Scribd
Upload
8 views

Multiple linear regression analysis with Stepwise method

Multiple linear regression analysis with Stepwise method

Uploaded by

manfoosah khalid
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
8 views

Multiple linear regression analysis with Stepwise method

Multiple linear regression analysis with Stepwise method

Uploaded by

manfoosah khalid
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Multiple linear regression analysis with Stepwise method

Example 1

Dependent Variable:

Job satisfaction

Predictors:
Age (continuous)

Income (continuous)

Education Level (categorical: High School/Bachelor's/Master's)

Hypothesis:

There is a relationship between job satisfaction and age, income, and education level

1. H1: Age is positively associated with job satisfaction.

2. H2: Income is positively associated with job satisfaction.

3. H3: Education level is positively associated with Job satisfaction.

Table 1

Stepwise Multiple Linear Regression Predicting Job Satisfaction by Age, Income and

Education Level (N=95).

variables Model 1 Model 2 Model 3

B β SE B β SE B β SE
Constant 36.84*** .45 35.45*** 0.74 34.93*** 0.76

Income 0.000*** .99 0.00 0.00*** .82 0.00 0.00*** .83 0.00

Age 0.15** .17 0.06 0.15* .17 0.06

Education 0.26* .03 0.13

level

R² 0.978*** .979** .981*

∆R² .001 .001

*p<.05, **p<.01, ***p<.001

The results of multiple linear regression by stepwise method showed that the assumption of

multicollinearity was found to be assumed for all variables except for age as tolerance value

was less than .2 which indicated that there is no multicollinearity in data except for the

predictor of age. The Durbin Watson value was found to be 2.060 which was in range (1-3)

showing that the assumption of independence of residuals was met. The assumption of

normality and homoscedasticity was also found to be assumed as shown by histogram and

relatively similar spread of data across the unit in scatter plot respectively. The assumption of

sample size and linearity was also found to be assumed. The assumption of outliers was also

found to be assumed as outliers were not present. The results showed that the predictors, age,

income and education level were entered in step 1. The model 1 accounted for 97.8%

variance as ∆F (1,93) =4206.94, p<.001. results showed that income was found to be

positively significant predictor of job satisfaction. Moreover, Model 2 explained 0.1% of

variance with ∆F (1,92) =5.606, p<.05 results showed that age was found to be significantly

positive predictor of age. Lastly, Model 3 accounted for 0.1% variance as ∆F (1,91) =4.283,

p< .05 which showed that education level was found to be a positively significant predictor of

job satisfaction. The overall model accounted for 98.1% variance with F= (3,91)= 1527.65,

p<.001.
Output:

You might also like