Psychology Week 6 Assignment: t-Test Application and Interpretation

profileGMcNeel0
Week_6_T_Test_Information_Sheet.docx

Week 6: Independent Samples t-Test

Statistical information sheet | For drafting the Capella DAA assignment

Data source: grades.jasp (105 student records). Calculations were performed in Python (SciPy), not JASP. Confirm that equivalent statistical output is acceptable for submission.

Section 1 — Data Analysis Plan

Variable

Type / role

Definition

Review

Categorical; independent variable

Review-session attendance: 0 = no; 1 = yes in the analyzed file

Final

Continuous; dependent variable

Number of correct answers on final exam

Research question: Is there a statistically significant difference in final exam scores between students who attended review sessions and those who did not?

Null hypothesis (H₀): The population mean final exam scores are equal for the two groups.

Alternative hypothesis (H₁): The population mean final exam scores differ between the two groups.

Planned analysis: Two-tailed independent samples t-test; significance level α = .05.

Section 2 — Testing Assumptions

Test

Statistic

p-value

Mean-centered Levene’s test

F(1, 103) = 4.276

.041

Levene’s test examines whether the two groups have equal variances.

Because .041 < .05, the homogeneity-of-variances assumption is violated.

Use Welch’s independent samples t-test rather than Student’s equal-variance t-test.

Section 3 — Results and Interpretation

Descriptive statistics

Review attendance

n

Mean (M)

SD

Did not attend

70

12.00

8.12

Attended

35

10.60

6.59

Welch’s independent samples t-test

Statistic

Value

t

0.95

Degrees of freedom

81.97

Two-tailed p-value

.346

α

.05

Mean difference (no review − review)

1.40 points

The no-review group had a higher sample mean, but the difference was not statistically significant.

Because p = .346 > .05, do not reject the null hypothesis.

These results do not establish that review attendance causes an increase or decrease in final exam scores.

Section 4 — Statistical Conclusions

Compared final exam scores for two independent groups; Welch’s test addressed unequal variances.

The observed 1.40-point difference did not reach statistical significance.

Limitations: unequal group sizes (70 and 35); possible differences in prior preparation, motivation, and study habits; no demonstrated random assignment; limited generalizability.

Alternative explanations: self-selection into review sessions, prior knowledge, study time, and other unmeasured student characteristics.

Section 5 — Application to Psychology and Human Resources

Component

Example

Independent variable

Employee training participation (attended vs. did not attend)

Dependent variable

Numerical employee job-performance score

Test

Independent samples t-test

Compare mean performance scores for employees who did and did not attend training.

Results could inform evaluation of training programs and professional development.

A difference between groups would not by itself prove training caused the difference.

Notes for preparing the final submission

Transfer these points into the required Capella DAA template and develop explanations in the student’s own words.

Include the actual statistical output tables and identify the software accurately; seek instructor approval if JASP output is required.

Check dataset coding against the course instructions before finalizing; the analyzed file used 0 = no and 1 = yes.