2.2.3. Language and literacy skills The Test of Preschool Early Literacy (TOPEL; Lonigan, Wagner, Torgesen, & Rashotte, 2007) was used to assess children’s early language and literacy skills. All three subtests were administered, which included Definitional Vocabulary, Phonological Awareness, and Print Knowledge. In the Definitional Vocabulary task, the child was presented a picture and was asked to label it and provide a definition. In the Phonological Awareness task, the child was asked to complete a variety oftasks involving manipulation of sounds in syllables and words. In the Print Knowledge task, the child was asked to identify various print concepts, name letters, and say sounds associated with letters. Points for the subtests were summed up to yield raw scores, which were used for the present analyses. According g to the testmanual(Loniganet al., 2007),the internal consistency () of the subsets of scores and composites ranged from 0.86 to 0.96 for 3- to 5-year old children. The three subsets of language and learning skills are correlated at the moderate level (Table 2). The scores of these sub-scales were totaled to represent children’s language and literacy skills in the actor-partner interdependence models. 2.2.4. Problem behaviors The Problem Behavior Scale ofthe preschool version ofthe Social Skills Rating System for teachers (SSRS; Gresham & Elliott, 1990) was implemented to measure children’s externalizing and internalizing behaviors. The problem behavior rating scale contains 10 items. The teacher rated each item on a 3-point rating scale (never, sometimes, very often). The externalizing factor measures children’s aggressive, disruptive, and defiance behaviors. The internalizing factor measures children’s loneliness, anxiety, and depression. The internal consistency () of the externalizing and internalizing behavior scores for the current data set was 0.82 and 0.75, respectively. Considering externalizing and internalizing problems follow distinctive developmental trajectories in young children (Gilliom & Shaw, 2004) and were associated with children’s learning-related behaviors (McWayne & Cheung, 2009) and language and literacy skills (Doctoroff et al., 2006), both of the behavioral indices were used in the current analyses.