Skip to main content

Table 4 Effects of standard and adjusted regression calculations on fail rates

From: Is the assumption of equal distances between global assessment categories used in borderline regression valid?

 

Station fails using standard regression calculation

Fail rate (standard calculation)

Station fails using adjusted regression calculation

Fail rate (adjusted calculation)

z-statistic

p-value (Critical p-value = 0.01)

Exam 1

198/3720

0.05

214/3720

0.06

1.17

0.243

Exam 2

118/2890

0.04

147/2890

0.05

2.73

0.006

Exam 3

167/2296

0.07

179/2296

0.08

0.96

0.335

Exam 4

206/2440

0.08

241/2440

0.10

2.44

0.014

Overall

689/11346

0.06

781/11346

0.07

3.62

< 0.001

  1. Legend: Data from four summative finals examinations at the institution studied. Null hypothesis that station fail rate using adjusted regression calculation = station fail rate using standard calculation (two-tailed test)
Back to article page

BMC Medical Education

ISSN: 1472-6920

Contact us