Using traditional ANOVA, you'd have to drop either cases or time points with missing data. Using linear mixed effects analysis, you'd be able to use all the data. LME also has the benefit of *not* assuming sphericity, which is good for data like yours (many measures across few cases) where the traditional ANOVA sphericity assumption is unlikely to hold.
Your dependent variable, % valid, suggests that there's some more raw representation of the data that may be better to look at. For example, if % valid is derived from, say, the success/failure rate of 10 observations per sample/timepoint, you might want to take a look the lme4 package (as suggested in a previous post: https://stat.ethz.ch/pipermail/r-sig-mixed-models/2008q3/001160.html ) On Tue, May 12, 2009 at 6:03 AM, Alan O'Loughlin <olo...@wyeth.com> wrote: > Hello, > > I'm trying to do a comparsion on a large scale say 10L bottle of liquid and a > small scale bottle of liquid 0.5L, I have 5 different samples from each and > they are measured over the space of 8 days as % viability and the % viability > decreases over time. However not all 10 samples got measured every day. How > would I do a two-way anova on this in R? > > Thanks for any help. > > Regards, > Al > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Mike Lawrence Graduate Student Department of Psychology Dalhousie University Looking to arrange a meeting? Check my public calendar: http://tr.im/mikes_public_calendar ~ Certainty is folly... I think. ~ ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.