Skip to main content
International Journal of STEM Education

Table 3 Outcome of ordered logistic regression for each survey question

From: Gender patterns in engineering PhD teaching assistant evaluations corroborate role congruity theory

Reference gender = M

TAs in analysis (n)

Estimate

Odds ratio

Lower–Upper CL

z statistic

Holms-Bonferroni adjusted p-value

1 Demonstrates command of the subject matter

610

0.424

1.529

1.314–1.743

3.878

0.0021

7. Provides clear, relevant and understandable responses to my questions

608

0.412

1.509

1.297–1.722

3.795

0.0028

16. Fair in grading

609

0.340

1.405

1.189–1.621

3.084

0.0349

20. Quality of your TA's teaching? (1 being poor, 5 being great)

590

0.335

1.398

1.203–1.593

3.372

0.0134

3. Provides clear and comprehensive explanations and instructions

608

0.333

1.394

1.172–1.617

2.930

0.0509

9. Actively helpful when students need assistance

611

0.3113

1.365

1.157–1.573

2.931

0.0541

15. Makes effective use of illustrations and examples

607

0.289

1.335

1.103–1.566

2.438

0.0888

12. Periodically checks to make sure students understand what was covered

607

0.281

1.325

1.105–1.545

2.506

0.1220

17. Provides helpful comments on my assignment

592

0.277

1.319

1.120–1.517

2.728

0.0892

13. Provides periodic summaries of what has been covered or discussed

582

0.263

1.301

1.091–1.511

2.454

0.0987

11. Communicates clearly

597

0.262

1.299

1.109–1.489

2.702

0.0896

4. Emphasizes the conceptual basis of the problem set or the lab experiment

597

0.258

1.294

1.092–1.496

2.502

0.1116

18. Makes effective use of visual aides (blackboards, overhead, slides etc.)

583

0.253

1.288

1.091–1.485

2.518

0.1298

14. Effective at relating lecture material to what is covered in section or lab

589

0.249

1.283

1.095–1.471

2.600

0.1118

10. Seems enthusiastic about teaching the material

595

0.247

1.280

1.080–1.479

2.416

0.0785

5. Encourages students to think in class by asking questions

608

0.232

1.261

1.043–1.478

2.088

0.1472

6. Makes me feel free to ask questions and express my opinions

581

0.231

1.260

1.077–1.442

2.481

0.1048

19. Divides his/her time equitably among laboratory groups

565

0.2066

1.229

1.033–1.426

2.057

0.1191

2 Fully prepared for class, laboratory or review section

603

0.182

1.199

0.975–1.424

1.587

0.2260

8. Evaluate this course as a whole. (1 being poor, 5 being great)

614

0.074

1.077

0.890–1.264

0.780

0.4350

  1. The Odds ratio is interpreted as: the odds that a male TA has a higher response level is (the Odds Ratio) times higher than it is for a female TA in our data set. 95% confidence intervals are also reported. Bonferroni-Holms post-hoc adjustment was applied to correct for multiple comparisons in the interpretation of statistical significance. Data were sorted so that odds ratios are listed from largest to smallest
  2. *Holms-Bonferroni corrected α used to correct for multiple comparisons. Data were sorted by unadjusted p-value (smallest to largest) and the corrected p-value was determined by multiplying the unadjusted p-value by total number of comparisions (20) for the most significant outcome and multiplying each successive p-value by one less comparision for each subsequent correction. Adjusted p-values ≤ 0.05 are in bold
Back to article page