A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Psychological constructs are abstract concepts (intelligence, depression, anxiety) that cannot be directly observed but must be carefully defined and measured. A rigorous construct definition specifies what is included and excluded theoretically, while valid measures operationalize that definition through observable indicators. The relationship between conceptual definition and empirical measure determines whether conclusions about the construct are justified.
Start with a domain definition: 'Depression is persistent sadness and loss of interest in activities,' then develop multiple items assessing different facets. Examine how established measures (BDI, DASS) operationalize the same construct. Test whether your measure correlates strongly with related constructs and weakly with unrelated constructs.
From your work on operationalization, you know that every variable in a study must be translated from a conceptual definition into something observable and measurable. For concrete variables — age, reaction time, number of correct answers — operationalization is relatively straightforward. Psychological constructs introduce a harder problem: concepts like "depression," "working memory capacity," or "implicit racial bias" have no direct physical referent. You cannot see depression; you can only observe behaviors and reports that you believe reflect it. The discipline of construct definition is about making that inferential chain as defensible as possible.
The process begins with a nominal definition — a clear theoretical statement of what the construct is and what it is not. A good nominal definition is explicit about boundaries: depression includes persistent low mood, anhedonia, cognitive symptoms, and somatic changes, but it excludes normal grief reactions of limited duration. Without this boundary-setting, a measure can drift and end up assessing something adjacent (demoralization, fatigue, negative affect) that is correlated with depression but not the same thing. This step is where most measurement failures are seeded: researchers skip the careful conceptual work and jump directly to item writing, then are surprised when their scale behaves oddly.
The operational definition translates the nominal definition into a specific set of observable indicators — items, behavioral tasks, physiological signals, or coded judgments. The guiding principle is content coverage: the indicators should sample systematically from the full domain of the construct, not just the easiest or most obvious facets. Depression has cognitive, affective, behavioral, and somatic components; a measure that only captures mood (the most salient symptom) will underrepresent the construct and may perform poorly in populations where somatic symptoms are primary. Comparing established measures like the Beck Depression Inventory (BDI), the Patient Health Questionnaire-9 (PHQ-9), and the Depression subscale of the DASS reveals how different design choices produce measures that overlap substantially but emphasize different facets.
The relationship between nominal and operational definition determines whether your measurement conclusions are valid. A mismatch creates construct-irrelevant variance (the measure captures something outside the construct's boundary) or construct under-representation (the measure misses important facets). Both undermine the ability to generalize findings. A reliably administered but invalid measure is worse than a noisy but valid one, because reliability creates false confidence: you are very precisely measuring the wrong thing. This is why construct definition must precede item writing, not follow it — once a scale has been deployed and accumulated validity evidence, its implicit construct definition becomes very hard to revise without starting over.