Measuring the sport club experience quality for migrants in Germany: Development and validation of a sport club experience quality scale
International Journal of Applied Sports Sciences, Vol.30, No.1, pp.62-79
https://doi.org/10.24985/ijass.2018.30.1.62
초록
In Germany, the national initiative titled ‘Integration through Sport’ has been implemented to encourage migrants to participate in sport clubs that provide a range of activities and help build mutual relationships. In spite of this effort, the number of migrants in the clubs has continued to decrease rapidly over the past decade. Successful implementation of the program requires practitioners and operators to understand which experience quality elements actually encourage migrants to participate in the activities offered and can help the clubs in retaining them as members. Thus, the aim of this research was (1) to conceptualize the notion of sport club experience quality for migrants (SCEQM) and (2) to develop a valid and reliable instrument to measure the SCEQM. Consequently, 15 items in five factors were developed: staff, peers, sport program, interaction program, and facility. The result of confirmative factor analysis indicated that the data fit the model well.
Abstract
In Germany, the national initiative titled ‘Integration through Sport’ has been implemented to encourage migrants to participate in sport clubs that provide a range of activities and help build mutual relationships. In spite of this effort, the number of migrants in the clubs has continued to decrease rapidly over the past decade. Successful implementation of the program requires practitioners and operators to understand which experience quality elements actually encourage migrants to participate in the activities offered and can help the clubs in retaining them as members. Thus, the aim of this research was (1) to conceptualize the notion of sport club experience quality for migrants (SCEQM) and (2) to develop a valid and reliable instrument to measure the SCEQM. Consequently, 15 items in five factors were developed: staff, peers, sport program, interaction program, and facility. The result of confirmative factor analysis indicated that the data fit the model well.
Introduction
Migration is a timeless phenomenon (Doel, 2012). According to a report by the United Nations, Europe hosted 76 million international migrants in the year 2015, which is nearly a third of the 244 million international migrants worldwide in that year (United Nations [UN], 2015). As European countries continued to experience increasing migration flows, socio-cultural adaptation and integration of migrants in host countries have become a priority issue (Guild, Carrera, & Luk, 2017; Makarova & Herzog, 2014). In recent years, some European countries have started to use sport as a means of promoting the integration of migrants (Elling, De Knop, & Knoppers, 2001; Entzinger & Biezeveld, 2003; Kelly, 2011). Because of a low threshold, uniform character, and internationally standardized rules, sport is one of the few social activities that human beings indulge in that can be recognized in virtually every community and culture around the globe as a vehicle for bringing people of different social and ethnical backgrounds together (Eitzen & Sage, 2003; Elling & Claringbould, 2005; Grove & Dodder, 1982). Furthermore, many researches indicate that sport participation helps migrants cope with acculturative stress and maintain physical and mental health (Kim & Iwasaki, 2016; Kim & Lee, 2010; Walker, Halpenny, & Deng, 2011; Zhou, Zhang, & Stodolska, 2017).
In Germany, sport is considered as ‘the motor of integration’ (German Federal Government, 2007). With the help and coordination of more than 90,000 sport clubs1 across the country, the national initiative titled ‘Integration through Sport’ has been implemented to recruit migrants as club members (German Olympic Sports Federation, 2012). Recently, more than 90% of all sport clubs in Germany agreed to offer sports to migrants (Breuer & Feiler, 2017). Despite this effort, there has been a significant decrease in the number of migrants in sport clubs over the past decade. The number fell from 2.76 million in 2008 to 1.72 million in 2014 (Breuer, 2009; Breuer & Wicker, 2011; Breuer & Feiler, 2015). Thus, one of the biggest challenges for German sport clubs is the recruitment and retention of migrant members (Breuer & Wicker, 2011).
The service experience, defined as the customer’s subjective response to the holistic encounter in any service setting, has recently been acknowledged as a critical success factor among service providers at various levels (Lemke, Clark, & Wilson, 2011, Verhoef et al., 2009). Previous researches in the context of sport and leisure suggest that the delivery of a high-quality experience triggers positive word-of-mouth and loyalty on the part of customers, which is a major indicator of customer retention (Alexandris, Zahariadis, Tsorbatzoudis, & Grouios, 2004; Howat & Assaker, 2016; Howat, Crilley, & McGrath, 2008; Theodorakis, Kaplanidou, & Karabaxoglou, 2015). Implementation of the national program requires practitioners and sport club operators to understand which experience quality elements of sport clubs actually attract migrants to participate in the activities offered and help the clubs retain migrants as members.
