2014_JSS A Model Linking Sources of Stress to Approach and Avoidance Coping Styles of Turkish Basketball Referees

8/12/2019 2014_JSS A Model Linking Sources of Stress to Approach and Avoidance Coping Styles of Turkish Basketball Referees

1/34

Anshel, M. H., Sutarso, T., Ekmekci, R., & Saraswati, I. W. Journal of Sports Sciences2013

The final version of this paper was published in: Anshel, M. H., Sutarso, T., Ekmekci, R., &

Saraswati, I. W. (2014). A model linking sources of stress to approach and avoidance coping

styles of Turkish basketball referees.Journal of Sports Sciences 32 (2), 116-128.

A Model Linking Sources of Stress to Approach and AvoidanceCoping Styles

Of Turkish Basketball Referees

Mark Howard Anshel1, Toto Sutarso2, Ridvan Ekmekci3, & Intan W. Saraswati4

Abstract

Purpose of this study was to externally validate and test a conceptual transient model

involving six paths that linked sources of acute stress to avoidance and approach coping styles

among Turkish basketball referees. The sample consisted of 125 Turkish basketball referees

ranging in age from 18 to 36 years (M= 25.58. SD= 3.69). The path analysis tested the

relationshipssimultaneously from stressors, in consecutive order, distractions, subpar

performance, and verbal abuse, to coping styles, first both avoidance-cognitiveand approach-

cognitive, and then approach-behaviour. Results of path analysis indicated that the model

achieved a good fit and that all paths tested simultaneously were significant. The distractions

stressor was positively related to subpar performance, which, in turn, was positively related to

verbal abuse. Verbal abusewas negatively associated with an avoidance-cognitivecoping style,

and positively related to the approach-cognitivecoping style. The results also supported a

crossover effect of both avoidance-cognitive and approach-cognitive on approach-behaviour.

One implication of this study is that coping should be studied in naturally occurring stages, a

process-oriented approach. Another implication is that approach and avoidance coping styles,

each sub-divided into cognitive and behavioural categories, provide a meaningful framework

which provides sports officials a coherent structure for learning and improving ways to cope with

acute stress experienced during the contest.

Keywords: stress, sources of stress, coping, coping style, path analysis, sports officials


2/34


A Model Linking Sources of Stress to Approach and AvoidanceCoping Styles

Of Turkish Basketball Referees

The classical and oft quoted definition of coping is constantly changing cognitive and

behavioural efforts to manage specific external and/or internal demands that are appraised as

taxing or exceeding the resources of the person (Lazarus & Folkman, 1984, p. 141). The

psychology literature is replete with empirical research on understanding (i.e., describing,

explaining, predicting) the coping process, particularly on identifying the mechanisms and

mediating factors that influence coping effectiveness (e.g., Nicholls & Polman, 2007). While the

predominant number of these studies has included individuals from various areas of human

performance, one group that experiences high intensity stress, but has received relatively rare

attention by researchers, is sports arbiters/officials (i.e., referees, umpires, judges).

Stress is inherent in sports officiating due to the perception of athletes, coaches, and

spectators that they (i.e., the official) commit errors and injustice, usually by applying a rule

inaccurately or unfairly, or by making a wrong call (Mark, Bryant, & Lehman,1983). Mark et

al. contend that acute stressors, such as judgment calls and decisions about rule infractions, are

predominant, inevitable, and likely to lead to poor performance by both officials and competitors

if officials fail to cope effectively with stressful events. The result of poor, ineffective coping is

slower information-processing, less accurate decision-making, improper performance mechanics,

burnout, and eventually, quitting (Anshel, 2012). Attempts to understand the coping process of

sports officials is likely to improve officials coping efforts, reduce their stress, improve health

and well-being, foster effective performance, and greatly reduce the rate at which officials drop

out of sport. Examining the relationship between sources of stress and coping styles has been

particularly neglected. Thus, examining the factors that contribute to perceived stress and to

improve our understanding of the coping process among sports officials appears warranted.

Previous studies of sports officials have addressed acute stressors and coping styles in separate


3/34


studies, but rarely as a sequential interaction, orrelationship, between both constructs in the same study.

