More vampirical than empirical" is one of the funniest lines I've ever read.  Agree that there needs to be a major overhaul in stats training in psychology starting at the late undergrad level.  What Diederik Stapel was able to get away with is criminal, and the fact that major journals fell for it is even more astonishing.  

[transfatfree]

10 hours ago, spunky said:

I don’t think it’s very common (yet) but I wouldn’t be surprised if something like this starts becoming more and more popular as time goes by.

I mean, there’s already research out there claiming that the level of statistical/methodological training one gets in Psychology graduate programs is… well… let’s just say not the greatest. Dr Leona Aiken’s famous survey comes to mind, although that’s a tad bit dated now. It’s also not a surprise that the overall misuse of statistics has, in part, contributed to the whole brouhaha of the Crisis of Replicability which for better or worse has been scaring people left and right (methodological terrorism, anyone? ) . So I sense that the discipline is (slowly) trying to make methodological/statistical training a little more central to the formation of psychologists as academics. 

Last year in the American Psychological Society, APS, annual meeting I remember a lot of post-workshop talks focused on how we can improve statistical and methodological training in graduate programs and a suggestion that kept coming up was identifying, from very early on, the different groups of students with different levels of expertise as far as statistics and methodology goes. My guess is that maybe programs can tailor their training to the different abilities that students have in order to maximize their time in graduate school? Who knows, a lot of things are up in the air right now. The consensus, however, does seem to be that the way in which things have been done in the past is no longer enough and something needs to change.  

2

Thanks for sharing the survey. It's an interesting read. I'm not sure if much can be done as there aren't many quantitative psychologists to begin with (as shown clearly in the survey. There are two quantitative psychologists in my department but both are Europeans). And many soft scientists just follow statistical conventions without understanding the underlying principles and remain users of statistics only.

I also agree that changes must be made at the undergraduate level as many programs (except for APA-accredited programs that have many more requirements) only require one or two years of coursework. It may be difficult to fit stat/method courses into the curriculum.

[OhSoSolipsistic]

I'd say just go in with what you know. As others have said - it sounds like more of an assessment to gauge if they'll need to form a course on either intro or moderate stats. I think a good chunk of first year graduate students in psych - if they've had a mehods and intro stats class - have a basic understanding of descriptive stats (GRE stuff) and perhaps t-tests and pearson's correlations, but nothing beyond basic ANOVAs or linear regressions. But especially if the program is accepting non-psych undergrads, I really don't think it's an assessment of ability. If you really want to study for it, I'd just review mean, median, mode, standard deviation, t-test, and person's correlation. They might ask about z-scores or histograms or confidence intervals or other stuff, but you really shouldn't be expected to know that off the top of your head (or very likely anything else). Hell, I need to review those terms and I've been working with hairy within-subjects doubly multivariate analyses for the past couple years. Good luck!

[spunky]

On 1/2/2017 at 8:05 PM, transfatfree said:

Thanks for sharing the survey. It's an interesting read. I'm not sure if much can be done as there aren't many quantitative psychologists to begin with (as shown clearly in the survey. There are two quantitative psychologists in my department but both are Europeans). And many soft scientists just follow statistical conventions without understanding the underlying principles and remain users of statistics only.

I also agree that changes must be made at the undergraduate level as many programs (except for APA-accredited programs that have many more requirements) only require one or two years of coursework. It may be difficult to fit stat/method courses into the curriculum.

Well, there are other ways (aside from increasing the number of quantitative psychologists) and many of them have already been put in place. Given the relevance that publishing has in academia, journal editors can have a lot of influence in shaping the way in which analyses are carried on and results are presented. For instance, during last year’s annual meeting of the Society for Personality and Social Psychology, Dr. Simine Vazire detailed some of the new policies and practices regarding data analysis/methodology that are being enacted in the journals where she serves as an editor in order to improve the quality of manuscripts they receive. Statcheck, in spite of its controversy, is becoming very popular very quickly for editors to automatically check whether or not the reporting of statistics in APA-formatted journals was done right. And I do know for a fact that this is being used by journals because I’ve been contacted by professors in the past asking me about these rather cryptic emails they get regarding whether or not they had "passed" or "failed" Statcheck.

There aren’t many quantitative-type psychologists, when compared to the more mainstream subfields like clinical or social/personality, that is definitely true. But this situation is also changing. The relevance that data has in society both inside and outside academia (that buzzword “big data” comes to mind) has definitely brought in a renewed interest in statistics or this new thing, “data science”. SAGE Publishing just conducted a study last year (in which I participated) regarding the role that big data will play now in psychology and the social sciences. They’re supposedly preparing a journal devoted entirely to this. Combine this excitement about data with the change in paradigm that the Crisis of Replicability is bringing about, and I truly thing it’s just a matter of time before statistics/methodology starts playing a more central role in the formation of academic psychologists. Heck, even in this forum I have seen the change. When I first joined, we barely ever had any questions related to Quant Psych/psychometrics. And now I’m pretty sure this is the first time ever we’ve actually had a quant psych applicants’ thread! I’m sure that this change will not happen overnight (although who knows, some exciting stuff is happening very quickly like the OP mentioned in the beginning) but I do think it’s coming. We need it.  

Edited January 6, 2017 by spunky

[Quantitative_Psychology]

I bet this thread is a little bit dead now. What ever ended up happening? I was curious if the "stats" was interpreting an output, doing a hand calculation, or interpreting results from a paper? I think it would be kind of silly if it was hand calculations, but I could see some interpretation.

@spunky is probably right that it had something to do with identifying different levels of incoming students and what classes are needed. At my Master's program there were wildly different levels in stats abilities upon entering, and I can only imagine it being equally as problematic for PhD programs.

Edited February 10, 2017 by mav160

[meep95]

On 12/27/2016 at 6:38 PM, lovepsych said:

Hello,

I received an interview day itinerary from a School Psychology Ph.D. program and it states that there will be a "short stats sample in the morning to supplement the materials you submitted with your application." Does anyone know what this means or experienced something similar during their interview? I'm slightly freaking out and would appreciate any advice! Thank you.

 

what did the stats sample end up being

[lovepsych]

It resembled a worksheet of GRE-esque quantitative problems. Fairly simple stuff that any psychology major should know. I only got 2 wrong and ended up receiving an admissions offer!! It definitely all worked out