Question

Using both the skew and kurtosis statistics and histograms, determine if there is a significant skew and/or kurtosis for the variables that are appropriate for this analysis. Are they negatively or positively skewed? Justify your answers.

Descriptive Statistics
	N	Minimum	Maximum	Mean	Std. Deviation	Skewness		Kurtosis
	Statistic	Statistic	Statistic	Statistic	Statistic	Statistic	Std. Error	Statistic	Std. Error
Sex	101	1	4	1.55	.591	.822	.240	1.347	.476
Age	101	24	64	41.92	7.951	.081	.240	-.240	.476
YearsSick	101	2	35	16.64	7.641	.277	.240	-.432	.476
Marital Status	100	1	8	1.49	1.020	3.321	.241	16.174	.478
Valid N (listwise)	100

Answer 1

For a normal distribution, the skewness value is in the range of (-2, 2) and kurtosis is in the range of (-3, 3).

The variable sex measured by the nominal scale of measurement. For categorical variables, frequency distribution along with a pie chart or bar chart is the best reviews and ratings for the data. With skewness and kurtosis value near 0, I can say that the distribution of sex is approximately normally distributed.

The variable age is set to be approximately normally distributed. This is because the skewness value is in the range of (-2, 2) and kurtosis is in the range of (-3, 3). The corresponding histogram for age is said to follow an approximately normal distribution.

The variable years sick is also set to follow a normal distribution. This is because the skewness value is in the range of (-2, 2) and kurtosis is in the range of (-3, 3). Hence the corresponding histogram is expected to be bell-shaped.

This skewness value for the marital status is greater than 3 with Skew = 3.3 indicating that its distribution is positively skewed. The value of kurtosis for marital status is 16.17 indicating that the distribution of marital status is leptokurtic. Hence I can say that the distribution of marital status is not normally distributed, but skewed to the right. A histogram of marital status is expected to have a tail towards the right.