The paucity of research that jointly examines the association between sources of stress and the persons

selection of a coping style is surprising, partly because sources and intensity level of acute stress

mediates coping (Nichols & Polman, 2007). The study of both conceptual frameworks in the same study

is called aprocess-oriented approach (Gaudreau & Miranda, 2010), or the transient nature of coping

(Nicholls & Thelwell, 2010).

The coping process has also been examined by categorizing responses to stress as a function of

coping style, that is, the interaction between the personal factors (e.g., coping style, cognitive appraisal,

perceived stress intensity) and situational characteristics (e.g., type of stressor), a process that describes

transactional theory (Porter & Stone, 1996). With respect to the former, Anshel and Sutarso (2007)

contend that linking categories of sources of stressors with the athletes coping style...has two distinct

advantages in understanding the coping process and creating effective interventions. First, most single

items of behaviour...have a high component of measurement error and a narrow range of generality.

Coping responses are more predictable. The second advantage is improved generalization of results

(which improves predictability of) the athletes responses following a set of stressors with similar

characteristics (p. 19).

The inclusion and relevance of environmental and situational factors that explain the coping

process and the social context in which coping occurs is intrinsic to transactional theory. This theory has

received extensive attention in the non-sport literature (Kosciulek, 2007). Situational factors such as the

type and intensity of a stressor are a critical factor in coping strategies and coping effectiveness.

Wethington and Kessler (1991) divided situational determinants into two types, the situational context

surround the individual who is coping (e.g., the social resources that are available to the individual,

characteristics of the individual), and the type of event that occurs (e.g., sources of stress, acute versus

chronic forms of stress). As Kosciulet has concluded, the type and severity of situational demands are

particularly relevant for evaluating the efficacy of different coping strategies (p. 78). Additional


4/34


research is needed to examine the extent to which the coping process can be explained and predicted by

recognizing the role of situational characteristics and coping consistency among sports officials.

Over the years, the coping process has been categorized and examined with respect to various

frameworks (see Nicholls & Thelwell, 2010, for a review). One coping framework that has received

extensive attention in the coping research literature over the years in both general psychology (e.g.,

Krohne, 1993, 1996; Roth & Cohen, 1986) and performance/sport psychology (e.g., Anshel, Williams,

& Williams, 2000; Krohne & Hindel, 1988) is approachand avoidancecoping. An approach coping

style in sport,also referred to as vigilant, attention, active, sensitization, engagement, and augmentation

(Krohne, 1993; Roth & Cohen, 1986), generally refers to the individuals behavioural(i.e., taking

action) and cognitiveefforts (i.e., thoughts, emotions) in dealing with stressful situations for the intended

purpose of reducing stress intensity and maintaining optimal performance (Krohne, 1996). Examples of

approach-behaviouralcoping in sports officiating include calling a penalty, defending a call (to a

player, coach, or colleague), overcoming assertive actions by coaches and athletes, and seeking support

from a colleague/partner/supervisor (Kaissidis & Anshel, 1993). Examples of approach-cognitive

activity include remembering a particular rule, self-statements that affirm a specific call or perspective,

or analysing the stressful event. Krohne (1993) contends that vigilant coping, a term used

interchangeably with approach, also consists of anticipation of negative events, information search,

or recalling negative events (p. 34).

The other category of coping is called avoidance, also called nonvigilant, passive,

desensitization, repression, disengagement, and avoidant (Krohne, 1993, 1996; Hoar, Kowalski,

Gaudreau, & Crocker, 2006; Krohne, 1993, 1996). An avoidance coping style reflects a persons efforts

to reduce or eliminate cues perceived as threatening or harmful (Krohne, 1993, 1996).

Similar to approach coping, avoidance has also been sub-categorized as behavioural or cognitive,

such as persons whophysically remove themselves from a threatening environment (Krohne et

al., 2000). In the parlance of Krohne (1993), avoidance-cognitivecoping reflects turning away


5/34


from threatening cues (p. 3). Examples includes discounting (i.e., reducing the importance or

changing the interpretation of a stressful experience), psychological distancing, which consists of

understanding the reasons of a stressful experience, or feeling detached from the source of the

stressor, thereby keeping it in perspective (Bramson, 1981), and using a distraction activity that

prevents rehearsal of the stressor (e.g. attentional diversion; Krohne, 1993).

