Hi there,
I am currently running an analysis for a master's psychology dissertation and I am conducting a one-way ANOVA. My Levene's (p<0.001) and Shapiro-Wilk's (p=0.016) tests have come back significant. My Q-Q Plot looks relatively normal, however. Based on this I was going to run a Welch test.
I have been using this book - https://davidfoxcroft.github.io/lsj-boo ... -way-anova
I have found it excellent and one paragraph for homogeneity of variance (Levene's) states:
"Similarly, if your sample size is quite small, then the homogeneity of variance assumption might not be satisfied and yet a Levene test could be non-significant (i.e. p > .05). What this means is that, alongside any statistical test of the assumption being met, you should always plot the standard deviation around the means for each group / category in the analysis…just to see if they look fairly similar (i.e. homogeneity of variance) or not."
I don't really understand how to plot the standard deviation around the means for each group. Is this the QQ Plot?
Thanks!