📝 Exercise: Worker Students in Dr. Zeppo’s Class
Dr. Zeppo is concerned about whether worker students, who may have less time to study due to their jobs, perform differently compared to the rest of the class. The dataset provides grades for worker students, and the general mean for non-worker students is known to be 80/100. Your task is to determine if there is a significant difference in performance between worker and non-worker students.
Dataset Location
Navigate to: Data Library > lsj-data > Zeppo
Instructions
1️⃣ Check the Data
- Summarize the Data:
- Calculate descriptive statistics for worker students (e.g., mean, median, standard deviation).
- Visualize the Data:
- Create histograms or boxplots to examine the distribution of grades for worker students.
2️⃣ Write the Hypothesis
- Null Hypothesis (H₀): The mean grade of worker students is equal to the general mean of 80/100.
- Alternative Hypothesis (H₁): The mean grade of worker students is different from the general mean of 80/100.
3️⃣ Choose the Appropriate Test
- Since the goal is to compare the mean of a single sample (worker students) to a known population mean (non-worker students):
- Use a one-sample t-test.
- Check Assumptions:
- Normality: Test whether the grades of worker students are approximately normally distributed (e.g., Shapiro-Wilk test).
- Scale/Continuous Variable: Ensure the grades are measured on a continuous scale.
4️⃣ Report the Results
- Include:
- Test statistic (e.g., t-value from the t-test).
- Degrees of freedom (df).
- P-value: Determine whether the results are statistically significant.
- Effect size: Calculate Cohen’s d to measure the magnitude of the difference.
- Interpretation:
- Does the data support the hypothesis that worker students perform differently compared to non-worker students?
- What insights can Dr. Zeppo derive to support worker students better?
5️⃣ Further Recommendations
- Suggest strategies to help worker students improve their performance based on the results.