While many scholars have made a concerted effort to understand experience quality, a few empirical studies have investigated the variables involved with the experience quality of sport clubs and focused on the demographic segment of migrants. Therefore, the purpose of this research was twofold: (1) to conceptualize the notion of sport club experience quality for migrants (SCEQM) and (2) to develop a valid and reliable instrument that can be useful for a better understanding of the SCEQM.
Theoretical background
Integration
Esser (2000) proposes that the notion of integration encompasses the process of ‘re-socialization’ and all aspects of the relationship between migrants and their host countries. Heckmann (2005) defines integration as a lasting process of inclusion and acceptance of migrants in the core institutions and statuses of the receiving society. It is an interactive process between migrants and receiving societies, and both of them have to learn new ways of interacting with each other and adapt their institutions to their needs (Heckmann, 2005). Also, Heckmann and Schnapper (2003) point out that an integration process concerns primarily migrants and their descendants, but it is an interactive, mutual process that changes the host society as well.
Likewise, the German federal government indicates that integration is the development of a common understanding regarding how to live together in society, which should be characterized by respect, mutual trust, shared responsibility, and a sense of community; integration can work only as a two-way process, which requires acceptance by the majority population and the willingness of immigrants to learn and respect the rules of the host country and to take responsibility for their own integration; integration should ensure that immigrants have equal opportunities and the chance to participate in all areas, especially social, economic, and cultural life (German Federal Office for Migration and Refugees, 2007).
As integration initiatives and a national-wide policy, the national integration plan was developed in 2007 (Böhmer, 2007). The ‘Integration through Sport’ has been implemented as part of the national integration plan with the German Olympic Sports Confederation (Deutscher Olympischer Sportbund; DOSB) as a program facilitator and the federal government as a financial sponsor working together. Regional sports confederations (Landessportbünde), 764 supporting clubs (Stützpunktvereine), and over 90,000 general clubs form the basis of the program (German Olympic Sports Confederation, 2012). The goal of the initiatives is to encourage people with a migration background-particularly girls and women, middle-aged adults, and senior citizens-to join sport clubs as active members or volunteers (German Olympic Sports Federation, 2014).
The idea of the program is that sport clubs with a range of integration programs can contribute to building meaningful interpersonal relationships and social support networks for migrants, and subsequently help improve their social skills (Breuer & Feiler, 2015; Shields & Bredemeier, 2007). By offering migrants and communities a place to interact (Keller, Lamprocht, & Stamm, 1998) and giving individuals a sense of belonging to teams, clubs, or a community (Ennis, 1999), sport clubs can play a crucial role in the social integration process. Thus, migrants’ participation and management of their experience in sport clubs are of vital importance to successful implementation of the integration initiatives.
Experience in the marketing context
To date, in the marketing research on service industries that include sport, fitness, and leisure, scholars have made exploratory attempts to conceptualize and measure customer experience (e.g., Brakus, Schmitt, & Zarantonello, 2009; Grewal, Levy, & Kumar, 2009; Verhoef et al., 2009). Abbott (1955) states that all products provide an experience and “what people really desire are not products but satisfying experiences” (p. 40). Berry, Wall, and Carbone (2006, p. 1) indicate that “by definition, a good customer experience is good customer service, thus the customer experience is the service.”
Recent marketing practice, however, has broadly defined the experience as “encompassing every aspect of a company’s offering-the quality of customer care, of course, but also advertising, packaging, product and service features, ease of use, and reliability” (Meyer & Schwager, 2007, p. 2). This expansive perspective considers customer experience as holistic in nature, incorporating the customer’s cognitive, affective, emotional, social, and physical responses to all interactions with a firm (Bolton et al., 2014; Lemke et al., 2011; Schmitt, 1999, 2003; Verhoef et al., 2009). Academicians and practitioners generally have come to agree that customer experience is a multidimensional construct (Schmitt, 1999, 2003; Verhoef et al., 2009). Hence, this research concentrates on the major accepted definition from Verhoef et al. (2009), and accordingly the experience of migrants in sport clubs is conceptualized as the migrants’ subjective response to their encounters in sport clubs.
Experience quality
Consistent with Zeithaml’s (1988) definition of quality, Lemke et al. define experience quality as “a holistically perceived judgement about the excellence or superiority of the customer experience” (2011, p. 849). Many researchers have developed measurements of the overall experience quality so far (Csikszentmihalyi & Larson, 1987, 2014; Csikszentmihalyi & LeFevre, 1989; Hormuth, 1986, Otto & Ritchie, 1996; Lemke et al., 2011). These measures, however, have yet to gain traction, either due to their recentness or due to the challenge in developing a single set of instruments that adequately captures customer experience across various industries (Lemon & Verhoef, 2016). Mittal, Kumar, and Tsiros (1999) state that service quality can be considered an antecedent of service experience. In this respect, Lemon and Verhoef (2016) suggest that it may be more fruitful to consider the existing service quality approaches that have been validated across many industries in order to define and create constructs that can be used to explore the measurement of experience quality. Hence, we have primarily referred to service quality literature and experience quality literature that deal with the procedures for investigating the constructs.
