Ninety children (42 girls, Mage = 12.16 years, SD = 1.32 years, age range: 9–14 years) and their parents participated in the study. They were recruited through 13 mainstream schools in Hong Kong: eight of them were secondary schools and five were primary schools. Informed consent was obtained from both the children and their parents (please see the Appendix for the sample of an informed consent form). All of the participants were ethnically Chinese and none was found to have neuropsychological disorders based on the Kiddie Schedule for Affective Disorders and Schizophrenia for School Aged Children – Present and Lifetime version (K-SADS-PL; Kaufman et al., 1997).
Multidimensional Anxiety Scale for Children (MASC). The MASC (March et al., 1997) is a self-report measure of the severity of anxiety symptoms in the past two weeks. It consists of 39 statements which are rated on a 4-point Likert scale, ranging from 1 (never true about me) to 4 (often true about me). A total score was derived from the sum of four subscale scores: Physical Symptoms, Social Anxiety, Separation Anxiety/Panic, and Harm Avoidance. A higher MASC total score indicated a higher anxiety level. Cronbach’s alphas were reported to range from .74 to .85, indicating moderate to good reliability (March et al., 1997). A Chinese version of the MASC was used in the present study.
Wisconsin Card Sorting Test (WCST). The WCST (Heaton et al., 1993) is a widely used measure of EF. It assesses the shifting ability which is to shift cognitive strategies in response to changing rules. The participants were asked to match 64 response cards with four stimulus cards one-by-one. Both the response cards and stimulus cards depict figures of different colours, forms and numbers of objects, and the participants had to sort the cards based on these three categories. However, they only received feedback regarding the correctness of each of their sorting responses and were not explicitly informed of the sorting rule which also changed regularly. Therefore, the participants had to adjust their sorting methods by inferring from the examiner’s feedback.
The number of perseverative errors was used to indicate the level of cognitive flexibility or EF. Perseverative errors were made when the participants persisted in sorting the cards based on a rule that was no longer correct as inferred from the examiner’s feedback. The raw scores of perseverative errors were transformed to T scores, and a lower T score indicated the commitment of more perseverative errors or weaker EF. The generalisability coefficient for Perseverative Errors was reported to be .52 (Heaton et al., 1993), which indicated moderate to good reliability.
Tower of London-Drexel University 2nd Edition (ToLDX 2nd Ed). The ToLDX 2nd Ed (Culbertson ; Zillmer, 2005) was used as another measure of EF, especially in the inhibition ability. It was intended for children between the ages of 7 and 15. In the ToL task, the examiner and participant each had a wooden board with a set of three beads (red, blue, and green). On each wooden board, there were three vertical pegs of different heights: the tallest peg could hold at most three beads, the medium peg could hold two beads and the shortest one could hold one bead only. The participant had to move beads from the starting configuration to the goal configuration as fast as possible, also with the fewest number of moves possible. Only one bead could be moved at one time. After the demonstration of the moves and rules using the examiner’s wooden board, two practice problems were presented. Then the test began and 10 test problems requiring a minimum number of three to seven moves were shown one at a time.
Performance on the ToL was indicated by several measures: Total Move (deduction of minimum number of moves from total number of moves made by the participant), Total Correct (number of test problems solved in a minimum number of moves), Initiation Time (the time from being presented the test problem to removing the first bead from a peg), Execution Time (the time from moving the first bead to problem completion), Total Time (the sum of Initiation Time and Execution Time), Time Violation (the number of times when the participant could not complete the test problem within 60 s), and Rule Violation (number of times when a rule is violated).
Dyadic Parent-Child Interaction Coding System-IV (DPICS-IV). A 15-minute recording of the interaction between each child and his/her parent was obtained. During the recording, the parent-child dyad was asked to plan an activity that both of them would be willing to participate together, such as going on a trip or to the cinema. The DPICS-IV (Eyberg, Nelson, Ginn, Bhuiyan, & Boggs, 2013) was used to code the parent’s verbalisations, and determine the number of Direct Commands (DC; declarative statements containing an order) and Indirect Commands (IC; suggestion in the form of a question/wish, or does not indicate the necessity of compliance) directed to the child. Parental demandingness was represented by a Demand score which was the sum of DC and IC. A higher Demand score was considered to indicate a higher level of parental demandingness. The inter-coder reliability coefficients of DC and IC were reported to be .82 and .66 respectively (Eyberg et al., 2013).
The participants completed two sessions in total (T1 and T2), approximately one year apart. Both sessions took place at the University of Hong Kong. At T1, the children completed the MASC, WCST and ToLDX 2nd Ed individually. Both the children and their parents also participated in the DPICS-IV observation task, and only one parent-child dyad was observed each time. The order of administration of the four tasks was randomised across the participants. At the T2 follow-up, the children’s anxiety level was re-evaluated by completing the MASC again, in order to investigate the effects of EF measured at T1 on the anxiety level at T2 as well as whether the relationship would be moderated by parental demandingness at T1, while controlling for T1 anxiety level. The entire experiment lasted approximately two hours and all the participants received $300 cash as compensation.