The proper selection and application of coping styles and strategies within those styles in

sport is often a function of personal and situational/environmental demands. While athletes and

coaches, for example, might use approach coping under certain conditions, and apply avoidance

coping under other conditions, sports officials must maintain situational control and, therefore,

are less likely to maintain an avoidance posture.Sports officials, for instance, typically

experience more verbal abuse than athletes and coaches. This often leads to less avoidance-

cognitive coping, and to more approach-cognitive coping, at least partially due to task demands

and responsibilities of game officials. Unlike the coping styles of skilled competitive athletes and

coaches, Anshel and Weinberg (1995, 1996) found that sports officials do not tend to use an

avoidance-behaviour coping style. This is because the primary role of sports officials is to

control the contest, and to ensure that rules and participant behaviour are enforced to ensure

fairness (Rainey, 1995). Ultimately, the sports official is held accountable for maintaining the

contests integrity. Not surprisingly, approach-behaviour coping, not avoidance-behaviour, is far

more common.

Understanding the coping process in particular areas of performance (e.g., sport, injury

and rehabilitation, the performing arts, various forms of physical activity) is especially important

in situations when the performer is required to rapidly take in, process, and respond efficiently to

an array of internal and external processes (Wrisberg, 2001). Wrisbergs review of related

literature supports the contention that elite performers selected, process, and retrieve information

more effectively and efficiently than their less-skilled counterparts.


6/34


Coping effectiveness can be similarly described as a function of skill level across

performance domains (Auerbach, 1992). Auerbach contends that stressors experienced

sequentially across several temporal stages form important determinants of subsequent appraisal

and coping processes. It may be surmised, therefore, that coping is a transitional process rather

than a series of separate, interdependent components. It is the nature of coping that has been

virtually neglected by previous researchers and serves as the focus of the present study.

Theoretical Framework and Hypotheses

It is widely known in the sport psychology literature that coping is a process consisting of

different models that depict various components and sequential stages (Nicholls & Thelwell, 2010).

Unknown, however, are the relationships between the early, antecedent stages of coping in relation to

subsequent components, specifically, the parallel paths that lead from sources of stressors to the use of

coping styles (Jones & Uphill, 2004). Experiencing stressful events during the sport contest is

continuous. While researchers have targeted the relationships between stressful events and the ways in

which sports competitors (e.g., athletes, coaches, contest officials) cope with these events, relatively

little is known about the temporal patterning of these relationships.

In their review of research on temporal patterning of emotion in sport, Cerin, Szabo, Hunt, and

Williams (2000) have called for more research into the temporal patterning of other emotions, including

stress. One area that has received relatively little attention by researchers is the temporal patterning

between stressful events in sport and the participants coping style, particularly among sports officials.

Jones and Uphill concluded in their review of related literature that assessing the emotion-performance

relationship experienced during sport competition will produce more valid data due to the transient

nature of emotions. One outcome of our improved understanding of this relationship, they claim, is the

development of strategies that promote emotional control.

While sport psychology researchers have examined the temporal patterns of emotions

experienced prior to, during, and immediately after competition, little is known about the transient


7/34


nature of the stress and coping relationship among sports officials (i.e., referees, judges, umpires).

Gaudreau and Miranda (2010) distinguish between stability and consistency in the coping process.

Stability in coping refers to the extent to which individuals are coping in a similar manner across time

within the same situation or context (p. 17). How a sports official copes with game-related stress across

a series of games or over the season provides one example of coping stability. Consistency, however,

represents the extent to which individuals use similar coping strategies or styles under different

conditions or within the same type of situation. Gaudreau and Miranda propose that these situations can

be divided in naturally occurring stages with distinct demands likely to generate changes in coping, what

they call a process-oriented approach. One area of study that has addressed mediating factors that

influence the coping process is the interaction between a persons cognitive appraisal and the selection

and effectiveness of coping strategies.

