Measures

The CDI was used to assess depressive symptoms. It is a 27-item scale, where each item consists of three choices (0, 1, and 2). Scores range from 0 to 20, with higher mean scores corresponding to higher depressive symptoms. An example item is “I do most things okay” (0), “I do many things wrong” (1), and “I do everything wrong” (2). The CDI scale has been found to display acceptable internal consistency, reliability, and validity (Kovacs and Beck, 1977; Kovacs, 1992). In the current study, the CDI demonstrated acceptable internal consistency at both time points [α = 82, and 80. at Time 1 (T1) and Time 2 (T2), respectively]. We used the clinical cut-off ≥12 for mild depression using CDI (Allgaier et al., 2012) to explore clinical levels of depression in the current sample. We found at T1 20 (9.4%) children in the current sample would be classified as having clinical levels of depression; at T2, that number was 35 (16.4%); 16 (7.5) children reported those clinical levels at both T1 to T2. Those percentages are slightly above that found in the Allgaier et al. (2012), paper, but fit with the latest UK statistics on mood disorders for children of this age (Sadler et al., 2018). When using a more conservative cut-off ≥20 (Rivera et al., 2005), we do not find any children at T1 or T2 who would be classified as clinically depressed. No parents or children reported receiving treatment for depression during the 12 months of the study.

Loneliness in relation to peers was measured using the peer sub-scale of the LACA. LACA is a 48 item scale, separated into 4 scales of 12 items The peer sub-scale was used in the present study and includes items “I feel isolated from other people” and “I feel excluded by my classmates.” Children are asked to indicate how often each item applies to them on a 4-point scale: 4 (often), 3 (sometimes), 2 (rarely), or 1 (never). Higher mean scores on the scale are indicative of greater loneliness in relation to peers. The LACA has been found to display acceptable internal consistency, reliability, and validity (Maes et al., 2015a,b). Although originally used with Dutch-speaking children it has also been used with English-speaking children (Terrell-Deutsch, 1999; De Roiste, 2000; Qualter and Munn, 2002, 2005; Harris et al., 2013; Qualter et al., 2013). In the current study, this sub-scale demonstrated acceptable internal consistency across the two waves of data collection (α = 79 and 0.82. at T1 and T2, respectively). There are no clinical cut-off points for loneliness, but to explore how many children were scoring at the higher end of loneliness, we explored how many children answered “often” to more than half the questions at T1 and T2. Findings showed that 22 children (10.3%) at T1 and 41 children (19.2%) at T2 reported feeling lonely often; 21 children (9.9%) reported those high levels at both T1 and T2.

The scale consists of 101 items (of which 97 are scored) that measure different aspects of AEI: (A) Perceiving Emotions – children identify certain emotions in photographed facial expressions, (B) Using Emotions – children rank, using a standardized scale, the extent to which different emotions impact behavior and decision making, (C) Understanding Emotions – children read vignettes and select the answer representing what emotion the protagonist is feeling, (D) Managing Emotions – children read several scenarios and pick, from several options, the best solution for managing emotions in each scenario. Multi-Health Systems, the test distributor, scored the data using expert norms, providing means for each branch of the MSCEIT-Yv, and a total MSCEIT-Yv score. Acceptable split-half reliabilities have been obtained for the MSCEIT, e.g., 0.67 (perceiving) to 0.86 (understanding) and 0.90 for total AEI (Papadogiannis et al., 2009), and the hierarchical four-factor structure has been recovered in data from youth aged 9 to 15 years (Rivers et al., 2012). In this sample, branch scores were significantly intercorrelated [0.24 (correlation between Perceiving and Using branches)] to 0.70 (correlation between Understanding and Managing branches); See Table 1.

Means and standard deviations with bivariate associations.

The TEIQue-CF comprises 75 statements, tapping 9 facets of emotional functioning (e.g., emotion expression; emotion regulation; affective disposition). The child is asked to respond to each item using a 5-point Likert scale, ranging from “completely disagree” to “completely agree.” Example items include “I find it difficult to understand what others are feeling” and “If someone makes me angry, I tell them.” The TEIQue-CF has satisfactory levels of internal consistency for the global trait EI dimension (e.g., α = 0.89), but not at facet level (Russo et al., 2012; Stassart et al., 2017). Additionally, data collected with younger children suggest the tool is unidimensional in nature (Russo et al., 2012). In view of that evidence, current analyses were restricted to the use of the global TEI score (scored data provided by the test developers). In the current study, α = 0.87 for the full TEIQue-CF.