Factors Influencing Student Performance

1. Does class size (small and medium classes only) influence STEM-related performance (math and/or science)? Provide the full analysis. [STEM=science, technology, engineering and math]

a. Methods

The General Linear Model (GLM) was used to conduct an analysis of covariance (ANCOVA). The STEM-related subjects (science and math) were set as the dependent variables (DV) while the small and medium classes were the independent variables (IV). Given that the predictors were categorical, coding was used to create the dummy variables with small class being variable 1 and medium class variable 2 (Field, 2013). If you require further assistance with your analysis, seeking statistics dissertation help can provide valuable insights into the methodology and interpretation of results.

b. Results

Parameter Estimates

Bootstrap for Parameter Estimates

Estimates

An increase in the covariate (small class and medium class) by one unit leads to an increase in the dependent variable (math) by 1.558 and 0.150 units respectively. Also, a one unit increase in the small and medium classes leads to an increase in the science performance by 3.758 and a decrease in the same by 0.842 units respectively.

c. Descriptive Statistics

Descriptive Statistics

d. Conclusions and Recommendations

The analysis shows that the class size influences STEM-related performance. The indications of the results are that, as the class size increases, performance in both math and science decreases (Jarman, 2013). On the other hand, while class size is constrained to be smaller than before, the performance in both math and science increases. The implications are that the STEM-related performance is driven by small teacher to student ratio. That is, math and science teachers need to serve a smaller number of students at every time for performance to improve.

2. Does the teacher ask for support differentially for boys versus girls (i.e., is there an association between gender and teacher request for support)?

a. Methods

A Chi-square test (2-tailed) was used to investigate the association between gender and request for teacher support. Both graphs and tables were used to present the results of the analysis. The ‘Descriptive Statistics’ function of the SPSS was used to conduct crosstab calculations as regards to the association between gender and request for support (Field, 2013).

b. Results

Gender * ReqSup Crosstabulation

Chi-Square Tests

a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 57.26.

b. Computed only for a 2x2 table

Symmetric Measures

a. Not assuming the null hypothesis.

b. Using the asymptotic standard error assuming the null hypothesis.

The Pearson Chi-Square Tests results (X2(1) = 22.708; p = .000; p < .001) show that there is an association between gender and teacher’s request for support. Therefore, the teacher asks for support differentially for boys vs. girls (Cramer, 2003).

The Phi (-.311) and Crammer’s V (.311) measures, despite the presence of the association, show that it (association) is weak.

c. Descriptive Statistics

Descriptive Statistics

a. Unless otherwise noted, bootstrap results are based on 1000 bootstrap samples

d. Conclusions and Recommendations

The Chi-Square analysis has fully addressed the query in that it fully provides evidence that teacher request for support is differential among boys and girls. The implications are that the gender is likely to influence the decision for teacher’s request for support. Despite the above, further analysis (on the tests for the association’s strength) imply that the difference under consideration is not significant (Field, 2013).

3. Is there an association between self-esteem and overall achievement?

a. Methods

Given that the data used was continuous, a Pearson correlation analysis was conducted. Using the correlate function of the SPSS a Bivariate correlation analysis was conducted. Both tables and graphs were used to present the summary of the results for easy understanding and description (Field, 2013). Additionally, descriptive statics were considered to communicate the characteristics of the data used.

b. Results

Correlations

**. Correlation is significant at the 0.01 level (2-tailed).

b. Unless otherwise noted, bootstrap results are based on 1000 bootstrap samples

There was a positive correlation between self-esteem and overall achievement, r = .491, n = 235, p = .000; p < .001. The results are summarized in the scatter plot below;

Overall, there was a weak positive correlation between overall achievement and self-esteem. Increases in self-esteem were correlated with increases in overall achievement.

c. Descriptive Statistics

Descriptive Statistics

d. Conclusions and Recommendations

The analysis fully addresses the query giving indications that self-esteem moderately correlates with overall achievement. Therefore, it is statistically correct to conclude that self-esteem plays a part in influencing the overall achievement of the student. Therefore, when self-esteem levels are high, this is likely to result in high levels of performance. Consequently, it is recommended that students’ levels of self-esteem be increased (through a variety of motivations) to achieve greater levels of overall achievement than before (Creswell, 2012).

4. Does parental support, wellbeing and gender contribute to overall achievement?

a. Methods

A multiple regression model (MRM) was used to conduct a regression analysis where the overall achievement was the dependent variables. The predictor variables included parental support, wellbeing and gender. The data consisted of both continuous and categorical predictor variables. The linear MRM was necessary in estimating the contribution of the identified factors to the overall performance of a student (Hilbe, 2014). The enter method was used in conducting the regression analysis in SPSS.

b. Results

Model Summary

ANOVAa

Coefficientsa

Bootstrap for Coefficients

It was found out that parental support, wellbeing and gender of the student explain a significant amount of the variance in overall achievement {F (3,231) = 15.78, p < .05, R2 = .17, R2Adjasted = .159}.

In particular, the regression analysis shows that parental support as well as gender did not significantly predict the level of overall achievement; {Parental Support: [Beta = .000, t (234) = .000, ns (p > .05)] and Gender: [Beta = -.027, t (234) = -.443, ns (p > .05)]}. On the contrary, student’s wellbeing significantly predicted the level of overall achievement {Beta = .413, t (234) = 6.876, p < .05}.

c. Descriptive Statistics

Descriptive Statistics

d. Conclusions and Recommendations

Parental support and gender do not significantly contribute to the overall achievement of the student. On the contrary, student’s wellbeing significantly contributes to their overall achievement. Despite the conclusions, the model used does not, to a significant extent, address the query because of a weak Adjusted R-square (15.9%) (Field, 2013). The implication of the low value is that only 15.9% variations in the dependent variable are explained by the independent variables (Cramer, 2003). Consequently, it is recommended that parental support for students focuses on increasing the levels of the students’ wellbeing to achieve overall excellent results (Creswell, 2012).

e. Standardized Equation (model) for the analysis

Overall Achievement = -28.827 + .000*Parental Support + .413*Wellbeing - .027*Gender + 2.906

f. Anticipated overall achievement

Overall Achievement (anticipated) = -28.827 + .000*9 + .413*85 - .027*0 + 2.906

Overall Achievement = -28.827 + 35.105 + 2.906

Overall Achievement = 9.184

References

Cramer, D. (2003). Advanced Quantitative Data Analysis. England: Open University Press.

Creswell, J. W. (2012). Educational Research: Planning, Conducting, and Evaluating Quantitative and Qualitative Research (4th ed.). Boston: Pearson Education.

Field, A. (2013). Discovering Statistics using IBM SPSS Statistics (4th ed.). Sage Publications Ltd.

Hilbe, J. M. (2014). Modeling Count Data. Cambridge University Press.

Jarman, K. H. (2013). The Art of Data Analysis: How to Answer Almost Any Question Using Basic Statistics. New Jersey: John Wiley & Sons.

Continue your journey with our comprehensive guide to Exploring the Relationship between Self-Esteem and Academic Stress.