Page 14 .618 382 .377 .44588 a. Predictors: (Constant), DGender, Violent, If a public defender was used, Number of prior convictions, Conviction Severity b. Dependent Variable: Log Incarceration ANOVA Sum of Model Squares df Mean Square Sig Regression 700.095 5 40.019 66.976 .ooob Residual 1131.000 541 2.091 Total 1831.094 546 a. Dependent Variable: Log Incarceration b. Predictors: (Constant), DGender, Violent, If a public defender was used, Number of prior convictions, Conviction Severity Coefficients Standardized Unstandardized Coefficients Coefficients Collinearity Statistics Model B Std. Error Beta Sig. Tolerance VIF Constant 2.656 353 -7.522 000 Number of prior convictions 124 015 286 8.356 .000 977 1.023 Conviction Severity 569 047 431 12.020 .000 886 1.128 If a public defender was used -.110 222 -.017 -.495 .621 984 1.017 Violent 919 134 243 6.867 000 .909 1.100 DGender 1.052 .239 150 4.393 000 984 1.016 . Dependent Variable: Log Incarceration O W ALIENWAREPage 14 36. A researcher is interested in whether there is a relationship between prior record and the length of incarceration in months. He obtains case files from an urban county on all criminal cases processed in a year. He decides to also control for other variables that may influence the sentence someone receives - such as attorney type (1=public defender), offense severity (as a scale), whether it was a violent offense (=1), and gender of defendant (male=1). Further, he realizes that his dependent variable (length of incarceration) is positively skewed and transforms it into the natural log for analysis. (10 points) a. State the null and research hypotheses. b. Assess the model fit by looking at the F-statistic and r2 - interpret each. c. Use the output to examine whether there is a significant relationship between prior record and sentence length. Interpret your results, including interpreting the slope. d. Examine the additional variables - are any of these statistically significant? If so, interpret these slopes as well. e. Of the 5 variables, which has the greatest influence on the dependent variable? Model Summary Adjusted R Std. Error of Model R R Square Square the Estimate 618 382 377 1.44588 a. Predictors: (Constant), DGender, Violent, If a public defende was used, Number of prior convictions, Conviction Severity b. Dependent Variable: Log Incarceration ANOVA Sum of Model Squares df Mean Square sig. Regression 700.095 40.019 66.976 000 Residual 1131.000 541 2.091 O m 40 of W ALIENWARE