Several folks are making the argument based on the 'anchor of history', that students won't be able to understand the existing research literature without knowing the fundamentals of null hypothesis testing and its manifestations through t-tests, ANOVA, etc. They are right, unfortunately; that argument carries a lot of weight.
Which is what takes us to the crux of the matter. Usually, by the time you've gotten to ANOVA, in one semester, you don't have any time for anything else. If the students are as motivated and bright as you suggest, I think that a week or two of the Bayesian approach might be quite a good thing to present, if the time is available after teaching the things they need to know to survive the existing literature. I was taught Bayesian approaches as a side note in my stats classes in graduate school. I don't know if it gets greater treatment now or not. One of my concerns about teaching null hypothesis testing, the way it is presented in nearly all books, is that students never quite get it. There are too many hypotheses for many of them to keep straight and they can get quite confused about it (it shows up when you ask them to interpret the null hypothesis rejection decision in the context of the originally posed research question). They reject the null hypothesis and write, 'therefore there is no effect of the treatment on the dependent variable'. They think they are testing the scientific hypothesis, no matter how many times they are told they aren't (and by the end of the semester they've been told that more or less 6 times: one sample z-test, one sample t-test, independent t-test, dependent t-test, one way ANOVA, test of correlation, etc.) In any case, I think that it is probably a good thing to give the students a sense of the bigger world of stats beyond the basics they are taught. I just wish we all had time to do so with enough depth to make it worth their while. Paul On May 19, 2011, at 9:11 PM, Gerald Peterson wrote: > I think even those arguing for some Bayesian stats also emphasize the need to > know which and when null testing and Bayesian analyses are appropriate. I > think you need to integrate and illustrate both in your text. > > > G.L. (Gary) Peterson,Ph.D > Psychology@SVSU > > > On May 19, 2011, at 7:36 PM, Marc Carter <[email protected]> wrote: > >> Hi, all -- >> >> Next year I've planned on developing a stats/methods integrated text (I have >> some sabbatical time). More and more, though, lately I've been reading that >> "we're doing stats wrong" and need to start moving to Bayesian stats. I >> understand and appreciate the arguments. I think they're right. The recent >> Psych Science has a bevy of articles about it, exacerbated, I'm sure, by >> Bem's JPSP article. >> >> Our program is essentially a grad-school-prep program, and the text will be >> for these students: all grad-school-bound, and smart. But most are going >> into the helping, rather than research-side of psych. But they'll get PhDs. >> >> Can I get a show of hands to help me decide whether or not I should a) >> include only Bayesian hypothesis testing, 2) both trad and Bayesian hypoth >> tests, or iii) just the trad stuff. >> >> It's a year-long course with a lab (I get them 6 hours a week for a year), >> and right now they come out knowing things all the way through mixed-model >> factorial ANOVA. >> >> Should I back off the hard-core experimental design (ANOVA) and move toward >> this recent (sorta) issue about how we have been doing hypothesis tests? >> >> What thinkest thous? >> >> m >> >> >> >> ------ >> Do not judge me before going to www.damnyouautocorrect.com. >> >> >>> >> >> The information contained in this e-mail and any attachments thereto >> ("e-mail") is sent by Baker University ("BU") and is intended to be >> confidential and for the use of only the individual or entity named above. >> The information may be protected by federal and state privacy and >> disclosures acts or other legal rules. If the reader of this message is not >> the intended recipient, you are notified that retention, dissemination, >> distribution or copying of this e-mail is strictly prohibited. If you have >> received this e-mail in error please immediately notify Baker University by >> email reply and immediately and permanently delete this e-mail message and >> any attachments thereto. Thank you. >> >> --- >> You are currently subscribed to tips as: [email protected]. >> To unsubscribe click here: >> http://fsulist.frostburg.edu/u?id=13445.e3edca0f6e68bfb76eaf26a8eb6dd94b&n=T&l=tips&o=10648 >> or send a blank email to >> leave-10648-13445.e3edca0f6e68bfb76eaf26a8eb6dd...@fsulist.frostburg.edu >> > > > --- > You are currently subscribed to tips as: [email protected]. > To unsubscribe click here: > http://fsulist.frostburg.edu/u?id=13441.4e79e96ebb5671bdb50111f18f263003&n=T&l=tips&o=10652 > or send a blank email to > leave-10652-13441.4e79e96ebb5671bdb50111f18f263...@fsulist.frostburg.edu --- You are currently subscribed to tips as: [email protected]. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=10653 or send a blank email to leave-10653-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
