Section 4: RESULTS AND FINDINGS
Analysis of pre- and post- questionnaires Students were asked to complete pre and post surveys to record their views on how confident, motivated and prepared they felt for the exam they were being tutored within.
It is worth noting, that though the school was asked to distribute this to all students invited to take part in the tutoring, a substantial number returned were from the maths group( with a relatively high amount of pre and posts able to be matched) as opposed to the English group.
Out of the potential 112 students, 72 completed either pre / post or both surveys with 10 from the English tutor programme, 3 didn’ t disclose and 55 from the maths tutor programme.
Of the 72 completed, 36 were matched for both pre and post surveys and have been used for the statistical analysis.
Testing for whether the 36 were representative, the averages suggested the 36 completers were not drastically different from the full 72 but when comparing variances, it is slightly lower for the 36 completers in most cases suggesting that they may be a more homogeneous group than the full 72. They are likely representative, but we need to proceed with caution. However, on conducting a t-test on the pre surveys filled in by all the students vs the 36 who completed both pre and post- survey there was no statistically significant difference so they can be considered representative.
For each question, we performed an effect size( Cohen’ s d) and a t-test( see Appendix 2) and then a correlation analysis on Survey Responses vs Final Grade and Sessions attended vs Survey Response Changes.
Survey Response Changes vs. Final Grade: Some changes in survey responses show weak correlations with the final grade, but no strong predictors stand out.
• " It is important to me to learn what is being taught "( 0.19) shows a slight positive correlation with final grades.
• " I think I know a great deal about this subject "( 0.24) has a stronger correlation with final grades.
• " I worry a great deal about tests "( 0.27) suggests that students who expressed more concern saw slightly better outcomes.
Sessions Attended vs. Survey Response Changes: Some areas, such as motivation and self-tracking of progress, show moderate positive correlations with the number of tutoring sessions.
• Students who attended more sessions showed small increases in learning importance and understanding confidence.