Health Sciences 2801A/B Lecture 6: Reading 6 – Retest Consistency and Bias.docx

• School: Western University
• Department: Health Sciences
• Course: Health Sciences 2801A/B
• Professor: Tsorng- Yeh Lee

Document Summary

A confidence interval is centered around an adjusted observed score (xa: the width of a confidence interval depends on, the desired level of confidence (z, the amount of unsystematic error (sem, sem depends on, the internal consistency ( alpha , standard deviation of total scores (sx) Confounds: a confound is an extraneous factor that produces systematic bias, placebo effect, a patient experiences a reduction in symptoms following an inert medication that they believe to be effective, maturation, natural changes that occur over time, including normal development, natural progression of illness, learning, etc, history, specific events that affect persons" beliefs and behaviours. An asymmetric ci describes the combined effect of unsystematic error and systematic error: positive systematic bias moves the upper bound of a confidence interval in an upward direction, negative systematic bias moves the lower bound of a confidence interval in a downward direction, an adjusted standard error (sea) is used to calculate the biased bound of the ci.