Adaptation Study of the Problem Solving Inventory on the Turkish Athlete Population

The aim of this study was to examine the reliability and validity of Problem solving inventory (PSI) with Turkish athletes. The PSI was designed to clarify utility of problem-solving constructs across sports environment and thus facilitate the development of more comprehensive acknowledges about problem solving and mental health. The subjects were 204 males and 109 females totaling 313 athletes. Participants voluntarily completed the 32 item problem solving inventory. Afterwards, exploratory factor analysis (EFA) and confi rmatory factor analysis (CFA) were undertaken. EFA results yielded a satisfactory three-factor solution, the same as the English version. Cronbach alpha (α) reliability indices for the respective subscales were as follows: problem solving confi dence (PSC) (.69), approach and avoidance (AA) (.73), and personal control (PC) (.68). The PSI explained 66% of the total variances. Moreover, CFA results provided the fi t indices as Chi-Square (X2)=912.8, df=461, X2/df=1.98, Root Mean Square Error Approximation (RMSEA)=.07, Comparative Fit Index (CFI)=.93, Goodness of Fit Index (GFI)=.92, Normed Fit Index (NFI)=.90, Non Normed Fit Index (NFI)=.89 with 32 items and 3 sub-dimensions. The fi t indices of the PSI in relation to EFA and CFA were at an acceptable level. The original 3-factor solution was supported by the Turkish athlete subjects. Results of the study introduced that the Turkish version of the Problem Solving Inventory is a valid and reliable measurement for Turkish athletes.


Introduction
Psychological skill set of athletes and mental preparedness are major decisive contributor of the performance in sport setting (Mahoney, Gabriel, & Perkins, 1987;Gee, 2010).Athletes' psychological abilities such as concentration, emotional control, motivational orientation, coping with stress and anxiety are essential dimensions for peaking the performance (Crust, 2007;Smith, Smoll, & Cumming, 2007).Harmison (2011) indicated that developing the necessary psychological skills is not solely important factor for peaking athletic performance, but adversity-coping skills are also decisive factor on athletic performance.
Acceptance, positive reinterpretation and problem solving are emphasized as adversity coping strategies in sport setting (Galli & Vealey, 2008).Although each of those strategies are important for athletic development and better performance, problem solving skills has unique construct due to the dealing with the stressor directly.Heppner and Krauskopf (1987) defi ned the problem solving as "the complex interplay of cognitive, aff ective, and behavioral processes for the purpose of adapting internal or external demands or challenges." In addition to problem-solving appraisal, researchers have used a number of strategies and measurements to assess applied problem solving (Heppner & Wang, 2003).First one has been to assess the nature and frequency of personal problems; the basic assumption has been that fewer problems suggest more eff ective problem solving.One of the earliest measures was the 330-item Mooney Problem Checklist (Mooney & Gordon, 1950); a second one has been to assess what is assumed to be problem-solving ability by providing a task that requires cognitive processes related to applied problem solving.Th e Means-End Problem Solving Procedure (MEPS) (Platt & Spivack, 1972) assesses one's cognitive ability to plan, step-by-step, the means of moving toward an eff ective solution of interpersonal problems.A third one has been to assess problem-solving attitudes and skills through self-report assessments, such as the Social Problem Solving Inventory (SPSI) (D'Zurilla & Nezu, 1980).A fourth strategy has assessed verbal reports of one's problem-solving activities that facilitate or inhibit progress toward resolving problems, such as the Problem Focused Style of Coping (PF-SOC) (Heppner, Cook, Wright, & Johnson, 1995).Th e multiple assessment strategies refl ect the multifaceted nature of applied problem solving.Only the PSI, however, is conceptualized as a measure of problem-solving appraisal and has been recognized as one of the most widely used self-report inventories in applied problem solving (A.M.Nezu, C.M. Nezu, & Perri, 1989).
In sport context, implementing executive functions requires problem solving skills (Diamond, 2006).Moreover, problem solving skills has a major role on actual motor performance, besides processing information and making decision, (Ripoll, Kerlirzin, Stein, & Reine, 1995;Hristovski, 2012).Participating regular sport activities may lead to better problem solving skills.To illustrate, Jacobsen and Matthaeus (2014) examined the executive functions by administering tests of problem solving, decision making and inhibition among athletes and non-athletes.Th eir results indicated that athletes demonstrated higher problem solving scores compared to non-athletes.Similarly, Senduran and Amman (2015) demonstrated that high-school aged athletes feel more confi dent when they encounter the problem comparing to non-athletes.Furthermore, this study also indicated that the type of the sport was also major factor eff ecting the problem solving skills of athletes.Th us, investigating the problem solving approaches of athletes become an important topic in order to explicate how to improve athletic performance.
Although evaluating psychomotor approaches in problem solving strategies are important (Ripoll et al., 1995), assessing the individual diff erences in problem solving is another necessary aspect in sport context.Heppner and Petersen (1982) developed to an instrument to explicate personal problem solving approaches of individuals.Problem Solving Inventory aims to diff er individuals with eff ective problem solving skills such as confi dence, personal control, avoiding problems from those who perceive themselves as ineff ective problem solvers (N.Sahin, N.H. Sahin, & Heppner, 1993).Th e fi rst adaptation of PSI into Turkish language was completed among university students (Sahin et al., 1993), and results of primary-component analyses revealed a six factor structure, namely thinking approach, avoiding approach, estimator approach, self-trust approach and planned approach, unlike to three factor structure revealed by Heppner and Petersen (1982).All Cronbach's Alpha values of sub-dimensions were larger than 0.65 in Turkish version of the PSI.Although Turkish version of PSI was implemented among students from the School of Physical Education and Sports (Sözen, 2012), high school athletes (Senduran & Amman, 2015) and sport referees (Karaçam & Pulur, 2016) and found reliable, none of the study analyzed the validity of the instrument among these populations.Because the PSI was not developed specifi cally for the sport context, the validity of the model of the Turkish version that used in sport studies might be a doubtful, especially for studies focused on athletes.
Th erefore, the purpose of this study was to investigate the reliability and validity of the Turkish version of the PSI among Turkish athletes in order to propose a valid model to measure problem solving skills.

