Validation of the Arabic Version of the Inventory of Coping Strategies of Competitive Sport ( ISCCS )

The aim of our study is to validate the Arabic version of the inventory of coping strategies of competitive sports (ISCCS) using factor analysis. After critical analyses of adapted sports versions, Gaudreau and Blondin (2002) proposed the ISCCS questionnaire to measure coping strategies in the field of competitive sports (André & Laurencelle, 2010). The ISCCS identify ten coping strategies that are divided in two dimensions: the coping task-oriented and the coping emotion-oriented. 419 athletes (273 men and 146 women; age 16.79 ± 3.82 years, 14 to 34) in different individual and team competitions, volunteered to participate in the study. Data were collected and analyzed for reliability and validity using the test-retest method, reliability, correlation analysis and confirmatory factor analyses. Statistical analysis was performed with SPSS version 22.0.0 the IBM AMOS. The confirmatory factor analyses showed good adjustments for coping models of task-oriented (chi-square 359.35, CFI: 0.92; TLI: 0.91; GFI: 0.93; RMSEA: 0.040 and p value < 0.001), and the coping oriented emotion (chi-square 215.45, CFI: 0.93; TLI: 0.91; GFI, 0.94; RMSEA: 0.054 and p < 0.001). In conclusion, ISCCS was translated and validated in various languages, including French, English, Spanish, and Turkish... Arabic version demonstrated good psychometric properties adjustments and can be used in other surveys in the area of sports competition.

In terms of performance optimization, the athlete must confront the stressful situation in order to maximize its chances of success (Molinero, Salguero, & Márquez, 2010).Lazarus and Folkman (1984) define coping as the set of cognitive and behavioral efforts to master the external and/or the internal voltages that mobilize or exceed human resources.Lazarus and Folkman (1984) propose the transactional model.It offers two dimensions of coping strategies: coping oriented problem (TOC) and coping oriented emotions (WCC) (Ponnelle & Lancry, 2002;Gaudreau & Blondin, 2002;Skinner, Edge, Altman, & Sherwood, 2003).The coping strategy is affected by personal characteristics (Costa, Somerfield, & Mac Crae, 1996), and the specific stressful circumstances (Chabrol & Callahan, 2004).
The second methodology is the Daily coping inventory developed by Stone and Neale (1984).It avoids the response distortion on an extended recall in the first methodology.In the Daily coping inventory, people are asked the same day the stressful situation.Recent studies develop and implement methods that would minimize the retrospective bias (Calmeiro & Tenenbaum, 2007;Calmeiro, Tenenbaum, & Eccles, 2010, 2014;Doron & Gaudreau, 2014;Evans, Hoar, Gebotys, & Marchesin, 2014;Gaudreau et al., 2010).
Generally all instruments adapted to the sports field: the WCQ (Folkman and Lazarus (1980)) as well as the COPE (Carver et al., 1989) have shown questionable results (Nicholls & Polman, 2007).The exception is the Athletic Coping Skills Inventory-28 (ACSI-28) developed by Smith, Schutz, Smoll, & Ptacek (1995), the Coping Style Inventory for Athletes (CSIA), developed by Anshel and Kaissidis (1997), the Coping Style Scale for Sport (CSSS), created by Anshel, Williams and Williams (2000) and the Coping Function Questionnaire for Adolescent and Sport developed by Kowalski and Crocker (2001).Gaudreau and Blondin (2002) have proposed an inventory of coping strategies in sports competition (coping strategies in sports competition inventory), specifically designed for the sport questionnaire.The ISCCS was created to evaluate athlete's actions to prepare for a competition (pre-competition) and those used for sporting competition (Gaudreau & Blondin, 2002).The ISCCS included a literature review of the coping actions used by sportsmen and a study analysis of the coping pressure, anxiety, and performance (André & Laurencelle, 2010).These two steps have allowed Gaudreau and Blondin to identify six subscales belonging to TOC: mental imagery, control of though-ts, relaxation, invested effort, research support, and logical analysis.Also, four subscales belonging to the WCC are defined: the extraction of negative emotions, the mental distraction, the resignation and the social avoidance.
The purpose of this study is to explore the validation of the Arabic version of ISCCS and examine its psychometric properties.

Methodology
The sample used in this study consists of 419 athletes (273 male and 146 female) that have voluntarily participated in this study (mean age 16.79 years and 3.82).They represent different competitive disciplines, including 294 team sports and 125 individual sports.
All our athletes regularly participate in official competitions nationwide, among them international required preparing to take part in the Olympic Games in 2016 (the Tunisian national team of canoe Kayak).Gaudreau and Blondin (2002) proposed the ISCCS.They identify 39 items, divided into ten subscales (mental imagery, control of thoughts, relaxation, invested effort, seeking support, analysis logic, negative emotions extraction, mental distraction, resignation and social avoidance).

