Using the Journal Dataset, consider the variable DP05 (days with precipitation that is more than 05 tenths of an inch). This is the number of days in each month that had 0.5 inches or more of precipitation. For example, a value of 3 in DP05 means that during three days of that month, it rained or snowed (equivalent) more than half an inch.
Draw a histogram of DP05. Does DP05 have an approximately normal distribution? Does it have an obvious skew (left or right) and any obvious outliers? If an analysis requires that DP05 have a normal distribution, do you think that your results will be valid?
Now consider the variable EMXP (extreme maximum precipitation). That is the extreme precipitation in each month, in tenths of a millimeter. For example, a value of 300 in EMXP means that it rained 30 millimeters on the rainiest day of that month.
Draw a histogram of EMXP. Does EMXP have an approximately normal distribution? Does it have an obvious skew (left or right) and any obvious outliers? If an analysis requires that EMXP have a normal distribution, do you think that your results will be valid?