[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Janet Hill <janethill73@yahoo.co.uk> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
Re: st: Re: Interpretation of contrasts |

Date |
Thu, 14 Feb 2008 08:42:28 +0000 (GMT) |

Dave, Thank you for your advice and the reference. I was wary of reducing the analysis to a oneway model but I will now pursue that approach. Janet ----- Original Message ---- From: David Airey <david.airey@vanderbilt.edu> To: statalist@hsphsun2.harvard.edu Sent: Wednesday, 13 February, 2008 4:05:25 PM Subject: Re: st: Re: Interpretation of contrasts .. I don't have the answer you asked for, but another approach. It sounds like you have one between subject factor (3 levels) and one within subject factor (5 levels) and you have a significant interaction to examine further. One way to follow up an interaction is to example the simple effects, that is, to look either within a given stress level at each repeated measure, which reduces to a oneway repeated measures model, or to look at a given time period between stress levels, which reduces to a oneway anova. It sounds like you want to do the latter, in which case just do "oneway" if time is equal to the last period. I've read that most packages use the error term from a simple oneway anova for testing the between subject simple effects following an interaction in your design, although the pooled term is more often used for testing within subject simple effects at a given between subject level following interaction. This is due to what violations are more often expected in behavioral research (Maxwell and Delaney, Designing experiments and analyzing data, 2nd edition, LEA publishers). -Dave On Feb 13, 2008, at 7:10 AM, Janet Hill wrote: > I have a repeated measures anova with 5 repeated time periods and > the response measured at 3 stress levels and I am estimating the > model as: > wsanova resp time,id(spec) between(stress) eps > test, showorder shows there to be 54 elements in the design matrix. > > My problem is that I want to look at time*stress interactions at the > last time period. I have constructed a test matrix for the > necessary constraints but I get the following error message: > . test, test(c6) > ( 1) time[5]*stress[1] - time[5]*stress[2] = 0 > ( 2) time[5]*stress[1] - time[5]*stress[3] = 0 > ( 3) time[5]*stress[2] - time[5]*stress[3] = 0 > Constraint 1 dropped > Constraint 2 dropped > Constraint 3 dropped > > . test, test(c6) mtest(sidak) > _mtest: 0 valid tests > > I can carry out this tests at other time periods. > > My question is how can I carry out this particular test. > > Thank you, > Janet > > > > > > ___________________________________________________________ > Support the World Aids Awareness campaign this month with Yahoo! For > Good http://uk.promotions.yahoo.com/forgood/ > > > * > * For searches and help try: > * http://www.stata.com/support/faqs/res/findit.html > * http://www.stata.com/support/statalist/faq > * http://www.ats.ucla.edu/stat/stata/ -- David C. Airey, Ph.D. Pharmacology Research Assistant Professor Center for Human Genetics Research Member Department of Pharmacology School of Medicine Vanderbilt University Rm 8158A Bldg MR3 465 21st Avenue South Nashville, TN 37232-8548 TEL (615) 936-1510 FAX (615) 936-3747 EMAIL david.airey@vanderbilt.edu URL http://people.vanderbilt.edu/~david.c.airey/dca_cv.pdf URL http://www.vanderbilt.edu/pharmacology * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ __________________________________________________________ Sent from Yahoo! Mail - a smarter inbox http://uk.mail.yahoo.com * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

- Prev by Date:
**st: Loop over all unique values of a variable** - Next by Date:
**Re: st: Regression question** - Previous by thread:
**Re: st: Re: Interpretation of contrasts** - Next by thread:
**st: LR-Test Tobit vs. Heckman two step** - Index(es):

© Copyright 1996–2014 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |