Skip to main content

Table 3 Variance components from partially crossed generalisability study

From: The reliability and validity of a portfolio designed as a programmatic assessment of performance in an integrated clinical placement

Component and their interaction

Explanation of interactions

Variance components

%

Student

The variance in marks due to true differences in the capability of the student

17.18

11.28

Task

The consistent tendency for one task to be marked higher or lower than others

17.81

11.70

Assessor

The tendency for one assessor to score a task highly and another to score the same task poorly

21.13

13.88

Assessor with student

The varying views that assessors have of students capability

5.60

3.68

Assessor with task

The tendency for an assessor to mark a task higher or lower in addition to their usual stringency/leniency

16.10

10.58

Student with task

The tendency for the student to engage with one task and not another

74.42

48.88

Residual error

Residual variation not explained by other factors

0.00

0.00