Skip to main content
Fig. 2 | BMC Medical Research Methodology

Fig. 2

From: The effect of missing data and imputation on the detection of bias in cognitive testing using differential item functioning methods

Fig. 2

DIF estimates for hearing impairment in the reference scenario (No Additional Missingness) compared to the six missingness scenarios for select cognitive test items in the Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS). Each dot represents a difference between the two groups (impaired and unimpaired) for a single threshold of the ordinal test score. When the difference in thresholds is significantly different from zero (indicated by an unfilled circle), this indicates DIF. Differences between the reference scenario and each of the missingness scenarios illustrates the magnitude of error due to missingness. Estimated thresholds in the random scenarios are generally stacked vertically on top of the estimates with no additional missingness, indicating no systematic DIF estimation error. However, estimates for scenarios with missingness among lower cognitive test scores are shifted to the left, indicating systematic DIF estimation error in these scenarios

Back to article page