The concept of service quality has been understood in the academic world as the difference between the expectations and the perceptions of customers’ outcomes; it is the comparison that customers make between their expectations about a service and their perception of the way the service has been delivered to them (Caruana, Ewing, & Ramaseshan, 2000; Grönroos, 1984; Lehtinen & Lehtinen, 1991; Parasuraman, Zeithaml, & Berry, 1985, 1988). The service quality model has received a great deal of attention from service marketing researchers and has been systematically proposed in recent years (Kelley & Turley, 2001; Martinez & Martinez, 2010). The SERVQUAL (Parasuraman et al., 1988), in particular, is a marketing theory that has had a major influence in practice (Cronin & Taylor, 1992). Furthermore, the scale has been a popular starting point to guide efforts towards an overall customer experience measurement approach because it can be applied to a wide spectrum of settings (Roberts, Kayande, & Stremersch, 2014). Since the development of the SERVQUAL, a variety of models for the measurement of service quality have been proposed in service industries (Brady & Cronin, 2001; Cronin & Taylor, 1992, 1994; Dabholkar, Thorpe, & Rentz, 1995).
Research into service quality in sport and leisure settings, however, has been limited compared that of business industries. Based on the SERVQUAL, a number of scales have been developed with significant attention by researchers (Howat, Absher, Crilley, & Milne, 1996; Howat, Murray, & Crilley, 1999; Mackay & Crompton, 1990; Wright, Duray, & Goodale, 1992). Researchers employed specific instruments in the context of sport and leisure, such as fitness clubs (Chelladurai, Scott, & Haywood-Farmer, 1987), recreational sport (Ko & Pastore, 2005), sport centers (Kim & Kim, 1995), and sporting events (Calabuig, Mundina, & Crespo, 2010; Kelley & Turley, 2001, Ko, Zhang, Cattani, & Pastore, 2011), to eliminate the limitations of SERVQUAL, which does not adhere to the original expectation of providing a generic measure that could be applied to any service (Carman, 1990).
Although service quality models have widely been utilized as an important standard for judgement for the services in the context of marketing, they normally do not take into consideration the consumer’s emotional or hedonic inclinations (Parasuraman et al., 1985). Moreover, while service quality refers to the attributes that are under the control of a supplier, experience quality involves not only the attributes provided by a supplier but also the attributes brought to the opportunities by other consumers (Chen & Chen, 2010; Lemke et al., 2011). According to the relationship marketing approach, customer experience includes the interaction between customers (Lemke et al., 2011). The reciprocal characteristics of the interactions among customers include common values, shared culture, and social contacts with other customers (Brady & Cronin, 2001; Verhoef et al., 2009).
Based on both service and experience quality, we have conceptualized the SCEQM in this research as the migrant member’s cognitive and affective perception (or judgement) of a specific service transaction and also of the various interactions in sport clubs.
Conceptual framework of the SCEQM scale
The first step when developing any scale is to define the constructs being measured and decide on the components which will comprise the constructs (Clark & Watson, 1995). The process of formulating constructs in the proposed model began with an extensive literature review. Field observation and interviews with sport club staff, trainers, and migrant members were conducted to gain additional insights. Various service and experience quality scales were investigated in an attempt to identify the construct of the SCEQM: service quality (Brady & Cronin, 2001; Chelladurai et al., 1987; Chelladurai & Chang, 2000; Howat et al., 1996, 1999; Kim & Kim, 1995, Ko & Pastore, 2005; Parasuraman et al, 1988, Rust & Oliver, 1994), and experience quality (Csikszentimihalyi & Lefevre, 1989; Lemke et al., 2011; Otto & Ritchie, 1996).
The dimensions developed in the literature could be grouped under three major constructs: personnel, program, and facility (Brandy & Cronin, 2001; Rust & Oliver, 1994; Howat et al., 1999). Consist with this three-factor model, our findings from the interviews, on-site observation, and the literature review comprised five dimensions: staff, peer, sport program, interaction program, and facility, which were categorized along with the major constructs. Specifically, the interviews with managers and members led to the categorization of program into two constructs (i.e., sport program and interaction program).
