Testing for Multiple Regression

Testing for Multiple Regression Student Name Institution Course Professor Date PART 1: Multiple Regression Analysis of Age and Economic Perception The dataset utilized in this analysis adopts the Afrobarometer Round 9 South Africa (2022) data to investigate whether the age of the respondents is predictive of the view of the current state of the economy in the country. Q4A (Country present economic condition) was the dependent variable, and Q1 (Age) was the independent variable. Despite having a single predictor, this model is based on the multiple regression design and can be used as a skeleton model to develop further analyses. The average age of the respondents was 39.51 years, which confirmed the analysis with this data. Figure 1:Model Summary and ANOVA for Multiple Regression Predicting Economic Perception from Age An enter method was used to analyze by way of multiple regression. This model summary showed that the relationship between age and economic perception was very weak, with R = .037 and R² =.001, implying that age was only able to explain 0.1% of the variance in perceptions of the economic situation of the country. The regression model in general was not statistically significant, F(1, 1578) = 2.15, p =.143. The results are an indication that age in itself is not a significant factor behind differences in the way respondents rate the national economy. Figure 2: Regression Coefficients for Age Predicting Economic Perception This conclusion is also supported by the regression coefficients. The unstandardized coefficient of age (B = -0.003) was not large, meaning that perceptions of the economy were slightly less negative with an increase in age (a year). The effect is, however, not statistically significant (p =.143), and the standardized beta coefficient (β = -.037) proves that the effect size was not significant as described by Pallant (2020). The age coefficient confidence interval of 95% ranged between -0.008 and 0.001, and the confidence interval meets the zero value, which is also evidence that there was no reliable effect. It evaluated assumptions of regression before interpretation. The dependent and independent variables were considered metric-level variables, and this met the measurement level in regression. The residual plots revealed no significant breach of linearity or homoscedasticity, and no outliers were extreme. Due to the large sample size, a slight deviation from normality was not an issue of concern for inference. In a social change, the results suggest that age does not significantly drive the perception of the economic status of South Africans regarding the country. Depending on this discovery, policymakers and social planners cannot use age as the only measure of differences when creating economic or social interventions. Rather, more focus should be given to more macrostructural and socioeconomic determinants that...