One relatively recent study (Anshel & Sutarso, 2007) examined the relationships between

sources of stress and coping style among 332 competitive athletes as a function of gender. Athletes were

asked to indicate their perceived stress intensity for common sources of acute stress and their subsequent

use of coping skills with the two most intense stressors. Theory-driven categories of acute stress sources

were labelledperformance-relatedand coach-related. Coping styles were labelled approach-behavioral

approach-cognitive, and avoidance-cognitive. Findings indicated that coping style three-factor model

showed a good fit with the data, that is, the type of acute stressor was associated with the athletes

respective coping style. A similar approach, using the coping transient model as a framework, was

generated and tested in this study of examining the relationships between sources of stress and coping

style among sports officials, a group in which the stress and coping process is consistent and continuous.

Thus, the purpose of this study was to examine the coping process using a process-oriented approach to

determine the relationships between sources of acute stress and coping styles among basketball referees.

The Coping Transient Model


8/34


Benson and Hagtvet (1996) contend that in the measurement of constructs...a construct

represents an abstract variable derived from theory or observation (and) represents a hypothesis

about the observables thought to represent the construct. Therefore, it is important not only to be

able to define and develop measurements of contrasts, but also to find and demonstrate

functional relations between measures of different constructs (p. 84). They endorse a three-step

approach to construct validation: (1) specify the domain of observables for the construct, (2)

determine to what extent the observables are correlated with each other, and (3) determine

whether or not the measures of a given construct correlate in expected ways with measures of

other constructs. Thus, identifying the relationships between constructs in the coping process

among sports arbiters first requires determining their sources of stress.

Along these lines, Dorsch and Paskevich (2007) concluded from their review of related

literature that two stress-related factors that appeared across all studies were fear of physical

harm (e.g., assault by a player/coach/spectator) and fear of failure/appearing incompetent

(e.g., making a bad call). In a subsequent similar study, Voight (2009) found the top sources of

stress experienced by soccer officials included making a controversial call and conflict

between officiating and work demands. Thus, it appears that the coping process consists of

three primary phases: (1) cognitive engagement that is both active and intense, consisting of

selected thoughts and emotions (e.g., fear and anxiety) that distractthe official from the task at

hand (i.e., distractions), (2) undesirable performance quality (i.e., subpar performance), and (3)

highly emotional and negative reactions by those affected by the referees decisions and actions,

including players, coaches, and spectators (i.e., verbal abuse).

As shown in Figure 1, the coping transient model is based ona logical sequence of events

in sports officiating, in general, and with basketball referees, in particular, moving from

experiencing the acute stressor to the use of a coping style. While we acknowledge that cognitive

appraisal is a mediator of coping (Krohne, 1996), the model tests the relationship between the


9/34


type of stressful event experienced during the sports contest and the basketball referees coping

style.

There were three stressors, or components, in the model as posited from previous

psychometric validation in the sports coping literature (e.g., Anshel & Sutarso, 2007; Rainey,

1995) and from the extant sport psychology literature (Anshel & Weinberg, 1995, 1996). These

sources of stress reflect a category of acute stressors called performance-related, as confirmed

in Anshel and Sutarsos study of male and female athletes.

The first segment, labelled distractions, consists of items that reduce the referees

effectiveness due to intense engagement with selected thoughts and emotions related to pain, fear

and anxiety that may distract the official from the task at hand. Studies by Rainey (1995), Anshel

and Weinberg (1995), Voight (2009), and Dorsch and Paskevich (2007) on baseball umpires,

basketball referees, soccer referees, and ice hockey referees, respectively, indicated that

distractions about physical harm, fear of failure, interpersonal conflict, and making a mistake

formed prevalent sources of stress. As Wrisberg (2001) reports from his review of related

empirical literature, and Weinberg and Richardson (1990) describe in their applied work,

distractions among sports officials results in the lack of performance consistency, or subpar

performance. While threats and abuse from others has been a ubiquitous theme in previous

studies examining sources of stress among sports officials (e.g., Anshel & Weinberg, 1996;

Kaissidas, Anshel, & Porter, 1997), researchers have not examined the relationships between

disruptive cognitive processes and performance outcome, at least among sports officials. Thus, it

is apparent that subpar performance is a manifestation of cognitive distractions, which formed

the second source of stress in the current study.

