Application: Random vs Constant Responding

In this section we will review two types of problematic responding: random responding and constant responding. Random responding is randomly entering numbers within a given range. Humans tend not to engage in true random responding, however. So the other extreme is for individuals to respond with a constant number across all questions (such as “3”). For each of the examples below, 100 of the 300 cases in the small Marsh SDQ data set (the same one we bootstrapped in Chapter 7) were either replaced with randomly generated cases with a uniform random distribution from 1-6 (integers only) or with a “3” to represent a 33% random responding or constant responding rate. Similar to previous chapters, we used ...

Get Exploratory Factor Analysis with SAS 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.