Psychometric Testing of the German Versions of Three Implementation Outcome Measures

Internationally comparable assessments of implementation outcomes are a prerequisite for the accurate and useful exchange of implementation science results. We translated the English versions of the acceptability of intervention measure (AIM), the intervention appropriateness measure (IAM), and the feasibility of intervention measure (FIM) into German following recent guidelines. We tested the structural validity, the internal consistency (Cronbach’s α), and the construct validity (known-groups validity) in a hypothetical situation study of secondary school teachers (n1 = 142) and the structural validity and the internal consistency in a pilot field study of primary school teachers participating in a school health promotion project (n2 = 61). The hypothesized three-factor solution (acceptability, appropriateness, and feasibility) revealed the best model fit in the hypothetical situation study (CFI = 0.97; RMSEA = 0.08). The second study tended to support these results (CFI = 0.99; RMSEA = 0.05) only when the error terms of two items from different constructs were allowed to correlate with each other (“seems like a good match” [IAM] and “seems possible” [FIM]). Furthermore, the intercorrelations between the three factors were higher in the pilot field study than in the hypothetical vignette study (0.71–0.86 vs. 0.06–0.35). The instruments’ internal consistency proved to be high in both studies (Cronbach’s α: 0.91–0.97). We gathered positive evidence for the known-groups validity, as participants who received different stimuli before answering the instruments rated them statistically significantly different. Our study showed that the German versions of the instruments could successfully be applied in health promotion research. We suggest using the four items per instrument rather than an abbreviated version and discuss recommendations for the instruments’ application.