Based on the extant sources of stress research in sports officiating literature (reviewed

earlier), and Weinberg and Richardsons (1990) descriptive narrative of the physical and

psychological qualities and demands of highly skilled sports officials, verbal abuse is typically


10/34


predicated on and is frequent reaction to, the referees actions which are perceived by critics as

inaccurate or erroneous (Kaissidis-Rodafinos & Anshel, 2000). It is less plausible to speculate,

we contend, that verbal abuse is an antecedent or predictor of poor performance. This conclusion

is supported by theoretical and empirical research on sport spectatorship (Wann, 1997).

In his review of related research, Wann (1997) contends that sport spectator aggression is

a function of several factors, including strong team identity, group affiliation (i.e., the fans need

for belongingness), and spectator self-esteem (i.e., home team success translates into

strengthened feelings of achievement and accomplishment) that may lead to making harsh

judgements of sports officials. This is likely because unfair or poor sports official

performance justifies the spectators highly critical, even aggressive verbal or physical reaction

leading to heightened stress and anxiety impair the officials quality of cognitive functioning,

such as attentional focusing or speed and accuracy of decision-making (Wrisberg, 2001). Thus,

in the present model, referees subpar performance is typically followed by harassment or verbal

abuse.

In summary, the stressors were labelled distractionsbecause they inhibited the referees

ability to efficiently process information and make rapid and accurate decisions, subpar

performancebecause external and internal distractions are known contributors to inefficient

information processing and reduced physical performance, and (3) verbal abuse by others(e.g.,

players, coaches, spectators). The model then posits three coping styles, described earlier, in

response to the stressors simultaneous paths. These are avoidance and approach coping in both

cognitive and behavioural forms, a framework consistent with Anshel and Sutarsos (2007) study

which validated the relationship between sources of stress and coping styles in sport.

The theoretical justification for the coping process prediction of avoidance cognitive

leading to approach behaviour is that, as Krohne (1993) explains, cognitive coping processes

(i.e., avoidance-cognitive) precede coping behaviour (i.e., approach-behaviour). That is, coping


11/34


with acute stress in sports officiating is initiated with the cognitive appraisal of the stressor -

thinking - followed by an action. Krohne (1993) contends that cognitiveforms of avoidance and

approach coping precede behaviouralforms of approach coping.

Coping styles that are compatible with referees performance include approach-cognitive,

in which the referee actively processes information in order to resolve the stressful situation and

maintains concentration on the task at hand (Krohne, 1993; Krohne et al., 2000), avoidance-

cognitive, in which the referee feels the correct decision was executed, perceives extraneous input

as a distraction, and uses discounting or psychologically distancing as ways to reduce the

stressors intensity, and approach-behaviour, in which the official maintains control of the

situation, such as giving a warning or technical foul or explaining/defending a call.

Examining a logical sequence of events in the coping process is important for researchers,

physical and mental health professionals, medical practitioners, and performers. Identifying

components of the coping process from categories of stressful events to responses to those events

(i.e., coping styles) provides insights in the sequence where the coping process becomes

problematic (e.g., experiencing a stressor is coupled with poor or ineffective coping that results in

unsatisfactory performance). Specific cognitive-behavioural interventions are tailored for

individual needs, a central feature of the matching hypothesis.

The matching hypothesis, as explained by Jones and Hardy (1989), posits that stress and

anxiety clog up information processing channels with irrelevant information, and that an

intervention should match the fundamental causes(s) of the persons cognitive or somatic

symptoms of state anxiety, or any other unfavourable cognition or affect. The implication in the

present study is that to achieve optimal coping effectiveness it is important to match specific

situational characteristics, or types of stressors, with particular coping responses, in our study,

using the approach-avoidance framework. As Anshel and Sutarso (2007) explain in their study of

competitive athletes, learning to respond effectively to a class of stressors, as opposed to


12/34


learning unique coping skills following single stressful events, would reduce the information load

required for storing and retrieving the proper use of specific coping strategies from memory (p.

5).

Thus, the following hypotheses (see Figure 1) were tested in this study based on the transient

model of experiencing, in sequence, selected stressors, followed by selected coping styles: (1) the

stressor, distractions, will be positively related to the stressor subpar performance; (2) subpar

performancewill be related to verbal abuse; (3) the referees will use less avoidance-cognitive coping

style, reflected by a negative relationship between verbal abuseand avoidance-cognitive coping; (4) the

referees will use more approach-cognitivecoping style upon receiving verbal abuse, reflected by the

positive relationship between these processes; (5) anavoidance-cognitivecoping style will be positively

associated with an approach-behaviourcoping style; and (6) there will be a positive relationship

between approach-cognitivecoping style and approach-behaviourcoping style.

