Re: Fisher's exact test for time allocation data

Bruce Weaver wrote:
Ray, apart from there being more categories, isn't this the same as
Example 1 in the Shaffer (1981) paper you've pointed to in the past
(reference below)? In that example, faculty members report the
proportions of time spent in teaching, research and administration.
Faculty members are sorted into groups by discipline (rather than
male/female), and the recommended approach is to have all 3 proportions
(summing to 1) as repeated measures in a mixed design ANOVA (with
discipline as the between-Ss factor).

Shaffer, JP. Journal of the American Statistical Association, Vol. 76,
No. 375 (Sep., 1981), 607-611. [available through JSTOR]

Yes, this is exactly the kind of data that Shaffer's paper is about.
To check for overall differences between the groups, he should look
at the Groups x Categories F-test in the anova. To check a specific
category, he should do a t-test (or a one-way anova, if there are
more than two groups) on the proportions for that category.

.

Relevant Pages

  • Hmm, thanks for the reference!
    ... I think I might be able to manage the ANOVA. ... Example 1 in the Shaffer paper you've pointed to in the past ... proportions of time spent in teaching, ... Faculty members are sorted into groups by discipline (rather than ...
    (sci.stat.edu)
  • Re: Repeated measures ANOVA ?
    ... >> I am mainly interested in determining wether or not the animals show a ... >> assumptions of ANOVA. ... > a separate d.v./analysis with its own error term. ... Once the first two proportions are known for a subject, ...
    (sci.stat.consult)
  • Re: Within-subjects proportional data
    ... > Non-linear mixed model with logit link! ... proportions is a lot simpler and would be a useful starting point ... before doing the more complex analyses. ... I'd also consider a simple repeated measures ANOVA looking at the ...
    (sci.stat.consult)
  • Re: The propotion mean...
    ... > If my data belong to this type, ... You can just about always "use ANOVA." ... Do the differences in proportions look like "equal intervals"? ...
    (sci.stat.edu)