3

I'm currently applying to graduate school for a career that will be greatly enhanced if I have a firm grasp of what constitutes an excellent, good, mediocre, bad, etc. experiment and results. The vast majority of the material I expect to analyze will be medical journal articles.

While I have taken the requisite courses to evaluate the articles (stats, ethics, probability, experimental design, etc.) I haven't had such a course in several years. I am curious if anyone knows of a thorough book; a sort of quick-reference guide that will help me sort out the chaff when a clear interpretation isn't possible (i.e. - studies with only a few dozen participants, studies that are not blind, studies where the results seem well interpreted but the design itself is questionable).

It doesn't need to be "How to Interpret p-Values 101", but something that would hopefully prevent me running to my old notes every time I forget what chi-squared represents in the context of an experiment.

MCM
  • 131
  • 3
  • From your last paragraph, it sounds like you want a stats cheat-sheet - a very concise summary of the sort of tests you might find in medical papers. Is that right? – 410 gone Dec 12 '12 at 09:01
  • Probability & Statistics For Engineers & Scientists, 8/E, by Wapole, Publisher: Pearson Education would be a good place to start. Its what we(undergraduates) were asked to use in our introductory class on Probability and statistics. –  Dec 12 '12 at 09:16
  • @EnergyNumbers - That's about right. I'd prefer something a little heftier that addressed some of the murkier aspects of experimental design as well, but I can usually spot questionable design better than questionable conclusions. –  Dec 12 '12 at 13:16
  • 1
    Very closely related: http://stats.stackexchange.com/questions/421/what-book-would-you-recommend-for-non-statistician-scientists – whuber Dec 12 '12 at 16:26
  • Related: http://stats.stackexchange.com/questions/179067/what-should-a-graduate-course-in-experimental-design-cover/179073#179073 – kjetil b halvorsen Apr 08 '17 at 22:29

1 Answers1

2

Statistics Explained by Hinton (ISBN: 0415332850) is a pretty good reference for something like this.

There's also a little bit meatier of a book "Statistics for the Behavioral Sciences" by Gravetter (ISBN: 0495602205) which I thought was easy to follow and refer to but you may find less useful because it's phrased more for psychology types of students.

Zai
  • 153
  • 4