Question 2
Download as DOCX, PDF
1 like288 views
This multiple regression analysis examined how motivation, grades in high school, parent's education, and gender predict mathematics achievement. The analysis found that the combination of these variables significantly predicted math achievement, with grades in high school having the strongest influence followed by motivation and gender. Specifically, the model explained 45.1% of the variance in math achievement, and grades in high school, motivation, and gender were individually significant predictors.
Question 2
- 1. Based on the SPSS 1. Based on descriptive statistics, Descriptive Statistics Mean Std. Deviation N math achievement test 12.5645 6.67031 75 motivation scale 2.8744 .63815 73 gender .55 .501 75 grades in h.s. 5.68 1.570 75 parents' education 4.3933 2.31665 75 a. Checking on assumptions 1. Correlations Statistics above 0.3 Statistics Value Sig Value Alpha-value Correlation Maths achievement x motivation scale .316 0.003 0.05 Correlated Maths achievement x parents .504 0.000 0.05 Correlated Maths achievement x gender -.301 0.004 0.05 Correlated Maths achievement x grades .389 0.000 0.05 Correlated So, all the variables are correlated. 2. Check on multicollinearity Look on Coefficient Tolerance value must be more than .1. Model Collinearity Statistics Tolerance VIF 1 (Constant) motivation scale .945 1.058 gender .867 1.154 grades in h.s. .895 1.117 parents' education .857 1.167 So, there is no multicollinearity. 3. Check on outliers, normality, linearity, homoscedaticity, independence of residuals So no major diversion from normality.
- 2. b. . Look at Model Summary Model Summary b Model R R Square Adjusted R Square Std. Error of the Estimate 1 .672 a .451 .419 5.08436 a. Predictors: (Constant), parents' education, motivation scale, grades in h.s., gender b. Dependent Variable: math achievement test R = .672, R squared = 0.451 = 45.1% variance in mathematics achievement 5. Hypothesis Ho There is no statistical significance in the multiple regression. Ha There is statistical significance in the multiple regression. Test statistics Sig Value Alpha-value Decision F = 13.981 0.000 0.05 Able to reject Ho, Accept Ha So, There is statistical significance in F (7,63) =13.981, p<0.05. c. Look at independent variables Beta must be the biggest Model Standardized Coefficients t Sig. Beta 1 (Constant) -1.485 .142 motivation scale .206 2.228 .029 gender -.260 -2.696 .009 grades in h.s. .467 4.917 .000 parents' education .186 1.921 .059 So, grades in high school has the strongest unique contribution towards mathematics achievement, followed by motivation and then gender which is also significant because their p<0.05 Report: A standard multiple regression has been used to analyze the combination of motivation, grades in high school, parents education and gender predict mathematics achievement. Based on the descriptive statistics, the highest mean is the grades in high school which is 5.68. The data screen has showed that the combination are correlated with each other. Further normality test shows that there is no major diversion as well as there is no multicollinearity within the independent variables. Regression results indicate an overall model of two predictors (gender and grades in high school) significantly predicted mathematics achievement R squared = 0.451, F (7, 63) =13.981, p<0.05. Therefore, the model which includes grades, motivation and gender explains 46.2 % of the variance in the mathematics achievement. Of these three variables, grades in high school makes the largest contribution (beta = 0.467) while motivation is 0.206 and gender has a beta = -.260. The beta values also indicate the increase of standard deviation 1.57 in grades, maths achievement statistics will also increase by 6.6.