constructs
arnold on Sep 16, 2008
Constructs are latent variables. A latent variable cannot be measured directly, but only through measurable indicator variables. When it comes to psychological traits, you cannot simply observe a person and be able to record a measure of these traits. For example, if you needed a measure of organizational commitment you could not simply sit a person down and observe their commitment to an organization. Additionally, there are many reasons that you cannot simply ask a person to self-report their organizational commitment because of the plethora of problems that occur with self-reporting such as acquiescence bias. Thus indicator variables, from a survey in our case, are used to form a construct. Put more simply, using a survey, a researcher could have five (or 3 or 7 or 20) questions that are used to form a construct of a psychological trait that is unobservable. While there are many ways to combine indicators to achieve a measure of a construct, all methods assume that what is being measured is a single entity, even if it is an abstraction like “efficiency” or “happiness.”
More Formally:
A construct provides an efficient and convenient method for labeling a number of similar behaviors. Through the use of constructs, the observer can begin to classify and group instances of similar behavior and communicate in compact terms what has been witnessed.
At a simple level, psychological theory is a statement of the possible relationship between two psychological constructs, or between a construct and an observable phenomenon of practical consequence.
Because psychological constructs are abstractions which can only be assessed indirectly, the design of instruments to measure such variables present major challenges.
Referenced from(CA, 1986, p. 5).