Procedure
We are referred to the back-translation procedure (see Vallerand, 1989) to get the translation of the Arabic ISCCS version.
Athlete participation was voluntary.We require managers and coaches authorization of each athlete.Their anonymity was assured, and parental consent is required for those under 18 years.
Before data collection, all athletes were widely informed about the purpose and study procedures, and were informed that the results would be made available at the end of the study.Participants respond to a Liker scale of 5 points, ranging from 1 (never) to 5 (always).Each participant completed a socio-demographic form indicating their age, gender, sport and its affiliate club.The ISCCS was administered 2 hours at maximum 1 day after competition.According to Gaudreau and Blondin (2002), the peculiarity of this questionnaire is to be used before, during and after competition.Unfortunately, we were forbidden to ask athletes before and during the competition.Therefore, our study is limited to the after competition period.

Data Analysis
Data descriptive analyses of study groups (mean, standard deviation) are provided in Table 1.
The ISCCS temporal stability We calculate the test-retest index to study the ISCCS temporal stability.The athlete is observed twice times by waiting a certain period of time between two collections (2 weeks).The stability is established by the degree of correlation between the answers provided by the same subject (N = 20), (Shrout & Fleiss, 1979).
Internal consistency was measured by calculating the Cronbach's alpha coefficient between the subscales of the questionnaire.It assesses whether each of the elements reproduced repeatedly and consistently measuring the same construct.
We observed a large number of participants (419) which is higher than the minimum number of 300 suggested by Tabachnick and Fidell (1996).Confirmatory factor analysis (CFA) was treated with AMOS 22.0.0, in order to validate the structure and arrangement of ISCCS factors.
We use several indices of adequacy (Hoyle & Panter, 1995;Kline, 2005) to evaluate the fit models to data collected such as the χ 2 statistic that overcomes the abnormality data (Sattora and Bentler, 1994), the compared fit index CFI (Bentler, 1990) and TLI (Tucker-Lewis Index), the Goodness of Fit Index GFI (Jöreskog & Sörbom, 1984) and the Root Mean Square error of approximation RMSEA (Browne & Cudeck, 1993).
Finally, we conduct a MANOVA analysis to examine the effects of gender and sport (collective and individual sports) on the use of coping strategies.

Descriptive Analyses
The mean scores and the respective standard deviations for each sub-level are presented in Table 1.

The Temporal Stability of the Instrument
The reliability data test-retest is studied by the correlation coefficients between the scores of the ISCCS subscales (see Table 2 and Table 3).

The Internal Consistency, Cronbach Alpha
Measured using Cronbach's alpha, the internal consistency of the Arabic version of ISCCS demonstrated good reliability (Table 4).The coefficients of different coping strategies are from 0.63 to 0.75.For the two dimensions (coping task-oriented and oriented coping emotion) coefficients are 0.81 and 0.64 respectively.Otherwise, the overall coefficient of the questionnaire is 0.78.

Relations between the Subscales of ISCCS
Significant positive correlations (p < 0.05) were observed respectively between mental imagery, the mind control, relaxation, the deployment effort, seeking social support, logical analysis, breakdown of unpleasant emotions and disengagement (between 0.109 and 0.606).The correlation coefficient between isolation and relaxation is 0.180 and between isolation and mental distraction is 0.189.As shown in Table 5, mental distraction, mental imagery, control of thoughts, Effort Expenditure, logical analysis, the breakdown of unpleasant emotions and disengagement are negatively and significantly correlated (p < 0.05).The scores range from −0.219 to −0.100.A negative correlation is observed also between isolation, the breakdown of unpleasant emotions and disengagement between −0.173 and −0.101.
Finally a multivariate analysis, MANOVA was conducted to examine the effects of gender and the kind of sport practiced.The results revealed an important effect of sex [Wilks' lambda = 88; F = 5.54; df = 408; p < 0.001] and the kind of sports practiced [Wilks' lambda = 77; F = 12.73; df = 408; p < 0.001] on the use of coping strategies.

