Table 3.
Variables that were significantly associated with female advantage or male advantage in univariate analyses
| Variable | B | SE | p * | OR |
| Females advantage | ||||
| Year | ||||
| 1995 | not sig | |||
| 1996 | not sig | |||
| 1997 | not sig | |||
| 1998 | not sig | |||
| 1999 | not sig | |||
| 2000 | not sig | |||
| 2001 | 1.773 | .536 | 0.001 | 5.89 |
| 2002 | reference | |||
| Theme A vs. all others | -.823 | .312 | .008 | 0.44 |
| Theme B vs. all others | -.816 | .260 | .002 | 0.44 |
| Theme C vs. all others | -.823 | .312 | .008 | 0.44 |
| Theme D vs. all others | 1.498 | .322 | <.001 | 4.47 |
| having some in-class assessment | 1.822 | .493 | <.001 | 6.19 |
| having some short-answer questions | 1.008 | .264 | <.001 | 2.74 |
| having some TFA questions | -1.700 | .345 | <.001 | 0.18 |
| Males advantage | ||||
| Theme A vs all others | -1.248 | .624 | .045 | 0.29 |
| Theme B vs all others | 2.315 | .622 | <.001 | 10.12 |
| Theme C vs all others | -1.248 | .624 | .045 | 0.29 |
| having some short answer questions | -1.548 | .745 | .038 | 0.21 |
| having some TFA questions | 3.260 | .744 | <.001 | 26.04 |
corrected for multiple comparisons
OR = Odds ratio; Theme A = the cell; Theme B = the man; Theme C = the society; Theme D = the doctor.