Aspects of Job Satisfaction Depending on Benefits - Research Paper Example

Unit –5 Regression Analyses

Introduction

This assignment conducts three linear regression tests for three pairs of independent and dependent variables.  The data used to conduct the tests were obtained from a survey conducted by AIU.  The regression tests were conducted using Excel’s built-in function.  The following paragraphs present the results and analyses of the tests.

Results of Tests

Table 1

Regression Output of Variables Benefits vs. Intrinsic Job

Satisfaction

Regression Statistics

Multiple R

0.030092219

R – Square

0.000905542

Adjusted R square

-0.004408791

Standard Error

0.876576061

Observations

190

Coefficient

Y-intercept

4.524522995

Slope

0.151207676

Note:  Benefits = X; Intrinsic job satisfaction = Y

Figure 1.  Regression line Benefits vs. Intrinsic job satisfaction

Table 2

Regression Output of Variables Benefits vs. Extrinsic Job

Satisfaction

Regression Statistics

Multiple R

0.026855348

R – Square

0.00072121

Adjusted R square

-0.004594103

Standard Error

1.024951959

Observations

190

Coefficient

Y-intercept

5.750215066

Slope

-0.157769935

Note:  Benefits = X; Extrinsic job satisfaction = Y

Figure 2.  Regression line Benefits vs. Extrinsic job satisfaction

Table 3

Regression Output of Variables Benefits vs. Overall Job

Satisfaction

Regression Statistics

Multiple R

0.010697607

R – Square

0.000114439

Adjusted R square

-0.005204101

Standard Error

1.072353233

Observations

190

Coefficient

Y-intercept

4.959981251

Slope

-0.065732877

Note:  Benefits = X; Overall job satisfaction = Y

Analysis of Results and Conclusion

The assignment conducted three separate linear regression analyses to establish a relationship between independent and dependent variables obtained through a survey.  The relationship between the two variables, in this case, is expressed through the linear regression equation, y = a + bx.  In this equation a is called the intercept of the Y-axis and b is called the slope of the regression line (“The University of New England”, n.d.). The slope indicates how changes in the values of the independent variable affect changes in the dependent variable.  The slope b may receive a positive or a negative value.  A positive slope defines that the dependent variable increases as the independent increases while the negative implies dependent variable decreases while the independent variable increases.  Table 4 displays one positive and two negative slopes.  Thus, Y = 4.52 + 0.15 X defines that both Benefits and Intrinsic job satisfaction move in the same direction, which suggests that the increase in benefits increases intrinsic job satisfaction.  However, Y = 5.75 – 0.16 X defines that the variables Benefits and Extrinsic job satisfaction move in different directions.  It means an increase in Benefits decreases extrinsic job satisfaction.  Regression equations Y = 5.75 – 0.16 X, and Y = 4.96 – 0.07 X demonstrate negative relationships between independent and dependent variables while Y = 4.52 + 0.15 X displays a positive relationship between independent and dependent variables.

The Excel regression statistics evaluate linear correlation coefficient r, which measures the direction and strength of the relationship between two variables.  Regression and correlation are two different concepts in statistical analysis.  Regression explains cause and effect relations between variables while correlation defines interdependencies of variables (“Mathbits”, n.d.).  In this case, through the correlation coefficient, it is possible to observe interdependency between benefits and intrinsic job satisfaction, benefits and extrinsic job satisfaction, and benefits, and overall job satisfaction.  The correlation coefficient r gets values between minus one and one (-1 < r < + 1).  The value of r may illustrate positive, negative, and no correlation between two variables.  When the value of r is close to zero, it indicates no or weak correlation between variables.  The values, in this case, are r=0.03 between benefits and intrinsic job satisfaction; r = 0.027  between benefits and extrinsic job satisfaction; r = 0.01 between benefits and overall job satisfaction.  So, it is proper to decide that the data presented in the survey values do not describe the dependencies of the tested variables.  Nevertheless, it also may be concluded that under the given circumstances among three variables r = 0.03 shows the strongest correlation of variables.  So, it can be concluded that benefits and intrinsic job satisfaction have the strongest correlation coefficient.  This correlation coefficient for a manager indicates that people like to have benefits and their intrinsic job satisfaction depends on benefits.

The Excel output of regression analysis also gives valuable information, which is noted through r 2; it gets the value 0 < r 2 < 1 and determines the coefficient of fluctuation (variances) between predictable and unpredictable variables.  In other words, it denotes the strength of the linear association of X and Y. Based on the values of r2 displayed in Table 4, it can be concluded that 0.09% of the total variation of Y can be explained by the linear relationship between the benefits and intrinsic job satisfaction variables.  In the other two cases, the r2 values are 0.00%, which shows that the total variation of Y cannot be explained by the relationships between independent and dependent variables.  This assignment recommends conducting a Chi-square test for nominal variables with the AIU survey data.