“…Rather, other constructs or sources of method effects additionally need to be represented as part of the model of measurement in order to account for the observed variation. For example, in reasoning scales, speed effects, difficulty effects, and item-position effects have been found to account for the response variation besides reasoning (e.g., Borter, Troche, & Rammsayer, 2018; Lozano, 2015; Zeller, Reiss, & Schweizer, 2017). Regarding questionnaires, the sources of the item wording and item position effects have been found to exert an additional influence on responding so that the one-factor model considering the trait only is likely to fail in an investigation into such data (e.g., Distefano & Motl, 2006; Knowles, 1988).…”