Testing for multiple regression
Name:
Institution:
Testing for multiple regression
Introduction
From the Afro-barometer dataset, this study aimed at testing the effect of countries present economy and the complications in the politics and government on the fear of crime at homes using multiple regression technique.
Research question
Does of countries present economy and the complications in the politics and government impact on the fear of crime at homes?
Hypotheses
Null hypothesis: Fear of crime at homes is not influenced by countries present economy and the complications in the politics and government.
Alternative hypothesis: Fear of crime at homes is influenced by countries present economy and the complications in the politics and government.
Analysis results and discussions
The regression model summary yielded an R-squared value 0.010 implying that the model fit was poor. The r-square value implied that 1.0% of the variations in the fear of crime at homes were explained by the complications in politics and government, and the countries present economic condition.
Summary of the Model analysis | ||||
Model | R value | R Square | Adjusted R Square | Standard Error |
1 | .100^{a} | .010 | .010 | 1.175 |
a. Predictors: (Constant), Q16. Politics and government too complicated, Q3a. Country’s present economic condition |
The results of the ANOVA yielded an F-statistic value equal to 244.043 with a significance value equal to 0.00. This implied that the model was statistically effective.
ANOVA results | ||||||
Model | SS | Degrees of freedom | MS | F-stat | Signf. | |
Regression | 673.645 | 2 | 336.823 | 244.043 | .000^{b} | |
Residual | 66394.878 | 48106 | 1.380 | |||
Total | 67068.523 | 48108 | ||||
a. Dependent Variable: Q9b. How often feared crime in home | ||||||
b. Predictors: (Constant), Q16. Politics and government too complicated, Q3a. Country’s present economic condition |
The summary of the results of the coefficient analysis yielded values equal to 0.966, -0.095 and -0.005 representing coefficients of the model constant, country’s present economic condition and complications in politics and government.
Coefficients | ||||||
Model | Unstandardized Coeff | Standardized Coeff | t-stat | Signf. | ||
B(coeff) | Std. Error | Beta | ||||
1 | (Constant) | .966 | .015 | 64.571 | .000 | |
Q3a. Country’s present economic condition | -.095 | .004 | -.100 | -21.952 | .000 | |
Q16. Politics and government too complicated | -.005 | .004 | -.006 | -1.287 | .198 | |
a. Dependent Variable: Q9b. How often feared crime in home |
The t-statistics associated with the coefficients of the regression terms were equal to -21.952 and -1.287 with significant values of 0.000 and 0.198 for the country’s present economic condition and complications in politics and government. Clearly the probability value associated with country’s present economic condition was less than 0.05 (5% alpha), hence, the study rejected the null hypothesis and concluded that Country’s present economic condition significantly influenced the fear of crime at homes. The absolute probability value associated with the complications in politics and government was equal to 1.98, greater than 0.05 (5% alpha), hence, the study failed to reject the null hypothesis and concluded that complications in the politics and government never influenced the fear of crime at homes (Berry & Feldman, 2014).
Recommendation
This study recommends that the society should embrace activities which aim at improving the countries’ economic condition since the study showed that good economic conditions reduced fear of crime at homes.
References
Berry, W. D., & Feldman, S. (2014). Multiple regression in practice. Beverly Hills: Sage Publications.
Multiple regression
Name
Institution
Multiple Regression
Research question
How well does family income in constant dollars predict the respondent’s prestige score?
Null Hypothesis
H_{0}: Family income inconstant dollars is not a significant predictor of the respondent’sprestige score.
Research design
The research design used for this question is causal design. Causaldesign can be used to measure the impact a certain change has on an existing assumption (Labaree, 2016).
Dependent variable and how it is measured
The dependent variable for the study was the respondent’s prestige score. The variable is measured using a ratio scale
Independent variable and how it is measured
The independent variable is family income in constant dollar. The variable is measured using a ratio scale
Control variables added in the model
The control variables added in the study included:
- Age of respondent
- Number of hours usually work a week
- Respondent’s highest degree
Justification for adding the variables
The researcher was not particularly interested in the control variables but acknowledges some relationship to the dependent variable. Zhou’s (2005) study on occupational prestige uses college education and work hours as control variables, and this justifies their uses as control variables. The researcher was also interested in establishing whether the dependent variable can be controlled for age.
Significance and stregth of effect
From table 2 in the appendix section, since the p-value is less than 0.05 this means that the model has reached statistical significance. However, since the variable family income in constant dollars has a p-value of 0.794 as shown in Table 3. The independent variable is not significant in predicting the respondent’s prestige score. Therefore, the researcher did not find any significance.
Explanation and research question
Table 1 from the appendix shows that the Adjusted R Square is 0.364. This means that 36.4% of the variation in the dependent variable can be explained by the variation in the independent variables. Table 3 also shows that only the variable respondent’s highest degree had a significant contribution to the model. Given this result, we fail to reject the null hypothesis and conclude that family income does not predict prestige score
References
Labaree, R. (2016, October 21). Organizing Your Social Sciences Research Paper: Types of Research Designs. Retrieved October 24, 2016, from USC Libraries: http://libguides.usc.edu/writingguide/researchdesigns
Zhou, X. (2005). The Institutional Logic of Occupational Prestige Ranking: Reconceptualization and Reanalyses1. American Journal of Sociology, 90-140.
Appendix
Table 1
Model Summary | ||||
Model | R | R Square | Adjusted R Square | Std. Error of the Estimate |
1 | .659^{a} | .434 | .364 | 10.343 |
a. Predictors: (Constant), RS HIGHEST DEGREE, NUMBER OF HOURS USUALLY WORK A WEEK, AGE OF RESPONDENT, FAMILY INCOME IN CONSTANT DOLLARS |
Table 2
ANOVA^{a} | ||||||
Model | Sum of Squares | df | Mean Square | F | Sig. | |
1 | Regression | 2628.403 | 4 | 657.101 | 6.142 | .001^{b} |
Residual | 3423.489 | 32 | 106.984 | |||
Total | 6051.892 | 36 | ||||
a. Dependent Variable: Rs occupational prestige score (2010) | ||||||
b. Predictors: (Constant), RS HIGHEST DEGREE, NUMBER OF HOURS USUALLY WORK A WEEK, AGE OF RESPONDENT, FAMILY INCOME IN CONSTANT DOLLARS |
Table 3
Coefficients^{a} | ||||||
Model | Unstandardized Coefficients | Standardized Coefficients | t | Sig. | ||
B | Std. Error | Beta | ||||
1 | (Constant) | 24.578 | 10.190 | 2.412 | .022 | |
FAMILY INCOME IN CONSTANT DOLLARS | -1.239E-005 | .000 | -.048 | -.263 | .794 | |
AGE OF RESPONDENT | .074 | .146 | .079 | .508 | .615 | |
NUMBER OF HOURS USUALLY WORK A WEEK | .211 | .157 | .202 | 1.340 | .190 | |
RS HIGHEST DEGREE | 5.980 | 1.446 | .628 | 4.134 | .000 | |
a. Dependent Variable: Rs occupational prestige score (2010) |