Applied Linear Regression - Essay Example

The author mentions that the range of values of satisfaction for the variable can range from a minimum of 3 to a maximum of 15. The chi – the square is tested for a significant difference in satisfaction between gender and majors. The hypothesis tested by the chi-square is to determine whether different samples are different enough in some characteristics. The statistical test to be used in the case is the probability test. The variables are gender which is a dependent variable and majors which are independent variables (Weisberg 2005).

The probability test articulates that there is no interaction because there is an intersection among the three majors (finance, accounting, and management) among the male and female where 0 is observed data and e is expected data. The variables are gender which is a dependent variable and the majors (finance, accounting, and management). If the chi-square value obtained is larger than the critical value at any chosen probability fault threshold, then data presented a statistically significant correlation between the variables that are used in the test (Walker & SAS Institute2010).

In regression -0.3 to 0 shows a weak spread, 0 to 0.3 shows a moderate spread of data along the regression line and above 0.3 shows a concentrated spread about the regression line. So 0.3 shows a moderate spread around the regression line. Kurtosis is the measure of skewness. A zero kurtosis is mesokurtic, a positive kurtosis is leptokurtic and negative kurtosis is platykurtic. Therefore, kurtosis = 0.3 is mesokurtic which is a normal distribution. Durbin Watson statistic has values from 0 to 4.

Whereby value 2 shows no autocorrelation in a sample selected. Any value approaching 0 shows positive autocorrelation while 4 shows a negative autocorrelation. So, in this case, d=0.3 indicates a positive autocorrelation. Shapiro – Wilks w is a measure of normality of random sample. For instance, if the value of p is more than 0.05 then it assumed to have a normal distribution hence accept the normality.