Factorial ANOVA HW Key
On question #1, how can you tell from the Q-Q plot or box plot that the distribution is negatively skewed? (I have not taken the test yet)
Thank you!
Brooke
posted by bfullmer @ 8:45 PM
This blog is for the use of students enrolled in either section of PG851. It is intended to be a forum for asking, discussing, clarifying, and helping.
posted by bfullmer @ 8:45 PM
1 Comments:
At 9:51 PM,
David said…
You can roughly tell the skew from the boxplot by looking at the position of the box (lines defined by the 25th and 75th percentiles) relative to the whiskers/outliers.
In the example with the recoded variable, the box is still slightly higher (more positive) than the middle of the whiskers. Thus it is still negatively skewed.
But skewness is probably the easiest to see when you plot the histogram of the variable and superimpose a normal curve on top.
Though the plots are sufficient, the most quantitative way would be to look at the skewness value. You can ask for it through Analyze -> Descriptive Statistics -> Explore -> Plots -> Normality Plots w/ tests. (This way you can also get your histogram). A negative skewness statistic confirms the conclusion from the plots.
Post a Comment
<< Home