mG2 was assessed for normality. Visual inspection of the histogram and QQ-Plot identified some issues with skewness and kurtosis. The standardised score for kurtosis (2.02) can be considered acceptable using the criteria proposed by West, Finch and Curran (1996), but the standardised score for skewness (-3.19) was outside the acceptable range. However 100% of standardised scores for mG2 fall within the bounds of +/- 3.29, using the guidance of Field, Miles and Field (2013) the data can be considered to approximate a normal distribution.
mG3 was assessed for normality. Visual inspection of the histogram and QQ-Plot identified some issues with skewness and kurtosis. The standardised score for kurtosis (1.11) can be considered acceptable using the criteria proposed by West, Finch and Curran (1996), but the standardised score for skewness (-5.63) was outside the acceptable range. However 100% of standardised scores for mG3 fall within the bounds of +/- 3.29, using the guidance of Field, Miles and Field (2013) the data can be considered to approximate a normal distribution.
The relationship between mG3(Final math grade achieved by students) and mG2(math grade achieved in Grade 2) was investigated using a Pearson correlation. A strong positive correlation was found(r=0.90,n=380,p< 0.001).
"An independent-samples t-test was conducted to compare the mean scores of mG3 for students who are trying for higher education and those who do not. A significant difference in the scores for mG3 was found ( M=10.63, SD= 4.56 for students who are trying for higher studies , M= 5.50,SD= 4.64 for students who are not trying for higher studies),(t(380)= -4.65, p = 0.01). A medium effect size was also indicated by Cohen's d value (-0.48)."
A one-way between groups analysis of Variance(Anova) was conducted to explore the impact of study-time of a student on mG3. study-time variable was divided in to 4 groups( Group: 1 - < 2 hours, Group: 2 - 2 to 5 hours, Group: 3 - 5 to 10 hours, or Group: 4 - >10 hours). There was not statistically significant difference in the G3 scores, p>0.05, which is also statistically insignificant .The effect size was calculated using eta-squared was (0.08) which is small.
A Chi-Square test for independence (with Yates’ Continuity Correction) indicated no significant association between students's family size and getting family support, χ2(1,n=380)=2.99,p=0.05, phi=0.084).
Multiple regression analysis was conducted to determine the student's final math grade(mG3).. Marks obtained in second grade(mG2),going for higher studies(higher) were used as predictor variables. In order to include the higher education in the regression model it was recorded dummy variable higher_edu (0 for no, 1 for yes).Examination of the histogram, normal P-P plot of standardised residuals and the scatterplot of the dependent variable, academic satisfaction, and standardised residuals showed that the some outliers existed. However, examination of the standardised residuals showed that none could be considered to have undue influence (95% within limits of -1.96 to plus 1.96 and none with Cook’s distance >1 as outlined in Field (2013). Examination for multicollinearity showed that the tolerance and variance influence factor measures were within acceptable levels (tolerance >0.4, VIF < 2.5 ) as outlined in Tarling (2008). The scatterplot of standardised residuals showed that the data met the assumptions of homogeneity of variance and linearity. The data also meets the assumption of non-zero variances of the predictors.
Multiple regression analysis was conducted to determine the student's final math grade(mG3).. Marks obtained in second grade(mG2),going for higher studies(higher) were used as predictor variables In order to include the higher education in the regression model it was recorded dummy variable higher_edu (0 for no, 1 for yes) and an interaction term was introduced by multiplying (inthigher * mG2) .Examination of the histogram, normal P-P plot of standardised residuals and the scatterplot of the dependent variable, academic satisfaction, and standardised residuals showed that the some outliers existed. However, examination of the standardised residuals showed that none could be considered to have undue influence (95% within limits of -1.96 to plus 1.96 and none with Cook’s distance >1 as outlined in Field (2013). Examination for multicollinearity showed that the tolerance and variance influence factor measures were outside acceptable levels (tolerance < 0.4, VIF > 2.5 ) as outlined in Tarling (2008).
Logistic regression analysis was conducted with higher studies as the outcome variable and sex and romantic(is student in relationship?) was used as predictors. The data met the assumption for independent observations. Examination for multicollinearity showed that the tolerance and variance influence factor measures were within acceptable levels (tolerance >0.4, VIF < 2.5 ) as outlined in Tarling (2008). The Hosmer Lemeshow goodness of fit statistic did not indicate any issues with the assumption of linearity between the independent variables as and the log odds of the model (x2(n=1)=0.008, p =0.92).
A principal component analysis (PCA) was conducted on the 20 items with orthogonal rotation (varimax). Bartlett’s test of sphericity, Χ2(190) = 2381.718, p< 0.001, indicated that correlations between items were sufficiently large for PCA. An initial analysis was run to obtain eigenvalues for each component in the data. Seven components had eigenvalues over Kaiser’s criterion of 1 and in combination explained 68.87% of the variance. The scree plot was slightly ambiguous and showed inflexions that would justify retaining either 2 or 4 factors. The group D had an low reliability of Cronbach's α = 0.34 and the group E also had low reliability, Cronbach’s α = .54