Validity

BoardResearch makes use of instruments to operationalize a number of themes. In the academic literature we refer to such themes as constructs. For instance, interpersonal 'trust' within a board can be considered a relevant theme to assess the functioning of boards. To measure 'trust' we use instruments that have been developed by behavioral scientists to measure the construct of trust. In particular, attention is paid to the 'construct validity' of the construct trust.

BoardResearch aims to develop instruments that have content validity, convergence validity, discriminant validity, and nomological validity. It should be mentioned that constructs usually are measured by using several statements (or questions) and that a single construct can comprise several dimensions.

Content validity:
Are all aspect of the construct covered?

Convergence validity:
Does the instrument measure what it intends to measure? This is also referred to as reliability.

Discriminant validity:
Is the construct sufficiently different from other constructs? That is, aren't we actually measuring the same thing twice but refer to it differently?

Nomological validity:
Does the construct behave as one would expect within the nomological net of related constructs? For instance, normally speaking we would expect a negative relationship between interpersonal trust and interpersonal tension.