Personnel: Staff and peer
A number of service and experience quality models put strong emphasis on the personnel. The typical service quality approach focuses on the attitude of service providers or on the interaction between the service providers and the customers (Brady & Cronin, 2001; Howat et al, 1996, 1999; Parasuraman et al., 1988; Rust & Oliver, 1994). In the SERVQUAL (Parasuraman et al., 1988), four of the five dimensions are related to the supplier (e.g., reliability, responsiveness, assurance, and empathy). Brandy and Cronin (2001) developed a three-dimensional model in which the quality of the interaction takes into account attributes such as attitude, behavior, and the expertise of the service provider. Howat et al. (1996), in their four-factor model of sport and leisure centers, explore the responsiveness and knowledge of staff in the dimension of personnel. Kim and Kim (1995), on the other hand, put an emphasis on employees’ attitude and reliability in the QUESC model of the sport center. Since the members of sport and leisure centers rely heavily on the expertise of staff or instructors, the constructs relating to the attitude, knowledge, and interaction between the staff and the members of the clubs have been mainly utilized (Howat et al., 1999, 2008).
Other researchers, however, stress the importance of the interaction within a group of customers in the case of teaching skills or coaching a team sport in the sport and fitness settings (Chelladurai & Chang, 2000; Ko & Pastore, 2005). The customer’s subjective perception towards a service can be influenced according to the attitude and behavior of other customers (Chelladurai & Chang, 2000; Zeithaml & Bitner, 1996). Chelladurai and Chang (2000) identify the critical aspect of interactions in a service operation as the ‘inter-clients interaction’. Lemke et al. (2011) indicate that the peer-to-peer construct is the perceived judgement of the superiority of consumers’ interaction among themselves and that it is a critical element in delivering a consumer experience.
Consequently, personnel can be divided into two types of dimensions in this proposed model: the staff and the peer. These dimensions consist of attributes such as friendliness, willingness to help, enjoyable interaction, and a harmonious relationship between the staff and the members or among peers.
Program: Sport program and interaction program
Program dimension is defined as the customer’s perception of the excellence of the program (Brady & Cronin, 2001; Howat et al, 1996, 1999). The program dimension was mainly employed in the literature of sport and fitness settings (Chelladurai et al., 1987; Howat et al., 1996, 1999; Kim & Kim, 1995), whereas it is hardly found in the generic service quality scales (Parasuraman et al, 1988). Kim and Kim (1995) address this dimension directly as ‘programs offered’, comprising children’s programs, family programs, play or goal-differentiated programs, variety of sports, and community activities. Howat et al. (1996) classify service and program as core service. The model developed by Howat et al. (1999), however, refers to the all attributes regarding physical facility itself as ‘core’ while service and programs are classified as peripheral which include a broad range of activities and up-to-date information, and on-time activities. The sport program dimension of this proposed model thus utilizes several attributes based on the aforementioned literature: convenient operating hours, convenient program schedule, interesting contents, and adequate level of sport program.
A significant and distinctive dimension in this proposed model is the integration program. German sport clubs organize social and cultural programs such as language/culture learning and outdoor activities as a means of integration (German Olympic Sports Federation, 2017). During these activities, each club member is expected to interact extensively with the other participants. Focused interviews with sport club managers and members also revealed that interaction activities were believed to be a crucial factor in the service experience of the sport club. Elliott and Hamilton (1991) identify that social interaction is an important factor when people decide to participate in sport and leisure programs. Similarly, Green (2005) states that participants at sport facilities can be motivated predominantly by social outcomes, suggesting that socializing with others can be a primary motivator for participation. According to Diehl and Berg (2006), retirees tend to participate in leisure programs that provide them opportunities to socialize with others.
Kim and Kim (1995), examining the items of socializing at sport centers, stress the social opportunity in the QUESC, among two attributes: interaction among members and opportunity to meet well-known people. Ko and Pastore (2005) support the dimension of interaction program, suggesting three items-opportunities for social interaction, a sense of family, and enjoyment of social interaction-as the main reasons that people sign up for the activities. Given the unique features of German sport clubs, it is important to investigate the interaction programs or the opportunities they provide as one of the dimensions. The specific attributes in the present study help define the dimensions of the Interaction Program: information on life, culture, and language, adequate socializing opportunities among members, and an adequate place for members to interact with each other.