Ethical administrative procedures
Prior to data collection, approval was obtained from the Ethics Committee of Nevşehir Hacı Bektaş Veli University.Th e author of the original scale was contacted to ask for permission which he kindly granted.During the questionnaire administration, each participant was briefed with regards to what is involved in the study and how their responses would be kept confi dential.Aft er participant briefi ng, athletes were informed that participation was voluntary and that they could withdraw at any time during the data collection process.Aft er this, data was only collected from athletes who gave consent and volunteered to participate.
Athletes who respond the PSI are provided with a likert scale and composed of 35 items, including 3 fi ller items and evaluated according to scoring system with the numbers of 1-6 (1=strongly agree to 6=strongly disagree).Th e lowest score that can be obtained from the inventory is 32 and the highest score is 192.Higher scores obtained from the scale indicate self-inability perception of individual about problem solving abilities and lower scores show that the one perceives himself/ herself as adequate.9 th , 22 th and 29 th items in scoring order are out of scoring in accordance with protocol.Moreover, there were inverse-scored items include; 1 th 2 th , 3 th , 4 th , 11 st , 14 th , 15 th , 17 th , 21 th , 25 th , 26 th , 30 th and 34 th items.

Procedure
Th e Turkish version of the PSI was only completed by Turkish athletes who voluntarily participated and signed consent forms.Th e translation procedure of the PSI was made by faculty members from English Language Department of Nevşehir Hacı Bektaş Veli University.Th ey forward-translated the English version into Turkish, and the translated versions of the PSI were compared for deviations.Th en another bilingual language expert back-translated the Turkish version into English to examine the retention of the meaning of the items.Th ree academics who are experts in the areas of physical education & sports sciences, sports psychology and psychometrics, reviewed the content of the preliminary PSI version to ensure that the questions were culturally appropriate to the Turkish athlete population.

Data analysis
Results of EFA & CFA analyses are presented in the following section.To determine the factor structure of the multidimensional questionnaire, exploratory factor analysis that conducted by Statistical Package for the Social Sciences (SPSS) 21.0 was fi rstly used.Following this, confi rmatory factor analysis was conducted by analysis moments of structures (AMOS) 18 to check statistical model.Comparative fi t index (CFI>0.90,acceptable), non-normed fi t index (NNFI>0.90acceptable), normed fi t index (NFI>0.90)and root mean square error of approximation (RMSEA<0.08,adequate model fi t) (Maruyama, 1998) were used to check the data.Internal con-sistency of the adapted scale was checked by computation of Cronbach's coeffi cient alpha.Fit indices scores were used in order to prove model fi ts Öcal, 2011.Th e measurement model based on the fi t indices was evaluated for construct validity.

Results
Descriptive statistics and Cronbach's alpha coeffi cients for the PSI scales are shown in Table 1 reasonable factorability.Secondly, the Kaiser-Meyer-Olkin measure of sampling adequacy was .70,above the commonly recommended value of .6, and Bartlett's test of sphericity was signifi cant (χ2 (397)=2011.09,p<.05).Given these overall indicators, factor analysis was deemed to be suitable for all 35 items.
For the fi nal stage, a principal components factor analysis of the PSI items, using varimax and oblimin rotations, was conducted, with three factors explaining 66% of the variance.An oblimin rotation provided the best-defi ned factor structure.Th e factor loading matrix for this fi nal solution is presented in Table 2. Th e factor labels proposed by Heppner and Pettersen (1992) suited the extracted factors.Internal consistency for each of the scales was examined using Cronbach's alpha.Th e alpha levels were moderate: problem solving confi dence (PSC) (11 items) (.69), approach and avoidance (AA) (16 items) (.73), and personal control (PC) (5 items) (.68).Th e PSI explained 66% of the total variances.
Aft er EFA, CFA was used to test the factor structure that shows the sub-dimensions of Problem solving inventory over the data gathered from Turkish athletes.Firstly, for a model with 3 factors (problem solving confi dence, approach and avoidance, and personal control) set in the original sub-dimension, goodness of fi t statistics were calculated.Th e results

