LEARN MORE...

Multiple Regression Analysis

Content type

User Generated

Subject

Statistics

School

Walden University

Course

RSCH 8210

Type

Homework

Uploaded By

OryyOregvr

Pages

Showing Page:

1/6

Multiple Regression Analysis

Individuals tend to have different levels of income. This study examines the factors that have a

significant impact on income or which may explain why individuals have different incomes.

Analysis carried out was in the form of multiple regression analysis based on the General social

survey dataset. The variables of interest for the study are; conrinc (Respondent income in

constant dollars), educ(highest year of school completed), emailhr(Email hours per week), hrs2(

Hours usually work per week), and Age of respondent. The research design that fits the research

question is quantitative research design as the aim is to evaluate the relationship between a

dependent variable and other independent variables (Apuke, 2017).

The research question of interest is therefore; “Does education, email hours per week, and hours

usually work per week have an impact on an individual’s income?” The null hypothesis is

therefore: Education, email hours per week, and hours usually worked per week have are not

significant predictors of an individual’s income.

The dependent variable is the respondent’s income while the independent variables are email

hours per week, hours worked per week, and highest year of school. Age of respondent. was used

as a controlling variable. All variables are ratio variable. According to Hays(2008), a ratio

variable is a variable that is numerical in nature and differences between adjacent values are

equal, and has a true zero. All variables in the study fit this definition and are therefore ratio

variables. The justification of controlling for age is that age may confound with the other

dependent variables. Pourhoseingholi et al. (2012) indicates that a confounding variable is one

that affects the variables being studied.

Analysis and Results

Showing Page:

2/6

A multiple regression analysis was carried out in SPSS to evaluate whether Education, email

hours per week, and hours usually worked per week are significant predictors of an individual’s

income. The results of the analysis were not statistically significant, F(4,2) =8.783, p

value=0.105. However, by considering the impact of individual variables on respondents’

income, the only statistically significant variable was hours usually worked per week, t=4.856, p

=0.04. The slope value for hours per week is 1180.717. The value indicates that a respondent’s

income increases by $1180.717 per year for every 1-hour increase in number of hours worked

per week.

Conclusion

Of the three independent variables studied while controlling for age, only number of hours

worked per week was found to have a significant impact on income. Numbers of hours worked

per week has a positive impact on income with income increasing by about $1181 per year for

vey additional hour worked per week. Email hours and education level have no impact on

income while controlling for age.

Showing Page:

3/6

End of Preview - Want to read all 6 pages?

Access Now

Unformatted Attachment Preview

Multiple Regression Analysis Individuals tend to have different levels of income. This study examines the factors that have a significant impact on income or which may explain why individuals have different incomes. Analysis carried out was in the form of multiple regression analysis based on the General social survey dataset. The variables of interest for the study are; conrinc (Respondent income in constant dollars), educ(highest year of school completed), emailhr(Email hours per week), hrs2( Hours usually work per week), and Age of respondent. The research design that fits the research question is quantitative research design as the aim is to evaluate the relationship between a dependent variable and other independent variables (Apuke, 2017). The research question of interest is therefore; “Does education, email hours per week, and hours usually work per week have an impact on an individual’s income?” The null hypothesis is therefore: Education, email hours per week, and hours usually worked per week have are not significant predictors of an individual’s income. The dependent variable is the respondent’s income while the independent variables are email hours per week, hours worked per week, and highest year of school. Age of respondent. was used as a controlling variable. All variables are ratio variable. According to Hays(2008), a ratio variable is a variable that is numerical in nature and differences between adjacent values are equal, and has a true zero. All ...
Purchase document to see full attachment

Tags: research design multiple regression Dependent Variables level of income survey dataset