Facility
This dimension encompasses both visual (e.g., cleanliness, modernity, size, and layout) and non-visual aspects (e.g., lighting, music, and temperature) (Brady & Cronin, 2001; Chelladurai & Chang, 2000; Howat et al., 1996, 1999; Kim & Kim, 1995; Parasuraman et al., 1988). Since services are inherently intangible and often require the customer to be present during the process, the physical environment of the clubs can have a significant influence on the customers’ perceptions of the overall quality (Bitner, 1992). Hence, a number of researchers describe the facility as a crucial factor in their service quality models (Brady & Cronin, 2001; Chelladurai et al., 1987; Howat et al., 1996, 1999; Kim & Kim, 1995; Parasuraman et al., 1988).
The SERVQUAL labels this dimension as ‘tangibles’, which include attractive facilities and up-to-date equipment (Parasuraman et al, 1988). Chelladurai et al. (1987) identify the primary goods in fitness services which include size of facilities, cleanliness, and convenience of use of equipment. According to Howat et al. (1999), the dimension of ‘core’ includes clean and well-maintained facilities, safe and secure parking areas, and high-quality equipment. Seven attributes for ‘ambience’ are proposed in the QUESC: comfortable temperature, adequate space, brightness, modern facilities, pleasant interior, cleanliness, and a locker room with a warm atmosphere (Kim & Kim, 1995). Based on prior researches, facility was developed with easy access by public transportation, a convenient-to-use facility and equipment, an easy-to-access location, cleanliness, modernity, moderate usage fee, easy-to-obtain information, and well-maintained equipment.
Guided by the standard scale development process advocated in literature (Churchill, 1979; Clark & Watson, 1995; Devellis, 2016), this research consisted of two studies. In Study 1, a pool of items was generated and a panel of experts evaluated the content-relevance items to ensure adequate representation of the constructs. Followed by item generation, a pilot study was conducted to examine the items and reduced their number in order to refine the measurement using the exploratory factor analysis (EFA). In Study 2, the new data set was analysed utilizing both the EFA and the confirmative factor analysis (CFA). The validity and reliability of the scale were established.
Study 1
Study 1 consisted of two distinct phases: a qualitative phase and a quantitative phase. A qualitative phase was conducted to formulate the item pool, and a pilot study as a quantitative phase was performed to refine the items.
Qualitative phase
According to the guidelines delineated by Clark and Watson (1995), of which the fundamental method to create an item pool is to systematically sample all content that is potentially relevant to the target construct, we conducted an extensive review of literature, on-site observation, and interviews with staff, trainers, and members in sport clubs. All face-to-face interviews were carried out in four sport clubs, and consisted of open-ended questions seeking opinions and insights of interviewees regarding their experiences at the clubs. As a result of the procedures, initial 50 attributes were generated. Identification, refinement, and content validity of the initial item set were completed by evaluation of an expert panel through a modified application of the Delphi technique, which is a “method of using expert opinion to help make decisions about practice, needs, and goals” (Thomas, Silverman, & Nelson, 2015, p. 280). Skutsch and Hall (1973) state that the opinions and judgements of experts will help gain specific information. To sum up, a set of 31 items were formulated to represent five dimensions consistent with the conceptual framework using a five-point Likert type scale ranging from 1 (strongly disagree) to 5 (strongly agree).
Quantitative phase
The main purpose of the quantitative phase was to understand the performance of the proposed model and purify the attributes using the EFA as a pilot study. Reducing the amount of items to a more manageable number was considered necessary so that the scale could be practically implemented by other researchers.
Data collection
The selection of migrant participation in the survey was made based on the following prerequisites: (a) The participant must be aged 14 and over, (b) he/she must have been living in Germany as a migrant for more than one year, and (c) he/she must have been a part of nonprofit sport clubs registered in the German Olympic Sport Federation for over three months. A self-administrated questionnaire was utilised and 250 questionnaires were distributed to migrant members belonging to 11 sport clubs in major cities in the state of North Rhine-Westphalia in Germany. Thirty-two responses were eliminated due to incomplete data, and 218 questionnaires were used for the analysis. Males made up 80% of the participants; of those, 51.8% were from European countries other than Germany and 34.3% had migrated from Asia. 78.5% (Mage = 21.2, SD = 6.26) of the participants were aged between 14 and 25 years. Roughly 40.8% of participants frequented sports clubs two to three times per week and 82.1% took part in some sports program for one to two hours per week (see Table 1).
Table 1.