Method

Participants and Procedures

Participants in this study, recruited through a basketball referees association in Turkey,

consisted of 125 basketball referees (108 males and 17 females), 64 of whom officiated at the local

level, and 61 officiated at the college level, all considered highly skilled based on their membership in

the referees association. They ranged in age from 18 to 36 yrs (M= 25.58 yrs. SD= 3.69), and whose

experience ranged from one to 12 years (M= 5.14 yrs, SD= 2.99). Each participant consented to

voluntarily engage in the study under the condition that they may withdraw at any time without negative

consequences. The university Institutional Review Board approved this study.

Materials

An inventory, called theBasketball Officials Stress Inventory(BOSI), was developed for this

study of Turkish basketball referees and consisted of two parts. The first part listed statements about the

referees sources of stress(RSS) of selected stressful events which were adapted from previous studies


13/34


in this area (e.g.,Anshel & Weinberg, 1995; Kaissidis & Anshel, 1993; Louvet, Gaudreau, Menaut,

Genty, & Deneuve, 2009; Tsorbatzoudis, Kaissidis, Partemian, & Grouios, 2005). Examples included

verbal abuse from coaches/players/spectators, making a wrong call, being in a wrong position on the

court, experiencing pain or injury, and problems working with my partner. The second part of the

inventory was to identify the referees coping styles(RCS) following three different stressors on

stressful events. Respondents were asked to indicate the level of intensity felt after experiencing each o

the stressful events on a Likert-type scale ranging from 1 (very low) to 5 (very high). The BOSI differed

from an earlier validated referee stress and coping inventory, called theBasketball OfficialsSources of

Stress Inventory(BOSSI; Kaissidis & Anshel, 1993) primarily in three ways. First, several BOSI items

were adapted from the BOSSI and not taken verbatim in order to be consistent with the Turkish

language. Second, one source of stress, sexual harassment, was added in the present study but not

included in the earlier BOSSI instrument. Third, the current study combined sources of acute stress and

coping strategies in the same inventory to identify their relationships. Coping style, however, was not

included in the BOSSI.

The BOSI consisted of nine items that were theoretically generatedfrom the existing stress

literature to examine coping with acute stress with athletes and modified specifically for basketball

referees and for sports officials by Kaissidis and his colleagues (1997, 1998), and Rainey (1995). This

process reflects an attempt to reconceptualise the sources of stress construct to make it more

compatible with a referees subsequent use of coping style. Items were conceptually categorized as:

(1) distractions, (2) subpar performance, and (3) verbal abuse.

The RCS consisted of ascertaining the referees coping style (RCS), each sub-categorized as

behavioural or cognitive coping responses to stressful events typically experienced during basketball

games. The nine RCS items used in this study were obtained and modified from several previous

studies using this framework in which all items were validated (e.g., Anshel, & Sutarso, 2007; Anshel,

Sutarso, & Jubenville, 2009; Krohne, 1993, 1996; Krohne et al., 2000). The coping style model


14/34


consisted of three factors: (1) approach-cognitive(e.g., I listened to the source of stress, I thought

that the situation was just part of the game, I kept my concentration on the game and focused on the

next task at hand), (2) avoidance-cognitive (e.g., I felt upset and wanted to quit, I kept thinking

about the situation, I felt the situation was unfair to me, and developed negative feelings), and (3)

approach-behaviour(e.g., I gave a warning/technical foul, I argued my point, I verbally

defended myself, I verbally responded to the situation. Results and Table 1 include the

psychometric properties of this inventory.

Data Analysis

Two primary statistical techniques were applied in this study to test the transient

(process-oriented) model. First, confirmatory factor analysis (CFA) was used to confirm the

factor structures and the items derived from the conceptual frameworks for refereessources of

stress(RSS) and the referees coping styles(RCS) as discussed earlier. Cronbachs alpha and

composite reliability for each factor of RSS and RCS were used to report the reliabilities of each

factor.Second, path analysis was used to determine the paths/relationships among factors of the

referees RSS and RCS.

