3
$\begingroup$

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.

Improve this question
$\endgroup$
5
  • $\begingroup$ 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? $\endgroup$
    – 410 gone
    Dec 12, 2012 at 9:01
  • $\begingroup$ 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. $\endgroup$
    – Naresh
    Dec 12, 2012 at 9:16
  • $\begingroup$ @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. $\endgroup$
    – MCM
    Dec 12, 2012 at 13:16
  • 1
    $\begingroup$ Very closely related: stats.stackexchange.com/questions/421/… $\endgroup$
    – whuber ♦
    Dec 12, 2012 at 16:26
  • $\begingroup$ Related: stats.stackexchange.com/questions/179067/… $\endgroup$
    – kjetil b halvorsen ♦
    Apr 8, 2017 at 22:29

1 Answer 1

Reset to default
2
$\begingroup$

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.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.