Descriptive statistics for the demographic variables
| Variable | Category | Study 1 (N=218) | Study 2 (N=409) | ||
|---|---|---|---|---|---|
| N | % | N | % | ||
| Gender | Male | 175 | 80.3 | 262 | 64.1 |
| Female | 43 | 19.7 | 147 | 35.9 | |
| Age | 14-17 | 83 | 38.1 | 110 | 26.9 |
| 18-25 | 88 | 40.4 | 121 | 29.5 | |
| 26-35 | 30 | 13.8 | 84 | 20.5 | |
| 36-45 | 14 | 6.4 | 64 | 15.8 | |
| Over 46 | 3 | 1.4 | 30 | 7.3 | |
| Country of origin | Europe | 113 | 51.8 | 209 | 51.1 |
| Africa | 27 | 12.4 | 32 | 7.8 | |
| America | 3 | 1.4 | 12 | 2.9 | |
| Asia | 75 | 34.4 | 156 | 38.2 | |
| Type of sport club | Team sport | 78 | 35.8 | 170 | 41.6 |
| Individual sport | 140 | 64.2 | 239 | 58.4 | |
| Frequency of participation | 1 to 2 per month | 3 | 1.4 | 18 | 4.4 |
| 1 per a week | 52 | 23.9 | 92 | 22.5 | |
| 2 to 3 per a week | 89 | 40.8 | 213 | 52.1 | |
| 4 to 5 per a week | 44 | 20.2 | 60 | 14.7 | |
| 6 to 7 per a week | 30 | 13.8 | 26 | 6.3 | |
| Duration on participation | 30 min – 1 hour | 0 | 0 | 36 | 8.8 |
| 1 - 2 hours | 12 | 5.5 | 303 | 74.1 | |
| 2 - 3 hours | 179 | 82.1 | 58 | 14.2 | |
| over 3 hours | 27 | 12.4 | 12 | 2.9 | |
Data analysis
With the samples (N = 218), the EFA was conducted to purify the scale and form a tentative factor structure using SPSS 21.0. Along with varimax rotation, a principal component extraction method was employed, which is widely utilized for grouping the variables under a few unrelated factors (Hair et al., 2010). A principal component analysis is most appropriate when data reduction is a primary concern, focusing on the minimum number of factors needed to account for the maximum portion of the total variance represented in the original set of variables (Hair et al., 2010). An appropriate number of factors was decided based on the Kaiser criteria, eigenvalues greater than 1 (Kaiser, 1960), and the scree test (Cattell, 1965; Zwick & Velicer, 1982). Items with a factor loading of .4 and above were selected to represent practical significance, while items loaded on more than one factor with factor loadings greater than .3 were deleted (Hair et al., 2010). Particular attention was paid to the number of significant items loading on the respective factors to avoid retaining factors with only one item or too many items that could be reasonably interpreted (Fabrigar, Wegener, MacCallum, & Strahan, 1999). The Kaiser-Meyer-Olkin (KMO) test helped in the measurement by comparing the magnitudes of the observed correlation coefficients with the magnitudes of the partial correlation coefficients.
Results
The KMO measure was .831 in this study, indicating that the sample was appropriate and ‘meritorious’ for a factor analysis (Kaiser, 1974). According to Hair et al. (2010), Bartlett’s test was conducted to test the hypothesis that the correlation matrix was an identity matrix, which was 5542.752 (p < .001); this means that the EFA of the data set was adequate to proceed. Based on the Kaiser criterion, seven factors with more than 1.0 eigenvalues emerged with 72.94% of the total variances explained (Kaiser, 1960). As for the item purification, 23 items of initial 31 items were retained and eight items with double loadings were eliminated (Hair et al., 2010). Overall, the EFA in the pilot study led to seven factors and 23 items.
Study 2
In Study 2, the CFA was conducted to investigate the reliability of the model and test if it was fit, along with the EFA, which determined the factor structure and construct validity. Although the EFA provides procedures for determining an adequate number of factors and the pattern of factor loadings primarily from the data using a separate sample, the CFA is required to test the hypotheses about the data and should be used to endorse the factor structure and provide further evidence of the validity of the construct (Fabrigar et al., 1999).
Data collection
A new self-administrated questionnaire was employed in 29 sport clubs of North Rhine-Westphalia state in Germany between July and December 2016. A total of 450 questionnaires were distributed. 41 incomplete surveys were discarded, while 409 were deemed appropriate, indicating a 90.89% response rate. The average age of the participants was 25.09 (SD = 9.3), with 64.1% male, 51.1% from other European nations, and 38.1% from Asia. More than half of these participated in the activities of the sport club at least twice a week and dedicated one to two hours to sport activities (see Table 1).
If the size of the sample in a single study is sufficiently large, the sample can be randomly split in half (Fabrigar et al., 1999). A sample size greater than 200 is considered adequate for the CFA (Barrett, 2007; Brown, 2006; Hair et al., 2010). Following Fabrigar et al. (1999) and Hair et al. (2010), using the random split function in SPSS 21.0, 409 samples were separated into two data sets: one for the EFA (N = 204) and the other for the CFA (N = 205).
