Student name

RSCH FPX 7864 Quantitative Design and Analysis

Capella University

Professor Name

Submission Date

t-Test Application and Interpretation

Having a systematic data analysis framework also provides a platform for the structure of how to conduct an organised research in the formulation of clear goals, identification of variables, and limitations to be noted in the course of data collection. The t-test is used to statistically compare the mean consumption of two samples to identify significant differences (Lee, 2022). The study examines the differences in performance between those students who attended and those who did not attend preparatory classes. Comparing the average score in the groups helps researchers determine the effect of the Review session on academic outcomes. In the examination, two variables are taken into account: Review as a categorical variable (1 = no, 2 = yes) that will indicate participation in the session, and Final as a continuous variable that will measure the number of correct responses to the assessment. The systematized study generates research-based findings about the effect of preparatory sessions on student achievement.

Data Analysis Plan

• Research Question

Is there a significant difference in examination performance between students who participate in review sessions and those who do not attend?

• Null Hypothesis

There is no significant difference in examination performance between students who participate in review sessions and those who do not attend.

• Alternative Hypothesis

There is a significant difference in examination performance between students who participated in review sessions and those who did not attend.

Testing Assumptions

• Levene’s Test Assumption Check

The results of the test (F = 0.740, p = 0.392) show that there is no significant difference in the variances of students who have attended and who have not attended review sessions (n=55; n=50, respectively). The standard independent samples t-test techniques are appropriate, as the p-value (0.392) is greater than 0.05. The test with F = 0.740, df1 = 1, df2 = 103 and p = 0.392 does not reject the null hypothesis of equal variances.

• Homogeneity of Variance Assumption

Given that the p-value (0.392) significantly exceeds the conventional alpha value of 0.05, researchers can assume the homogeneity of variances between groups of students using the Levene test. The null hypothesis is not rejected, so the assumption of equal variances is supported and we are justified in continuing to use the standard independent samples t-test and not using something like Welch’s test (Vankelecom et al., 2024). The findings, out of the 105 total participants (df2+2) show that the assumption of the equality of variance is met and, as such, the variability of the scores between groups is statistically equal, regardless of the differences in the means of the performance.

Results and Interpretations

• Descriptives

• Independent Samples T-Test

In order to explore possible differences in assessment outcomes concerning preparatory course attendance, an independent group comparative analysis was performed. Students in two separate groups formed the research participants. Group 1 included 55 students who attended preparatory course and Group 2 included 50 students who had neither attended preparatory course nor attended the supplementary workshop. According to the descriptive statistics, the mean score (M= 62.160, SD= 7.993) among students who did not attend the supplementary session (n= 50) was slightly higher than the mean among students who attended the supplementary session (n= 55, M= 61.545, SD= 7.356) respectively. After checking the equality of variance assumption by means of the Levene test ( F= 0.740, 103, p= 0.392), the standard Student t-test was appropriately used.

• Independent Samples t-Test Results

Errors in the independent samples t-test t(103) = -0.410 p = 0.682 showed that there was no statistically significant difference between mean final examination scores of groups. As the p-value of 0.682 significantly lies above the traditional alpha value of 0.05, we can understand that the study does not reject the null hypothesis, indicating similar academic performance among students who decided to attend the supplementary sessions and those who did not attend. The average difference in scores (0.615 points) is small in both statistical and practical terms and indicates less than 1% difference in performance. The relatively close standard deviations (7.356 vs. 7.993) are another indication of a similar distribution characteristic of the groups. Results indicate no significant impact of attendance at the extra sessions on the final assessment scores of the students, dispelling the notion that supplementary practice activities would raise the academic performance of the students.

Statistical Conclusion

The study examined the correlation between attendance at review session and final exam performance in students. Statistical analysis showed that there was no significant difference between participants, (n= 55, M= 61.545, SD= 7.356) and non-participants, (M= 62.160, SD= 7.993) respectively. The results of the test of equality of variance (F= 0.740, p= 0.392) and the values exceeded the 0.05 threshold and supported the use of a regular independent samples t-test. The statistical analysis shows that there were no significant differences between the groups: t(103) = -0.410, p= 0.682, and that the difference between the means of two groups is insignificant, 0.615 points. It therefore retained its null hypothesis of no difference in exam achievement. Results indicate that the existing review session format (M= 61.545, SD= 7.356 vs. M= 62.160, SD= 7.993; t(103)= -0.410, p= 0.682) might not be sufficient to boost academic performance, and some changes may be needed to the format of the supplementary teaching sessions or to the implementation style and format of the review sessions.

Limitations and Alternative Explanations

The statistical analysis determines several methodological constraints that can affect the understanding of results. Even though the independent samples t-test is suitable when comparing group means, the test does not control for confounding variables, including previous academic performance or additional study behaviors (Zarei et al., 2021). The sample size used (n = 105) raises concerns over statistical power to detect small, but significant differences between groups. The effects of frequency of participation or length of session may be obscured by binary classification of review session attendance. Selection bias is a substantial limitation, as it means that session participants are potentially different from non-participants in terms of their motivation or academic commitments than they are otherwise (Lesala et al., 2025). Results in student achievement and their perceived benefits may be affected by the timing of review sessions as it relates to final examinations. Further research designs must use multivariate methods to deal with confounding factors. Research into the influences of review sessions on diverse student groups and distinct academic settings needs to be continued.

Application

Medication adherence rates in independent samples can be effectively analyzed using independent sample t-tests when variables are divided by healthcare facilities using various strategies to educate patients. This statistical approach allows researchers to determine whether there is a significant difference in adherence outcomes between the two educational strategies. The independent variable would be the educational mode (comprehensive medication counseling against regular information provision) and the dependent variable would be the rate of medication adherence over a period of time. The analysis is important as medication non-adherence is a serious, highly preventable problem that remains a challenge affecting patient outcomes in healthcare facilities (Piña et al., 2020).

• Educational Interventions and Patient Safety

Directly related to patient safety, systematic educational interventions are effective in reducing medication errors based on better understanding and facilitating comprehensive training to help students and practitioners improve on adherence (Jaam et al., 2021). Results would give healthcare facilities a basis for creating effective educational protocols that would minimize treatment failures while reaffirming the effectiveness of quality assurance systems in clinical departments. Research shows that, by implementing appropriate education practices, it is possible to reduce the complications of treatment and also improve the performance of healthcare teams due to more effective communication practices (Zajac et al., 2021). The results greatly empower health care delivery systems as well as professional adherence to delivering safe and effective patient care using evidence-based practices in education.