It might seem that the ideal IDA scenario occurs when identical measures are used across contributing studies (i.e., all studies measure the same variable in precisely the same way). In this scenario between-study heterogeneity in measurement may initially seem an irrelevant concern. However, even when identical measures are used across studies, distinct subpopulations may interpret or respond to the same item in different ways, above and beyond any actual differences in the underlying construct. These differences may reflect systemic influences of local norms in how participants’ view their research participation (i.e., participants in one location or sociocultural context may respond with less veracity than participants in other locations or contexts), of how items are interpreted within the context of the larger assessment battery of each study (e.g., the content of surrounding items; Rivers et al., 2009; Tourangeau et al., 2000) or of how items are administered across study (e.g., by interviewer, paper and pencil, or computer; Meade et al., 2007; Richman et al., 1999). In such cases, despite the fact that the item is identical, the values obtained from the different study samples would not necessarily have an identical scale or meaning.2
