The present study wants to investigate the relationship between difficulty in emotion regulation and alexithymia with hostile attribution bias and anger its method is cross-sectional, specifically path analysis.
Participants
The participants were includes adolescents aged 12 to 18 years in Zanjan, Iran 2021. Using Cochran's formula, the sample size was estimated to be 345. The inclusion criteria included informed consent and confidentiality of the questionnaires online without mentioning the name, being a teenager, and exclusion criteria for the research were distorted and incomplete questionnaires. The convenience sampling was used to select the sample. Recruitment occurred from February 1th to April 10th 2021. Out of 345 participants in this study, 35 (10%) were 12 years old, 38 (11%) were 13 years old, 47 (13%) were 14 years old, 55 (15%) were 15 years old, 49 (14%) 16 years old, 61 people (17%) are 17 years old, 60 people (17%) are 18 years old. More details are reported in Table 1.
Table 1. Demographic characteristics of the subjects
Variable
|
Number
|
Percentage
|
Age
|
12
|
35
|
10
|
13
|
38
|
11
|
14
|
47
|
13
|
15
|
55
|
15
|
16
|
49
|
14
|
17
|
61
|
17
|
18
|
60
|
17
|
Educational status
|
Student
|
280
|
81
|
Dropout
|
65
|
18
|
Mother's education
|
Reading and writing
|
108
|
31
|
Secondary
|
163
|
47
|
University
|
74
|
21
|
Father's education
|
Reading and writing
|
90
|
26
|
Secondary
|
171
|
49
|
University
|
84
|
24
|
Number of children
|
1
|
101
|
29
|
2
|
83
|
24
|
3
|
105
|
30
|
4
|
56
|
16
|
Procedure
This is a cross-sectional study and was done in several steps. The first step was performed with the aim of identifying students with high bullying in a sample of 1150 people. Participants completed the Illinois Bullying Scale. In the next phase, participants who had a high bullying score in the first phase were asked to participate in the second phase. 345 of these people participated in the second stage. In this phase, participants completed four Difficulties in Emotion Regulation Scale, Toronto Alexithymia Scale, Questionnaire of Hostile attribution bias and State-Trait Anger Expression Inventory-2™ Child and Adolescent (STAXI-2™ C/A). All rights of the participants were protected during this study. All procedures performed in study involving human participants were according to the ethical standards of the National Research Committee. The executive process of this research has been approved by the Ethics Committee of Kermanshah University of Medical Sciences under No. IR.KUMS.REC.1400.608.
Instruments
Difficulties in Emotion Regulation Scale (DERS)
This scale was developed in 2004 by Gertz and Roemer(41) to assess the difficulty in emotion regulation, which has 36 items and 6 dimensions (Nonacceptance of Emotional Responses, Difficulties Engaging in Goal-Directed Behavior, Impulse Control Difficulties, Lack of Emotional Awareness, Limited Access to Emotion Regulation Strategies and Lack of Emotional Clarity .(each of the items of this scale in the 5-point Likert From very rare to almost always is arranged and the minimum and maximum scores are 36 and 180, respectively. The results indicate that this scale has good internal consistency and Cronbach's alpha was obtained above 0.80 in Gertz and Roemer research(41). Based on research conducted by Pandey, Saxena and Dubai(31), the Cronbach's alpha of this scale was 0.90. Also, in Ghasemi, Beyrami and Vahedi(42) research, the reliability of the test through Cronbach's alpha for the whole test was 0.95, which is a very desirable value. The reliability of DERS in the present study was obtained with Cronbach's alpha coefficient of 0.77.