Path analysis assumptions. The assumptions of path analysis were met in the following

ways. First, all relations are linear and additive; the sequence paths are shown in the path

diagram. Specifically, in Figure 1, it indicates that the hypothesized paths model from stressors

to coping styles clearly follow the sequence paths. Second, the residuals are uncorrelated

with the other variables in the model and with each other. As Figure 1 clearly shows, each error

(e1 or e2 or e5) does not have any relation (path) to other error(s). Also, each error does not

correlate with other variables. For example, e1 does not show a path to distractions, verbal abuse,

avoidance-cognitive, approach-cognitive or approach behaviour. Third, the sequence flow is one-

way, as shown in Figure 1. Fourth, the results show a very good fit between the model and the

data with comparative fit index (CFI) and Tucker-Lewis index (TLI) > .90 (.97 and .95,


15/34


respectively, in the results); and root mean square error of approximation (RMSEA) < .05 (i.e.

.04, in the results).

Fifth, the variables are measured on interval scales or better. Each variable in the model

is a factor or subscale. As mentioned later, only subscales or factors derived from items resulted

from the CFA mentioned earlier were included on the path analysis model. Thus, each variable is

a composite score of its items with very good composite reliability. Consequently, each variable

has at least an interval scale properties. Finally, each variable has good reliability Cronbachs

alphas and Composite as shown in Table 1.

On the path analysis, for simplicity and clarity, only major variables or factors are

included on the model (see the hypothesized model). Structural model normally requires a larger

sample size then included in this study (N=125). However, sample size in path analysis depends

on the number of parameters that can be estimated by the saturated model (Anshel & Sutarso,

2010; Byrne, 1993; Kline, 1998; Tang, Luna-Arocas, & Sutarso, 2004). Because the present path

model has six variables, or factors (three factors of RSS i.e., distractions, subpar performance,

and verbal abuse, and three factors of RCS, i.e. avoidance-cognitive, approach-cognitive, and

approach-behaviour), the number of parameters is estimated to be 21. This is based on the

formula, [p(p+1)]/2, wherepis the number of variables. A recommendation for the minimum

sample size is to include five data points for every estimated parameter (Kline, 1998; Tang,

Luna-Arocas, Sutarso, & Tang, 2004). Hence the sample size required for this model is 105. This

would make the current sample size (N=125) adequate for conducting the path model.

Chi-Square (2) statistic, the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI),

and the Root Mean Square Error of Approximation (Anshel & Sutarso, 2010; Bentler, 1990;

Schumacker & Lomax, 1998; Tang, Sutarso, Davis, Dolinski, Ibrahim, & Wagner, 2008) were

used to assess model fit. Low values and a non-significant Chi-Square would indicate a better fit.

However, this statistic is very sensitive to larger sample sizes. For example, Chi-Square may lead


16/34


to rejection of a model with good fit in larger sample sizes (Schumacker & Lomax, 1998).

Joreskog (1969) proposed the Normed Chi-Square, that is, Chi-Square adjusted by the degrees of

freedom (2/df) to assess model fit. The Normed Chi-Square values between 1.0 and 5.0 are

considered to fall within the level of acceptance (Schumacker & Lomax, 1998).

The CFI and TLI are indicative of data that are a good fit to the model if the fit is equal to

or greater than .90 (.90 = the lower bound of a good fit, .95 or higher = an excellent fit). The

Root Mean Square Error of Approximation (RMSEA) values below .08 are considered as

indication of a good fit (i.e., .08 is the upper limit of a good fit, whereas .05 or less is an

excellent fit; Browne & Cudeck, 1989).

The critical ratio (CR) is used to determine whether the path is significant (Byrne, 1993;

Tang, Sutarso, Akande, Allen, Alzubaidi, Ansari, et al., 2006; Tang, Sutarso, Davis, Dolinski,

Ibrahim, & Wagner, 2008). A path is significant at .05, .01, or .001 when the critical ratio (CR) is

equal to or greater than 1.96, 2.58, or 3.50, respectively.

