DSRT 734: Comparing Teacher Evaluations Before and After Training

DSRT 734 Question

Question 1
5 Points

A school has 52 teachers who have been teaching last year and this
year. The summer between the two years they had a training for all the
teachers to be more effective in their teaching. The school wants to
test if the teachers scored significantly better this year on their
course evaluations compared to last year. What is the best test for
them to run? The evaluations scores are scored between 0 and 100 and
are normally distributed.

Option A

ANOVA with a post hoc test


Option B

Wilcoxon signed-rank test


Option C

Paired samples t-test


Option D

Pearson's correlation

What this question assesses

This item assesses whether the relationship between observations, the number of measurement occasions, the quantitative outcome, and the distributional statement are used together when selecting an inferential procedure.

How to approach this type of question

Build a design description before considering the option labels. Identify who is measured, how many evaluation scores each teacher contributes, and whether last year’s and this year’s observations can be linked teacher by teacher. Distinguish a within-person comparison from samples composed of different people. Then note that the outcome is a 0-to-100 score and record the prompt’s distributional information, while remembering that assumptions for a change analysis concern the pattern of paired differences. The phrase “scored significantly better” indicates a directional research claim, but direction does not determine whether measurements are dependent. Compare each candidate’s required data structure with this design: some methods compare several independent groups, some summarize association, and others evaluate repeated measurements. Also consider whether nonparametric reasoning would be needed if the stated conditions did not hold. This page leaves the final option selection to the reader.

Check that all 52 teachers have two identifiable scores and that no staff turnover breaks the implied pairing. If some teachers lack one year, document the resulting complete-pair count. Separate the question of method family from the direction of the alternative and from causal attribution to training. Even a supported score change could reflect other year-to-year conditions, so a final interpretation should remain within the comparison design.

Related DSRT 734 questions

• DSRT 734: Choosing a T-Test for Two Independent Branches
• DSRT 734: Comparing Midterm and Final Exam Scores in JASP
• DSRT 734: Comparing Grades From Different Students Across Two Years

Related DSRT 734 resources

• University of the Cumberlands course resources
• DSRT 734 Inferential Statistics in Decision-Making
• DSRT 734 Questions: Inferential Statistics in Decision-Making
• How to Choose an Inferential Statistical Test