Reading focus
Syllabus: How Do We Describe and Measure Individual Differences?
Trace the route from a construct definition to an item pool, structural analyses, and validation evidence.
Why you’re reading it. Learn the general logic of building a self-report measure before following one construction project in detail.
A question to consider. A draft conscientiousness scale becomes more internally consistent after the team drops every item about keeping commitments to others. What has been gained, and what might the scale no longer measure?
Why this reading
Scale construction begins with a defensible account of the attribute and its boundaries. Clark and Watson connect that account to item selection, sampling, dimensional structure, and evidence that a proposed measure adds something useful.
Before you read
An item pool is the set of candidate questions. A factor represents shared variation under a statistical model. Unidimensionality concerns one underlying dimension; internal consistency concerns relationships among items. They are not interchangeable.
Read for
- Follow how the definition determines which content belongs in the initial item pool.
- Ask why closely related constructs and heterogeneous validation samples matter.
- Distinguish improving a coefficient from improving coverage of the intended attribute.
Consider afterwards
- A shorter questionnaire has more similar items but loses an important aspect of a construct. What evidence would justify keeping or rejecting it?
Editorial context
This updates the authors’ 1995 paper, with an emphasis on language-based instruments relevant to clinical assessment. Its recommendations require judgment about constructs and populations; they do not constitute a universal mechanical recipe for every kind of measure.