Reading focus
Syllabus: How Do We Describe and Measure Individual Differences?
Follow the distinction between structural questions, about how people differ, and process questions, about why they differ, and the chapter’s comparison of different sources of evidence about people.
Why you’re reading it. Establish what the field is trying to do before examining how its measures are built, and meet an early challenge to treating questionnaire numbers as measurements.
A question to consider. The team’s pilot asks learners to rate “I keep my work organized” from 1 to 5, and the average is 4. What does that number allow the team to say, and what would Cooper’s discussion of Michell caution against?
Why this reading
This short opening chapter sets out what the psychology of individual differences tries to do. It separates describing how people differ from explaining why, compares several sources of evidence about people, and ends by asking whether questionnaire scores are measurements in the ordinary sense.
Before you read
Much of psychology looks for regularities that hold across people; individual-differences research studies systematic variation between them. Cooper calls accounts of how people differ structural models and accounts of how those differences arise process models. Psychometrics is the study of psychological measurement. A rating scale assigns numbers to answers such as “agree” or “strongly agree”; whether those numbers behave like measurements of length or weight is the question Cooper raises through Michell’s (1997) critique.
Read for
- Examine the four reasons given for studying individual differences, including the claim that measuring them can make experimental comparisons more sensitive.
- Compare clinical theories, close knowledge of individuals, armchair observation, and test-based assessment as sources of evidence, noting the weakness attributed to each.
- Follow the argument that research on processes must wait for agreement about structure, and consider whether description and explanation might instead develop together.
Consider afterwards
- One respondent agrees with a questionnaire item, scored 4, and another disagrees, scored 2. What, if anything, would justify saying the first has twice as much of the trait, and what can such scores legitimately support?
- Cooper argues that researchers should establish whether a dimension such as sociability matters before studying how it works. Could evidence about a proposed process help show whether the dimension is worth measuring?
Editorial context
This is the opening chapter of a textbook, and it argues its case briskly. Several claims are best read as positions to examine: not every other branch of psychology assumes that people are much the same; structural and process research often inform each other rather than proceeding in strict order; and including a measured trait as a covariate can make a treatment comparison more sensitive, whereas asking whether a treatment works differently for different people requires testing an interaction. Michell’s critique, which Cooper presents sympathetically, remains debated. Responses include treating rating-scale scores as ordinal and using measurement models whose assumptions can be tested rather than presupposed.