TESTING FOR CORRELATION AND BIVARIATE REGRESSION

Testing for Correlation and Bivariate Regression Student’s Name Affiliation Course Number: Course Name Instructor Name Due Date Testing for Correlation and Bivariate Regression In this discussion, I utilized the Afrobarometer dataset to test for correlation and bivariate regression. My selected variables were: Q1. Age, which is the independent variable (IV), and the Lived Poverty Index (average index of 5 poverty items), which is the dependent variable (DV). I created two subsequent research questions. Research question 1 (RQ1) was for correlation, while the second one, RQ2, was designed for the regression analysis. The means for Q1. Age, was 37.23. See the table below. Table 1 One-Sample Statistics N Mean Std. Deviation Std. Error Mean Q1. Age 12821 37.23 14.613 .129 Correlation Analysis Research Question 1 (RQ1): What is the relationship between Q1. Age and the Lived Poverty Index (average index of 5 poverty items)? My null hypothesis (H0) was “There is no relationship between Q1. Age and the Lived Poverty Index.” Table 2 Correlations Q1. Age Lived Poverty Index (average index of 5 poverty items) Q1. Age Pearson Correlation 1 .056** Sig. (2-tailed) .000 N 12821 12637 Lived Poverty Index (average index of 5 poverty items) Pearson Correlation .056** 1 Sig. (2-tailed) .000 N 12637 12747 **. Correlation is significant at the 0.01 level (2-tailed). To test whether there was a significant relationship between IV (Q1, Age and DV (Lived Poverty Index), I conducted a Pearson correlation analysis. The results are displayed in Table 2 above. From the SPSS data set, both variables are measured on an interval scale, which makes them suitable for Pearson correlation. My sample was 12821, which also made it a better data set for robust statistical tests. From the results (Table 2), the Pearson correlation coefficient r is 0.056 with a significance level of p < .01. This shows that between the two variables (i.e., Q1. Age and the Lived Poverty Index), the statistical significance stands at the 1% level. Despite the statistical significance, the effect size can be deemed very small. The r2 is .003, which suggests that less than 1% of the variance in the Lived Poverty Index is explained by age. The correlation coefficient 0.056 also indicates a very slight to no positive linear association between poverty and age, which suggests that as age rises, the Lived Poverty Index is also inclined to increase, but with a minimal amount. The effect size reveals that only about 0.3% of the variation in poverty levels c...