feedback Sharmin

 

Evaluation Question

Does participation in the NYAP program lead to a significant reduction in internalizing symptoms from pre-test to post-test among youth participants?

2. Determining the Statistical Test

Because my evaluation involves pre-test and post-test scores for the same participants, I will use a within-subjects (repeated measures) statistical approach.

Before choosing the final test, I will check whether the difference scores (Post – Pre) meet the assumptions for parametric testing:

Normality Assumption Check

In Jamovi, I will:

• Create a new variable representing the difference scores
• Run the Shapiro-Wilk test
• Inspect the histogram and Q-Q plot
• Review skewness and kurtosis values

Decision Guideline

• If the difference scores are normally distributed:

I will use a Paired Samples t-test, which compares the mean pre-test score to the mean post-test score.

• If the normality assumption is violated:

I will use the Wilcoxon Signed-Rank Test, which is the appropriate non-parametric alternative.

This step ensures that the statistical method aligns with the characteristics of my dataset and provides the most accurate interpretation of change.

3. Checking Dataset Quality (Missing Data & Outliers)

Before running the analysis, I reviewed my dataset for issues that could affect results:

• Missing Data: I confirmed that all participants had both pre-test and post-test values. If any were missing, I would decide whether listwise deletion or imputation was appropriate, but my sample is complete.
• Outliers: Because the PSC-17 uses small item-level scoring (0–2), extreme outliers are unlikely. However, I will still evaluate boxplots for both variables to confirm there are no unusual total scores.

Ensuring a clean dataset strengthens the credibility of the evaluation.

4. Presenting My Results in the Final Report

For my Week 6 written assignment, I plan to present results clearly and visually:

• A paired comparison table showing means, standard deviations, and the test statistic (t-value or Wilcoxon Z).
• A bar chart or line graph displaying pre-test and post-test average internalizing scores.
• A brief narrative summarizing whether scores decreased and whether the change was statistically significant.