Validity

A meta-analytic review of achievement goal measures: Different labels for the same constructs or different constructs with similar labels?

“…Without adequate conceptualization, achievement goal measures may be perfectly reliable, but they may not measure the intended construct. It is only when constructs have been adequately defined and measured that we can make valid inferences to theory (Barron et al, 2008; Campbell, 1969). Our review clearly indicates that there is a lack of both conceptual and operational consistency, which has resulted in an overall misalignment between theory and measurement.…”

“…Conceptual disagreements and measurement imprecision may also contribute to the lack of consensus over the most appropriate theoretical framework for achievement goal theory (e.g., Elliot, 2005; Kaplan & Maehr, 2007), and whether the mastery-avoidance goals are a necessary addition (Cury et al, 2006; Easter et al, 2008; Hulleman, Rhee Bonney, et al, 2006). One way to determine which model most accurately reflects the available data is to examine the predictive and discriminant validity of model components (Barron, Brown, Egan, Gesauldi, & Marchuk, 2008). For example, if performance-approach and performance-avoidance strivings are distinct constructs, then we would expect them to be relatively uncorrelated with each other (discriminant validity) as well as to have different patterns of associations with outcomes (predictive validity).…”

“…In this approach, achievement goals were operationally defined by the constituent items in survey instruments. Although the instruments are imperfect representations of the theoretical constructs, the individual items provide the most proximal representations of the construct (Barron et al, 2008; Campbell, 1969; Shadish, Cook, & Campbell, 2002) and enabled us to test the theoretical underpinnings of the differing frameworks. Specifically, we first quantified differences in the operationalization (i.e., measurement) of achievement goals.…”