Health Sciences 2801A/B Chapter Notes - Chapter 5-6: Observational Error, Standard Score, Confidence Interval

• School: Western University
• Department: Health Sciences
• Course: Health Sciences 2801A/B
• Professor: Tara Lee

Document Summary

Value between 1 and 10: zero means that the two items are completely unrelated. Only representing unsystematic error: one means they are completely related. Unsystematic error: a type of measurement that increases or decreases individual scores by different amounts. To calculate it you subtract your alpha value from one: the amounts are unpredictable. The score is equally likely to increase or decrease as a result of unsystematic error. True score: a score with no unsystematic error, this is only hypothetical. As every i(cid:374)dicator is a(cid:374) i(cid:373)perfect sig(cid:374) of a perso(cid:374)"s true score. Confidence interval: probable range of true scores associated with an observed score, calculating it. You use 1-alpha (or the unsystematic error) Z score of 1. 96 = 95% confident. Z score of 1. 64 = 90% confident. Adjusted x: always going to be closer to the mean than the actual x value. Item universe: large, hypothetical collection of indicators. Every instrument has a set of a number of indicators.