Are Testlets Created Equal: Examining Testlet Construction Strategies in Sem.

Validity and construct contamination of the Racial Identity Attitude Scale—Long Form.

“…For each of the three subscales, we first created a set of “testlets,” or subsets of items, to serve as indicators. Following recommendations by Snell, Hall, and Foust (1997), we created testlets by combining conceptually related groups of from 3 to 5 items, which resulted in three testlets for each subscale. We constructed testlets by grouping items that appeared to the first two of us to reflect “secondary constructs” (Snell et al., 1997) as follows.…”

“…Following recommendations by Snell, Hall, and Foust (1997), we created testlets by combining conceptually related groups of from 3 to 5 items, which resulted in three testlets for each subscale. We constructed testlets by grouping items that appeared to the first two of us to reflect “secondary constructs” (Snell et al., 1997) as follows. For Preencounter, Testlet 1 involved anti-Black sentiments and included Items 4, 9, 12, 31, and 32; Testlet 2 items pertained to idealization of Whiteness and included Items 8, 25, 29, 40, and 41; Testlet 3 contained material involving discomfort with Black and White people and included Items 21, 44, 46, and 47.…”

Item Parceling Strategies in SEM: Investigating the Subtle Effects of Unmodeled Secondary Constructs

“…The multiple indicators of the latent constructs may be at different levels, as suggested by Bagozzi and Edwards (1998) All of these aggregation depths have been employed in organizational research (Snell, Hall, & Foust, 1997). In cases where the construct of interest is broadly defined or the measurement scale has multiple subscales or subfacets, indicators at a high level of aggregation may be most appropriate.…”