Discussion: Correlation and Bivariate Regression

Student 1

COLLAPSE

Top of Form

In this analysis, the author of A Comparative Analysis of Illinois, Ohio, Colorado and South Dakota Park Districts and Parks and Recreation Departments to Wisconsin, Iowa, Missouri, Kansas, Indiana, and Michigan Parks and Recreation Departments used quantitative method design with comparative means-testing. The study evaluates park district services efficiencies and effectiveness in comparison to services provided by municipal governments.  Emanuelson (2008) used bivariate and multivariate linear regression to test the relationships between service levels and efficiency levels.

This design and test are appropriate.  The difference between correlation or bivariate regression measurements is that “correlation measures the degree of a relationship between two variables, whereas regression is how one variable affects another” (Calvello, 2020). The author used bivariate regression analysis because it analyzes two variables to establish the strength between the service levels and efficiency levels and how one affected the other (Emanuelson, 2008).  Bivariate regression analysis is “useful for examining the ability of the independent variable or predictor variable to predict the dependent of criterion variable” (Rockinson-Szapkiw, 2013).

The author displayed all of the data through multiple tables and figures and included his questionnaire in the Appendix.  Tables 12 and 13 are two examples of data provided.    

Table (12) shows the Linear Regression with the Structure of Government as the Independent Variable with predictor, or constant as the department of the city, the independent variable is not statistically significant.  The article states a weak relationship between parks and recreation and budgetary efficiency, as seen in R-Square .008.  Coefficients – dependent variable: Budgetary efficiency

 

Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
1	.087 (a)	.008	.003	1.0565985

 

Mode 1		Unstandardized Coefficients	standardized Coefficients	t	Sig.
		B	Std. Error	Beta
1 (constant) department of city	.812 .184	.101 .144	.087	8.025 1.275	.000 .204

 In another table (13) the Linear Regression with Structure of Government as the Independent Variable with the predictor or constant as the park district.  The data is “neither meaningful nor statistically significant as a predictor of staff efficiency” (Emanuelson, 2008).  R value is .039.  Coefficients – dependent variable: Budgetary efficiency

Model	R	R Square	Adjusted R Square	Std. Error of the Estimate
1	.039 (a)	.002	-.003	42.89480

 

Mode 1		Unstandardized Coefficients	standardized Coefficients	t	Sig.
		B	Std. Error	Beta
1 (constant) Park district	25.196 -3.464	3.587 6.038	-.039	7.024 -.574	.000 .567

This study provides meaningful data and affects current literature, and provides useful data for social change.  "The data regarding differences in means of levels of service and efficiency support the theories contained in the current literature of public choice and metropolitan ecology theory" (Emanuelson, 2008).   

While not every variable that was run is considered meaningful, the data's conclusion indicates information obtained and can be applied.  Emanuelson (2008) concluded that park districts provide greater effectiveness in services than parks and recreation departments. In contrast, per capita, spending on parks and recreation services is a good predictor of services, and training is essential (Emanuelson, 2008). 

 

References:

Calvello, M. (2020, January 2). Correlation vs. Regression Made Easy: Which to Use + Why. Retrieved January 21, 2021, from https://learn.g2.com/correlation-vs-regression

Emanuelson, D. N. (2008). A Comparative Analysis of Illinois, Ohio, Colorado and South Dakota Park Districts and Parks and Recreation Departments to Wisconsin, Iowa, Missouri, Kansas, Indiana, and Michigan Parks and Recreation Departments. Conference Papers -- Midwestern Political Science Association, 1–42.

Rockinson-Szapkiw, A. J. (2013, October 4). SPSS Tutorial: Bivariate Regression. Retrieved from https://www.youtube.com/watch?v=qfilvBlg0iI