Discussion
For stability over time, we find that the test-retest reliability scores are generally greater or equal than 0.9, indicating a very good repeatability.Usually a high correlation means that the questionnaire remains rather stable over time and that an overall score very close in consecutive reversals.The score ranges between 0.70 and 0.80 for mental imagery and isolation and correlation is rather weak only for mental distraction strategy (r < 0.70).
The coefficients of reliability for mental imagery, control of thoughts, deployment efforts, seeking support, relaxation and isolation are high (0.60 < α < 0.70; Kline, 1998).Also the Cronbach alpha coefficients are reasonable, as they range between 0.70 and 0.80 for logical analysis, the breakdown of unpleasant emotions, mental disengagement and distraction.
There are no big differences between the Arabic and the original form version based on internal consistency.Generally according to DeVellis (1991), alpha values that are The inter-scale correlations are significant as the average correlations vary between 0.20 and 0.40 (Briggs & Cheek, 1986).According to Klein (2005), if the correlation between two factors equal to 1.00 (more than 0.85), the two factors are equal, which implies the omission of one of these two factors.In our study, the inter-scale correlations were in the acceptable range.
For the confirmatory factor analysis, chi-square as well as normed X 2 (X 2 /df) statistics values were important for both models.They are in an acceptable range (X 2 ≤ 1/df ≥ 3); see Wheaton, Muthen, Alwin and Summers (1977) and Tabachnick and Fidell, (2007).
The adjustment indices of the two models (the six factors model and the four factors model): CFI, TLI and GFI are lower than 0.90 (Roussel, Durrieu, Campoy, & El Akremi, 2002).The RMSEA is respectively lower than 0.05 (Brown, 2006) and 0.06 (Hu & Bentler, 1999), for the model with six factors and four factors.
For the exploratory factor analysis, standardized factor loadings can be interpreted as Pearson correlations estimated between an indicator and factor (Kline, 2005).Factor loads of 0.30 and above are commonly used (Brown, 2006).
For the model of coping oriented emotion, we distinguished saturations with 3 factors.Consequently, we estimated two models: the four factors and the three factors models [X 2 = 268.46;p < 0.001; df: 101; X 2 /df = 2, 66; CFI = 0.89; TLI = 0.88; and RMSEA = 0.063].The results, suggest the best match of the original four-factor model that we used in the empirical study.
Finally united various analyses have concluded that mental imagery, control of thoughts, research support, the breakdown of unpleasant emotions, and disengagement vary with the athletes sex.More precisely, we find that most men use mental imagery and seeking support, whereas women use control thoughts, the breakdown of unpleasant emotions, and disengagement.
For the practiced sport, the coping strategies: research support, relaxation, ventilation unpleasant emotions, disengagement and mental distraction vary according to the kind of sport practiced.Formally athletes belonging to individual sports use more of the coping strategies indicated above.
However, our study suffers from some limitations.The first limitation concerns the timing of the administered questionnaire: The ISCCS was administered two hours at maximum one day after the competition (retrospective method).We know that this could be a source of bias, the output of the competition (success or failure) could have influenced the answers.
Also, the factor of stability ISCCS through the different phases of a sporting competition has not been tested (Gaudreau & Blondin, 2002).
Further, we did not ISCCS cross-validation with other scales that invest coping, which affect the choice of the coping strategies used by our participants.

Conclusion
In this research, we propose an Arabic version of the inventory of coping strategies in sports competition.The loyalty of the ISCCS being excellent, the majority of the variables used for the validity and sensitivity to change are acceptable.The results of this study suggest that ISCCS is an evaluation tool in the sports field.Moreover, it is valid for assessing coping task-oriented and the coping emotion-oriented in the collective as well as in the individual sport before, during and after the competition.Psychologists and sports practitioners in the Arab world could use ISCCS to differentiate the coping strategies used by athletes and for preventive interventions.

Figure 1 .
Figure 1.The hypothesized model of 6-factor (TOC) of the Arabic version of ISCCS.

Figure 2 .
Figure 2. The hypothesized model of 6-factor (TOC) of the Arabic version of ISCCS.

Table 1 .
Mean scores and the respective standard deviations for each subscale of the ISCCS.

Table 2 .
The test-retest coefficients for the subscales TOC of the ISCCS.

Table 3 .
The test-retest coefficients for the subscales EOC of the ISCCS.

Table 4 .
Reliability coefficients for the ten factors in the Arabic version and original version of ICSSC.

Table 5 .
Correlations between the subscales of ISCCS.

Table 6 .
Confirmatory factor analysis and models of the ISCCS.

Table 7 .
Standardized solutions for factor loadings for the task oriented coping (TOC).

Table 8 .
Standardized solutions for factor loadings for the emotion oriented coping (EOC).
above 0.60 are considered as acceptable.Cronbach alpha coefficients of the Arabic version are very close to their original version counterparts.