Vormistamine ülesanne 3
discussion of the distinction between constructs and survey questions follows these line of
reasoning.
To bridge the gap between theory and measurement, two distinct research strategies are advocated:
a theory driven or top down strategy, which starts with constructs and works toward observable
variables and a data driven or bottom up strategy, which starts with observations and works towards
theoretical constructs (cf. Hox & De Jong-Gierveld, 1990). For examples of such strategies we refer
to Hox (1997).
When a final survey question as posed to a respondent fails to ask about what is essential for the
research question, we have a specification error. In other words, the construct implied in the survey
question differs from the intended construct that should be measured. This is also referred to as a
measurement that has low construct validity. As a result, the wrong parameter is estimated and the
research objective is not met
annaabi.ee 