Toronto Alexithymia Scale (TAS-20)
Toronto Alexithymia Scale (TAS-20) was developed by Taylor, Ryan and Bagby(43) and has three subscales (Difficulty in recognizing emotions, difficulty in emotion regulation and objective thinking) and 20 items. The items to score based on a five-point Likert scale from a score of 1 for "strongly disagree" to 5 for "strongly agree". The minimum and maximum scores are 20 and 100, respectively. And Cronbach's alpha coefficient in Chen, Xu, Chen and Chang(44) research for the whole scale was 0.83, test-retest reliability coefficient was 0.87 and the correlation of the three factor coefficients with the total scale score was from 0.72 to 0.82 Also, in Ghorbani and Bagheri(45) research, the reliability of the whole instrument by Cronbach's alpha method was 0.82. The reliability of TAS-20 in the present study was obtained with Cronbach's alpha coefficient of 0.75.
Questionnaire of Hostile attribution bias
HAB was measured using the SIP–ABQ, which was rated on a 4- point Likert scale from not at all likely (0) to very likely (3). The English language questionnaire was translated into Chinese and then back-translated by three bilingual speakers. This questionnaire consisted of eight vignettes. Each vignette described an ambiguous social situation followed by four questions that assess direct Hostile attribution bias (e.g., ‘The club members wanted to ignore me’), indirect Hostile attribution bias (e.g., ‘The club members wanted me to feel unimportant’), instrumental nonhostile intent (e.g., ‘The club members were more interested in talking among themselves’), and benign intent (e.g., ‘The club members didn’t hear me say ‘‘Hi”’). The sum scores of the 16 items that assess direct and indirect Hostile attribution bias represent the degree of HAB Coccaro et al(46) The alpha reliability in the sample was good (α = 0.89) and In the research of Wang et al(47) Cronbach’s alpha in the was 0.88 at Time 1 and 0.91 at Time 2.
State-Trait Anger Expression Inventory
The present questionnaire is taken from STAXI-2, which was designed by Spielberger (1) in 1988 to measure anger in adults. In this research, the Persian version of STAXI-2 C / A has been used to measure the experience, expression and control of anger in children and adolescents. In this research, the Persian version (STAXI-2 C / A) has been used In Heidari Nasab, Barabadi and Moghaddasin(49) research, Cronbach's alpha coefficient for State of anger and trait of anger was obtained from 0.94 to 0.88. The Cronbach's alpha coefficient for the clinical group on the anger scale scale was 0.94, the anger trait was 0.88, the outward anger was 0.84, the inward anger was 0.74, and the anger control was 0.89. The results show that this tool has a good reputation. In the Persian version, seven factors of feeling anger, verbal and physical expression of anger, angry temperament, and angry reaction, expressing anger outward, expressing anger inward and controlling anger were identified. Brunner and Spielberger(48) reported a Cronbach's alpha coefficient of 0.74 for the whole test, 0.87 in State of anger and trait of anger, 0.70 in outward anger, 0.71 inward anger, and 0.79 in anger control. In the present study, the reliability of this instrument was obtained with Cronbach's alpha coefficient for seven Persian scale factors of 0.70, 0.73, and 0.79. 0.72, 0.85, 0.80, and 0.72 respectively.
Statistical analysis
To analyze the data, descriptive statistical indices including mean and standard deviation as well as Pearson correlation coefficient and path analysis were used to test the hypotheses. The analysis was conducted with SPSS v.26 (IBM), and LISREL v10.2. Due to the high sample size, the normality is less important(50) Despite this, the skew index for the research variables was between -1 to +1, which indicates that the distribution of variables is normal. Therefore, there was no limit to the use of Pearson correlation coefficient and path analysis with LISREL software. Path analysis with ordinal data was conducted using the diagonally weighted least squares method (WLSMV). We examined the hypothesized model using all data obtained by the original questionnaire. The model fit indices were Chi square statistics, Chi square/df, Root Mean Square Error of Approximation (RMSEA), Comparative Fit Index (CFI), Tucker-Lewis Index [TLI, also known as the Non-normed fit index (NNFI)], Goodness of Fit Index (AGFI) and Adjusted Goodness of Fit Index (AGFI). The model was judged as having good fit when the overall picture of fit indices indicated good fit and excellent if all of them indicated well fit: RMSEA ≤ 0.05, CFI and TLI ≥ 0.95, and WRMR < 0.90 (51).