Hi everyone, I am new to this forum and also new to statistics and I would appreciated it if someone would take some time to answer my question.
I am analyzing companies in regard to their leverage. I categorized the companies into 3 groups: small, mid and large. For the group small, I have 55 debt multiples, for mid 42 and for large 72. (Unfortunately I can not provide my data because it is confidential.) I am now trying to find out whether the mean debt multiples are significantly different for the 3 different groups. For this reason I calculated an anova table with the aov function and to display the results for each "pair" I did the tukey.hsd function. Now my question: Am I allowed to use these functions given that my data is unbalanced? Can use I read several times that aov is only valid for balanced data? If not, is there another function that I can use? Thank you very much for your answers. Call: aov(formula = Debt.Ebitdax ~ Company.Size, data = Anetdebtx2003) Terms: Company.Size Residuals Sum of Squares 302.3089 926.2174 Deg. of Freedom 2 166 Residual standard error: 2.362123 Estimated effects may be unbalanced > summary(Anovanetdebtx2003) Df Sum Sq Mean Sq F value Pr(>F) Company.Size 2 302.3 151.15 27.09 6.58e-11 *** Residuals 166 926.2 5.58 --- Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 > TukeyHSD(Anovanetdebtx2003) Tukey multiple comparisons of means 95% family-wise confidence level Fit: aov(formula = Debt.Ebitdax ~ Company.Size, data = Anetdebtx2003) $Company.Size diff lwr upr p adj Mid Market Buyout-Large Buyout -1.446292 -2.530922 -0.3616617 0.0054123 Small Buyout-Large Buyout -3.112143 -4.112545 -2.1117420 0.0000000 Small Buyout-Mid Market Buyout -1.665852 -2.810574 -0.5211300 0.0021037 -- View this message in context: http://r.789695.n4.nabble.com/Anova-tp4650940.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ 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.
