Question: Stereograms, revisited. Because of the skewness of the distributions of fusion times, we might consider a reexpression. Here are the boxplots of the log of fusion times. Is it better to analyze the original fusion times or the log fusion times? Explain.