Data analysis
The EFA with principal component extraction and varimax rotation was conducted using the samples (N = 204). Followed by the EFA, the CFA was employed based on the final five-factor model. Using AMOS 21.0, the second data set (N = 205) was submitted for the CFA to validate the models. Kline (2005) suggests using multiple fit indices to generate adequate information and assess the overall fit data. Hoyle and Panter (1995) also recommend that data analyses include absolute and incremental fit indices. Hu and Bentler (1999) note that an absolute fit index assesses how well an a priori model reproduces the sample data. Absolute fit indices include the Goodness-of-Fit Index (GFI; Jöreskog & Sörbom, 1989), the Root Mean Square Error of Approximation (RMSEA; Steiger, 1990), and the Standard Root Mean Squared Residual (SRMR; Steiger & Lind, 1980); In contrast, incremental fit indices measure the proportionate improvement in fit by comparing a more restricted baseline model with a target model (Bentler & Bonett, 1980). The Comparative Fit Index (CFI; Bentler, 1990) and the Tucker-Lewis Index (TLI; Tucker & Lewis, 1973) were utilized as incremental fit indices in this research.
The cut-off criterion for acceptable model fit is as follows: GFI > .9, RMSEA < .10, SRMR < .05, TLI > .9 and CFI > .9 (Bentler, 1990; Byrne, 1998; Diamantopoulos & Siguaw, 2000; Hu & Bentler, 1999; MacCallum, Browne, & Sugawara, 1996; Shevlin & Miles, 1998).
Result
Exploratory factor analysis
In the EFA, the KMO measure was .868 and the Bartlett test of sphericity was 7163.618 (p < .001). According to Hair et al. (2010), this data set was appropriate for the application of the EFA. As shown in Table 2, the second factor identification contained five factors that accounted for 70.63% of the total variances following the Kaiser criterion and the scree plot test (Cattell, 1965; Kaiser, 1960; Zwick & Velicer, 1982). Six double-loaded items were eliminated and 17 items were presented (Hair et al., 2010).
Table 2.
Summary results for the SCEQM scale in Study 2.
| Factor | ITEM | Factor Loadings | Eigen values | Variance | Cumulative Variance | |
|---|---|---|---|---|---|---|
| Factor 1–Staff [α = .882, AVE = .733] | ||||||
| Staff are reliable | ST1 | .815 | 9.325 | 38.853 | 38.853 | |
| Interacting with staff is enjoyable | ST4 | .797 | ||||
| I have a harmonious relationship with staff | ST5 | .762 | ||||
| Staff are willing to help me | ST2 | .750 | (.349) | |||
| I can have a nice conversation with staff | ST3 | .669 | (.357) | |||
| Factor 2–Peer [α = .895, AVE = .756] | ||||||
| Peers are reliable | PE1 | .832 | 2.919 | 12.164 | 51.017 | |
| I can have a nice conversation with peers | PE3 | .815 | ||||
| Interacting with peers is enjoyable | PE4 | .679 | ||||
| Peers are willing to help me | PE2 | .652 | (.381) | |||
| I have a harmonious relationship with peers | PE5 | .647 | (.355) | |||
| Factor 3–Sport Program [α = .909, AVE = .794] | ||||||
| The operating hours are convenient | PR4 | .858 | 2.124 | 2.124 | 59.868 | |
| The contents of sport program is interesting | PR1 | .838 | ||||
| The level of sport program is adequate | PR2 | .769 | ||||
| The sport program schedule is convenient | PR3 | .734 | (.291) | |||
| Factor 4–Interaction program [α = .811, AVE = .678] | ||||||
| The opportunities to interact among members are offered | IN2 | .807 | 1.500 | 6.251 | 66.120 | |
| It is easy to have life/culture/language information | IN3 | .739 | ||||
| The place to interact among members is provided | IN1 | .741 | ||||
| Factor 5–Facility [α = .873, AVE = .629] | ||||||
| It is convenient to use facility and equipment | FA6 | .803 | 1.083 | 4.511 | 70.631 | |
| The usage fee is moderate | FA4 | .781 | ||||
| It is easy to access by public transportation | FA1 | .740 | ||||
| The facility is clean | FA5 | .723 | ||||
| The information for using facility is easy to obtain | FA7 | .677 | (.296) | |||
| The facility are located in easy-to-access place | FA2 | .636 | (.354) | |||
| The equipment is well-maintained | FA3 | .622 | (.305) |
Raubenheimer (2004) recommends a minimum of three items per a factor. Short, Sullivan, and Feltz (2005) propose that it is desirable if each factor in the hypothesized model has a similar number of items for the efficiency and validity of the scale. In addition, Nunnally (1967) suggests that “in some cases the domain of related variables is so small that any one of the few observable in the domain will suffice to measure the construct” (pp. 85-86). Based on the aforementioned literature, although each set of four items in the sport program and facility dimensions was derived by the EFA, only three items with the highest factor loadings were retained in order to formulate a more manageable number of items and refine the proposed model. Finally, 15 attributes were designed to measure five dimensions: staff (three items), peer (three items), sport program (three items), interaction program (three items), and facility (three items).
