12 Assessing Reliability
12.1 Overview
When discussing measurement errors in Chapter 5, we differentiated errors that are random in nature from errors that are systematic in nature. Random errors are unpredictable and inconsistent, and they affect (increase) variability of responses but not the mean response, whereas systematic errors are predictable and consistent and they affect the mean but not the variability of responses. The amount of random errors is inversely proportional to reliability of scores, and the amount of systematic errors is inversely proportional to validity of scores. Sources of random and systematic errors are the problems in the sampling design (Sampling Design Error—SDE) and process (Sampling ...
Get Survey Development now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