Discussion
Th e goal of the study was to evaluate the psychometric properties of the Problem Solving Inventory (Heppner & Petersen, 1992) on a Turkish population.Exploratory and confi rmatory factor analysis results supported the initial structure of the questionnaire for the overall model.
A crucial personal resource for dealing with psychological health, stressors, and barriers etc. in sports is problem solving abilities.Th e competent problem solvers are fl exible, adaptable people and they are able to create or fi nd proper strategies to solve problems.It is to be expected problem solving abilities are very relevant for all sports stuff (athletes, referees, managers, sport psychologists, coaches etc.).
A number of measurement tools exist in the literature that has been developed to evaluate problem solving abilities in sports.As explained earlier, such measurements tools (e.g. the Mooney Problem Checklist, the MEPS, the SPSI and the PF-SOC) have been trialed in diff erent environments.
Th e PSI has been found to have acceptable internal consistency estimates across a number of populations and cultures (e.g.Heppner, 1988;Heppner, Pretorius, Wei, Lee, & Wang, 2002).Th e inter-correlation among these three factors ranged from .39 to .69 across a range of studies on PSI (Heppner, 1988).Results suggest that the factors are not only interrelated but also independent enough to be considered as separate factors (Heppner & Wang, 2003).Consistent with our fi ndings, subsequent studies using either exploratory factor analyses indicate that the PSI factors tend to replicate well across diff erent age groups from various backgrounds, such as mid-western White college students (Cronbach's Alpha reliability values from .72 to .90)(Heppner, Baumgardner, & Jackson, 1985); French Canadian adults (Laporte, Sabourin, & Wright, 1988), Turkish college students (Cronbach's Alpha reliability values from .69 to .78 and total internal consistency was .88)(Sahin et al., 1993); Black South African college students (Cronbach's Alpha reliability values from .71 to .84 and total internal consistency was .89)(Heppner et al., 2002); Egyptian college students (Cronbach's Alpha reliability values from .76 to .88 and total internal consistency was .75)(Soliman, 2014); Romanian adults (Cronbach's Alpha reliability values from .78 to .84)(Marian & Roşeanu, 2012), and Mexican American high school students (Cronbach's Alpha reliability values from .66 to .77 and total internal consistency was .86)(Huang, 2005).Moreover, in another Mexican American high school students Cronbach's Alpha reliability values were as follows: .86 for PSI total, .77for PSC, .76 for AAS, and .66 for PC (Huang & Flores, 2011).Summing across studies, the PSI total obtains average alpha coeffi cients around .80s, whereas two of the factors (PSC and AA) obtain average alpha coeffi cients in the low to mid .80s,and the third factor (PC) obtains average alpha coeffi cients in the low .70s.Th ese results suggest that the PSI is internally consistent across diff erent forms of the PSI used across diff erent cultural groups (Heppner & Wang, 2003).
Additionally, to results of exploratory factor analysis, the confi rmatory factor analysis results' indicated that the factor structure of the PSI supported the three factors of the PSI and a general problem-solving factor with Turkish athlete population.Th at is consistent with some other scientifi c studies those aimed to fi nd factor structure model of the PSI with diff erent samples would support the use of the three factors of the PSI include; Heppner et al. (2002) study on South African college students, Huang and Flores' (2011) on Mexican American high school students, Marian and Roşeanu's (2011) on Romanian adults, and Soliman's (2014) on Egyptian college students.Moreover, this study provides additional support for the generalizability of the PSI factor structure to a Turkish athletes sample.All these results suggest considerable consistency of the PSI factor structure across various cultures and diff erent samples.In summary, similar to the current fi ndings, other validity and reliability studies demonstrated that PSI is a valid data collection tool in diff erent cultures.
Th e results of the present study introduced that Problem Solving Inventory-Turkish version has relatively strong psychometric properties, and is a valid and reliable test instrument to evaluate mental skills levels in Turkish sports contexts (athletes, university students, etc.).
Th e present study has some limitations.Th e sample of the study were formed by Turkish amateur athletes from a number of team and individual sports.Additionally, professional level athletes should be evaluated for applicability for use in a high level sport environment.Th us, continued evaluation of the PSI is necessary.Moreover, additional types of invariance testing (e.g., temporal, cross-cultural, and sport type), as well as other ongoing construct validity evaluation, needs to be considered in future research to gather new evidence on problem solving abilities.

Table 1 .
Descriptive statistics for the three Problem Solving Inventory factors (N=313)

Table 2 .
Factor loadings based on a principal components analysis with oblimin rotation for 32 items from the PSI (N = 313) Legend: Problem solving confi dence (PSC), Approach and avoidance (AA), and Personal control (PC) PSYCHOMETRIC PROPERTIES OF PROBLEM SOLVING INVENTORY | B. O. MICOOGULLARI & R. EKMEKCI