Results

Confirmatory Factor Analysis

Multiple factors model of Confirmatory factor analysis (CFA) were used (i.e., 9-item, 3-

factor of the referees sources of stress and 9-item, 3-factor of the referees coping styles) to

examine the fit between the model and the data (Table 1). The CFA confirmed the factor

structures resulting from the earlier conceptual framework of the refereessources of stress

(RSS) and the referees coping styles(RCS). The CFA achieved all the statistical criteria

discussed earlier (2 = 153.66, df= 131,p= .09,2/df = 1.17,CFI = .97, TLI = .96, RMSEA =

.04). These results also indicated that both RSS and RCS scales achieved construct validity. All

standardized factor loadings from items included were significant, ranging from .62 to .91, which

passed the criterion of .32 or higher for factor loadings(Tabachnick, & Fidell, 2007). This resultprovided support for convergent validity. Composite reliability for each factor was very good; in


17/34


fact, composite reliability for each factor or sub-scales was superior then those of the Cronbachs

reliability.The items, factor loadings, statistical measurements, mean, standard deviation, and

reliability coefficients (i.e. Cronbach alpha and composite reliability) of each factor for both

scales are listed in Table 1.

Path Analysis

For simplicity and clarity, only subscales or factors derived from items resulting from the

CFA were included on the path analysis model (Anshel, Sutarso, & Jubenville, 2009; Anshel,

Sutarso, & Sozen, 2012). As shown in Figure 2, results of the path analysis tested two major

issues: (1) The model achieved a good fit, and (2) there were significant paths from referees

source of stressors to their coping styles.

The model achieved a good fit based on all the criteria mentioned in the method section.

The2and Normed2criteria of the model were very good [2 = 10.71; df = 9;p= .30 and

Normed2 (2/df) = 1.19]. The other fit indices were also supported the goodness of fit model

with the data (CFI = .97, TLI = .95, RMSEA = .04). Figure 2 includes the 2 and other fit

indices.

The significant paths were achieved on all paths of the model simultaneously (i.e.,

distractionssubpar performanceverbal abuseboth avoidance-cognitiveand approach-

cognitiveapproach-behaviour).Path analysis tested all the paths simultaneously (see Figure 2). To clarify these results and to

address each of the six hypotheses in this study, the analysis was divided for each path. The first path

showed a significant and positive relationship between distractionsand subpar performancewith

standardized regression estimate = .33 and CR = 3.92 or equivalent top< .001. This path supported

hypothesis 1 that distractionswas positively associated with subpar performance. Subpar performance

served as a mediatorof the relationship between the distractionsand verbal abuse. The second path,

association between subpar performanceand verbal abusewas positive and significant (.23 and CR =


18/34


2.63;p< .01). This finding supported hypothesis 2, in that subpar performancewas associated with

additional verbal abuse. Thus, the three previous factors each reflected the referees source of stress

(RSS) in this study. The next path will link RSS to RCS.

Verbal abusewas related to two paths, avoidance-cognitiveand approach-cognitive. The path

from verbal abuseto avoidance-cognitivewas significant (-.18 and CR = 1.99; p < .05). This path

indicated a significant and negative association from verbal abuseto avoidance-cognitive. In other

words, in support of hypothesis 3, the more referees received verbal abusethe less likely they were to

apply avoidance-cognitivecoping. The referees also appliedapproach-cognitivecoping. The path from

verbal abusetoapproach-cognitivecoping was positive and significant (.39 and CR = 4.66;p< .001).

Thus, in support of hypothesis 4, experiencing more frequent verbal abusewas associated with greater

use of an approach-cognitivecoping style.

Paths to an approach-behaviourcoping style were generated from both avoidance-

cognitiveand approach-cognitiveforms of coping. The fifth path, in support of hypothesis 5,

showed a significant and positive relationship between avoidance-cognitiveto approach-

behaviourcoping styles (.26 and CR = 3.10;p< .01); avoidance-cognitivecoping was positively

associated with approach-behaviourcoping. Finally, the sixth path between approach-cognitive

to approach-behaviourcoping styles was positive and significant (.28 and CR = 3.32 or p

Documents

2014_JSS A Model Linking Sources of Stress to Approach and Avoidance Coping Styles of Turkish Basketball Referees