next »

[kohelet]

I'm working on preparing for another stats review session (I mentioned the first session in the SD vs AAD thread, http://chronicle.com/forums/index.php/topic,30047.0.html, which I abandoned once y'all started talking about ASCII, but I'm glad you continued to enjoy it).  This question might be more appropriate for "In the Classroom," but after that last thread, I know that just the right, well, geeks, are hanging out here.

I've spent a couple of hours looking for something I KNOW is in a book somewhere on my bookshelves:  A decision flowchart that students can use to guide selection of the appropriate statistical analysis tool.  It probably starts with something like "What are you trying to do--describe one variable or describe a relationship between two or more variables?" and eventually has stuff like "Do you have parametric data?," eventually pointing students to the Mann-Whitney test or whatever.

Here's an example of a flow chart for deciding among the varieties of t-tests--I'm looking for something more comprehensive, but the same idea:  http://tinyurl.com/ygdg6s (WARNING, if you're virus-phobic, it's a .doc file--go all the way to p. 9 to see what I'm talking about).  There's another example on pp. 20-21 of Hair, Anderson, Tatham, and Black's _Multivariate Data Analysis_ (5th edition), but it's too advanced.  SPSS's "Statistics Coach" takes you through a similar process (in fact, I'm considering just stealing their organization and making my own flowchart if I can't find the one I'm looking for).

I remember from forever ago really appreciating this flowchart as a student--it flipped learning statistics on its end, from starting with specific statistical tests to starting with the purpose of the analysis and type(s) of data.  This would be a great approach for my students--they need something practical and something to jog their memories from long-ago stats courses.

Does anyone know what I'm talking about?  Can you point me to it, or to something similar online?

[edwidge]

Kif_, after reading the first paragraph of your post, I dashed to my bookshelf and excitedly pulled out my text with its lovely flowchart. Then I continued reading your post and realized that what I have (the one in the Hair et al. text) is not what you want....

But then I pulled out something else: a photocopy of a flowchart to end all flowcharts--I think this is what you are looking for. It's the first chapter from a text (the chapter is about 28 pp. and is called "The decision tree: Questions and answers leading to appropriate statistics or statistical techniques"):

Andrews, Frank M., Klem, L., Davidson, T. N., O'Malley, P. M., & Rodgers, W. L. (1981). A guide for selecting statistical techniques for analyzing social science data (2nd ed.).

My photocopy of the first chapter doesn't include the complete publisher information, but it looks like it was published by the Institute for Social Research at the University of Michigan.

I hope this helps!

[kohelet]

Edwidge, you're awesome!  I just checked, and our library has two copies, so I'll be excitedly running there as soon as I hit "post" here.  Thanks VERY much!

[edwidge]

Aww shucks, it was nothin'. I hope it's the right source!

[kohelet]

Just for the sake of those who might stumble upon this thread in the future, I thought a follow-up post might be helpful . . .

This flowchart is insane!  It's an absolute gem.  28 pages of decision tree guiding the reader to the correct statistical technique.  28!  And then several pages of cites and a glossary.  It needs to be updated--they note a few "new" techniques in an appendix, some of which have become standard fare since 1981.  It's amazing, though.  I ended up using to it construct my own simplified (3 pages) decision tree (which I'm quite proud of!)--if a professor had done this for me as a master's student, I would have trudged through the Ph.D. at least a little more easily.

So, if you're interested, the full cite, with additional kudos to Edwidge and a standing ovation for these authors:

Andrews, F. M., Klem, L., Davidson, T. N., O'Malley, P. M., and Rodgers, W. L. (1981). A Guide for Selecting Statistical Techniques for Analyzing Social Science Data (2nd edition). Ann Arbor: University of Michigan, Institute for Social Research, Survey Research Center.

I also found the following helpful:

Field, A. (2000). Discovering Statistics using SPSS for Windows. London: Sage.

Hair, J., Anderson, R., Tatham, R., and Black, W. (1998). Multivariate Data Analysis (5th edition). Upper Saddle River, New Jersey: Prentice Hall.

Motulsky, H. (1995). Intuitive Biostatistics. Oxford: Oxford University Press.

And I'll offer an endorsement for my new favorite software--very intuitive freeware (make a donation!) for making beautiful flow charts, among other chart types:  MeeSoft Diagram Designer by Michael Vinther (http://meesoft.logicnet.dk/, also available at www.download.com).

[tamiam]

thank you thank you thank you thank you.

[edwidge]

Last night I checked on Amazon and found an updated version of this text by the same authors:

Selecting statistical techniques for social science data: A guide for SAS (1998). Published by the SAS Institute.

It looks like the decision tree is still standing in this updated version, but the authors have updated some of the stats and incorporated SAS into the mix.

I know you are perfect strangers, but somehow it feels very nice to have done a good deed this week!

[edwidge]

AND at my current torrid pace, I might be able to achieve 100+ posts soon!! Do I get three stars by my moniker then?

[kohelet]

I know you are perfect strangers, but somehow it feels very nice to have done a good deed this week!

To make you feel even a little better, I used the decision tree in class tonight and heard several under-the-breath "awesome"s, "wow"s, and "cool"s--they REALLY appreciated it.  Makes working on it in my office until 10:30 last night worth it.

Thanks for the updated cite.

[psychle]

Thank you, kif__ and edwidge, for some very helpful information. The next time I head over to the library, I'll try to remember to see if they have that book.

[psychle]

And I'll offer an endorsement for my new favorite software--very intuitive freeware (make a donation!) for making beautiful flow charts, among other chart types:  MeeSoft Diagram Designer by Michael Vinther (http://meesoft.logicnet.dk/, also available at www.download.com).

Very nice software! I played around with it tonight and it was very easy to use. I actually had a need for software to make flowchart-like figures for a recent manuscript, but I ended up just making them manually (i.e., drawing lines and aligning each individual item). It was time-consuming and annoying, but the final product was satisfactory. Anyway, thanks for the information, and perhaps I'll use this software in future work...