STA 550 Testing for Multiple Regression

Assignment 1: Testing for Multiple Regression

Institutional Affiliation:

Student’s Name:

Date:

Research question

Do education category, gender, and employment status have a significant impact on internet usage?

Hypothesis

Null hypothesis: Education category, gender, and employment status have no significant impact on internet usage.

Alternative hypothesis:Education category, gender, and employment status have a significant impact on internet usage

Results

To determine whether education category, gender, and employment status have a significant effect on internet usage, a multiple linear regression was used. The dependent variable was internet usage while independent variables were; education category, gender, and employment status.

From table 1, R-square =0.226 means that 22.6% of the changes in the internet usage is explained by education category, gender and employment status of the respondent. The value of R-square is below 50% hence the model is not best fitted(Freedman, 2005). Table 2, illustrates the overall relationship between dependent and independent variables. From ANOVA table F=4875.148 corresponding to p-value=0.000 which is less than 0.05 .Hence we reject the null hypothesis and conclude that education category, gender, and employment status have a significant impact on internet usage. Table 3 illustrates the impact of each independent variable on internet usage. There is a significant negative relationship between gender and internet usage (p-value=0.000).The findings indicate that females use less internet than male students. An increase in the education by one level leads to increase in internet usage by 0.278units.Educational level is a significant factor in explaining internet usage( p-value=0.000<0.05).Employed persons use the internet more than unemployed individuals. A change from unemployed to employed increases person's internet usage by 0.066 units. Employment status is a significant factor in explaining internet usage (p-value=0.000 less than 0.05).Social implication is that education level, gender, and employment status affects internet usage of the respondents.

Table 1:

Model R R Square Adjusted R Square Std. Error of the Estimate
1 .475a .226 .226 .51116

Table 2

Sum of Squares df Mean Square F Sig.
1 Regression 3821.411 3 1273.804 4875.148 .000b
Residual 13110.766 50178 .261
Total 16932.177 50181

Table 3:

Standardized Coefficients t Sig.
B Std. Error Beta
1 (Constant) .913 .009   106.969 .000
Q101. Gender of respondent -.037 .005 -.032 -8.052 .000
Education Category .278 .002 .455 112.240 .000
Employment Status .066 .005 .053 13.111 .000

REFERENCE

David A. Freedman (2005). Statistical Models: Theory and Practice, Cambridge University Press.