Data Analysis Plan
RSCH FPX 7864 Assessment 3 The Final and review variables that were utilized in the t-test and Levene’s test were utilized. The final variable is the number of correct answers on the final exam, while the review variable is determined by the students who attend the review sessions, which is coded as 1 = no; 2 = yes. The review is a two-level categorical variable, but the Final is treated as continuous for this study.
The review question is, are the mean test scores of the group of students who attended the review session (yes) and the group of students who did not attend (no) significantly different on the final exam? Null hypothesis is that the mean test scores of students who attended the review session and the students who didn’t attend the session are not different significantly from each other. Lastly, the alternate hypothesis is that students who attended the review session and students who did not differ significantly in mean test scores on the final exam.
Testing Assumptions
Assumption Checks
Test of Equality of Variances (Levene’s)
| Test | f | df1 | df2 | p |
| final | 0.740 | 1 | 103 | 0.392 |
A t-test tests the assumption of equal variances between two groups through Levene’s test (Kim, 2015). The “final” variable in the output was subjected to Levene’s test to determine whether there was equal variance between the two groups (attendees and non-attendees of the review session). There are df1=1 and degrees of freedom (df1, df2), and a test statistic (F) of 0.740. The p-value of the test is greater than 0.392, the default alpha of 0.05. The null hypothesis, i.e., the two groups’ variances are equal, cannot thus be rejected. That is, the result of the Levene test does not indicate a violation of the assumption of homogeneity of variances. The homogeneity of variances necessary for the ttest is therefore met, and the variances of the two groups are equal.
Results & Interpretation
Independent Samples T-Test
| t | df | p |
| -0.410 | 103 | 0.682 |
Note. Student’s t-test.
Descriptives
| Group | N | Mean | SD | SE | Coefficient of variation |
| Attended review session | 55 | 61.545 | 7.356 | 0.992 | 0.120 |
| Did not attend the review session | 50 | 62.160 | 7.993 | 1.130 | 0.129 |
Every group’s means and standard deviations are given below:
• Participant group in the review session: m= 61.545, sd = 7.356
• Did not attend the group review session: sd = 7.993, m = 62.160
To find whether there was any difference between the means of two groups, the independent samples t-test was utilized. t-test findings are listed below:
• t = -0.410
• df = 103
• p = 0.682
Because there is not sufficient evidence, the null hypothesis cannot be rejected, as indicated by the p-value of 0.682, which is greater than the conventional alpha level of 0.05. Thus, we have the conclusion that there is no statistically significant difference in the end test scores among students who went to review sessions and those who did not. We cannot reject the null hypothesis based on statistical evidence, and the alternative hypothesis is weakly supported.
Statistical Conclusions
The objective of this study was to determine if students who participated in the review sessions and those who did not have statistically significantly different final exam scores. An independent samples t-test was conducted to test the equal variances assumption, which was acceptable. Based on the research, it was determined that there was no statistically significant difference between the mean final test scores for the two groups. Consequently, there was no proof to support the hypothesis that attending review sessions would assist in enhancing examination scores, and the null hypothesis couldn’t be verified.
RSCH FPX 7864 Assessment 3 t-Test Application and Interpretation
Independent samples t-Test can compare up to two groups at a time. tTest can compare one independent and dependent variable only; multiple comparisons aren’t supported. Yet another constraint of the t-Test is that it is able to hint at carryover effects, whereby instead of revealing group differences, it might lead to indications that there is something wrong with repeating the test. There might exist other factors like previous performance or ability to learn that have determined the results yet were not left out during the study. Beside this, there might also be additional factors for the results. For example, maybe review sessions did not go as planned or students might not have attended as frequently as required to impact their test scores.
Application
In applied behavior analysis (ABA), two distinct therapeutic interventions for the management of a given behavior problem in children with autism or other developmental disabilities can be contrasted using the independent samples t-test. Independent samples t-Test application is used in the practice of ABA by behavior analysts because of the following reasons. Student’s aggressive behavior intervention and design significantly depend on the result of t-tests of dependent variables.
Every subject in this specific student group is assessed twice on a result measure, like a pretest-posttest design, which is the essence of the t-Test application in their treatment. By investigating the mean differences in the dependent variable between treatment groups, I would be able to ascertain if the intervention was more successful in alleviating the targeted behavioral problem. It would be simpler for health workers, psychologists, social workers, and caregivers to make the most and best choices on intervention methods to be applied for their loved ones or patients with the help of this information.
References
Capella University (n.d.) 7864 Course Study Guide.
RSCH FPX 7864 Assessment 3 t-Test Application and Interpretation
Kim, T. K. (2015). T test as a parametric statistic. Korean Journal of Anesthesiology, 68(6), 540. https://doi.org/10.4097/kjae.2015.68.6.540
