0

Possible Duplicate:
Difference between generalized linear models & generalized linear mixed models in SPSS
Repeated measure problem

I have 6 subjects. For each subject, we measured two variables over six-week period once a week. Let’s call the variables A and B. Variable A is binary (male or female), and variable B is also discrete (blood pressure: High, medium, and low). Now I want to see if females have more highe blood pressure levels than males. What kind of statistical method do you suggest here (keep in mind that we’re dealing with repeated measure)?

Community
  • 1
user9292
  • 1,324
  • 2
  • 17
  • 31
  • If you just want to see if females have more high blood pressure than males, then you can probably ignore the repeated data and just take the proportions at time 1, or the average over all the times, or something similar. But since you have repeated measures, you can ask more interesting questions including whether male and female blood pressure changed in different ways over time. – Peter Flom Dec 30 '12 at 23:33
  • Thanks for the note, Peter. This is different from the previous problem in the sense that both variables are discrete. – user9292 Dec 30 '12 at 23:36
  • OK, well, if you have a continuous variable you should not categorize it. That will lead to loss of information and an increase in type I and type II error. – Peter Flom Dec 30 '12 at 23:39
  • This is a different problem. I'm not categorizing any continuous variable :-) – user9292 Dec 30 '12 at 23:45
  • Oh, OK. Sorry. They just seemed so similar. – Peter Flom Dec 31 '12 at 00:04
  • I agree w/ @PeterFlom, someone is categorizing a continuous variable; *blood pressure* (a continuous variable) has been turned into "High, medium, & low" (discrete categories). However, we can also ignore these details if you're just wondering about how to analyze repeated discrete measures in general. The answer requires specifying how you're thinking about the problem. I discuss those issues here: [difference between generalized linear models & generalized linear mixed models](http://stats.stackexchange.com/questions/32419//32421#32421). More info exists at other threads linked there. – gung - Reinstate Monica Dec 31 '12 at 03:31

0 Answers0