Confirmative factor analysis
The CFA model is shown in Figure 1. It was conducted to analyse the proposed five-factor model using maximum likelihood estimation in AMOS 21.0. The result of CFA in the five-factor model suggests that fit indices exceed the minimum recommended criterion. Generally, the data fit the model well; GFI = .920, RMSEA = .080, SRMR = .033, CFI = .940, and TLI = .921 (see Table 3).
Table 3.
The result of CFA for the five-factor model
| GFI | RMSEA | SRMR | CFI | NFI | TLI | |
|---|---|---|---|---|---|---|
| criteria | >.9 | <.10 | <.05 | >.9 | >.9 | >.9 |
| result | .920 | .080 | .033 | .940 | .919 | .921 |
Reliability estimates
The reliability was assessed using Cronbach’s alpha correlation coefficient (Nunnally, 1967) and the average variance extracted (AVE; Fornell & Larker, 1981). Cronbach’s alpha indicates internal consistency by testing the homogeneity of all the items in the instrument subscale (Nunnally, 1967). Nunnally and Bernstein (1994) suggest that the Cronbach’s alpha of any construct should be greater than .7. The AVE was conducted to identify the reliability of the constructs in the amount of variance attributed to the measurement error (Fornell & Larcker, 1981). The criterion for construct reliability is that the AVE measures should exceed .5 (Fornell & Larcker, 1981). As shown in Table 2, all dimensions in the five-factor model met the .7 minimum Cronbach’s alpha criteria ranging between .811 and .909. Also, the AVE values all emerged above the recommended criteria over .5, ranging from .629 to .794. Moreover, Table 4 shows the results of discriminant validity, which are the correlations among five factors comparing the squared correlations against the AVE. All the inter-correlations of the factors are lower than the r < .85 (Kline, 2005) and all of the AVE scores for each construct exceed the squared correlations (Fornell & Larcker, 1981).
Table 4.
Correlations among the constructs
| Staff | Peers | Sport Program | Interaction Opportunity |
Facility | AVE | |
|---|---|---|---|---|---|---|
| Staff | 1 | .733 | ||||
| Peer | .294*** | 1 | .756 | |||
| Sport program | .223*** | .278*** | 1 | .794 | ||
| Interaction program | .142*** | .253*** | .132*** | 1 | .678 | |
| Facility | .116*** | .168*** | .175*** | .145*** | 1 | .629 |
Discussion and Conclusions
Discussion
The current study proposes a sport club experience quality model for migrants in Germany, developing a systematic framework of factors that help comprehensively capture the migrant’s perceptions of the service experience. Guided by the standard scale development process (Churchill, 1979; Clark & Watson, 1995; Devellis, 2016), two studies provide the significant and empirical evidence for the reliability and validity of the SCEQM.
In Study 1, the research generates the item pool and purifies the items through a qualitative phase and a pilot study. Using the EFA, Study 2 identifies five dimensions with 15 items that are important service encounters for migrant members in German sport clubs. The findings indicate that migrant members base their perceptions of a sport club experience on five dimensions: staff, peer, sport program, interaction program, and facility. In terms of the validity, the results of CFA support the factorial structure of the proposed model, providing a good fit for the data. The findings suggest that migrant members evaluate sport club experience quality at a dimensional level and an attribute level.
This study increases support for the use of dimensions in previous research. The dimensions in this study indicate that the construct of experience quality at sport clubs is similar to the multi-dimensional model of service quality in the sport setting as well as other service settings: staff (Brady & Cronin, 2001; Howat et al, 1996, 1999, Parasuraman et al, 1988), peer (Chelladurai & Chang, 2000; Lemke et al., 2011), sport program (Chelladurai et al., 1987; Kim & Kim, 1995), and facility (Brady & Cronin, 2001; Chelladurai & Chang, 2000; Howat et al., 1996, 1999; Kim & Kim, 1995; Parasuraman et al., 1988).
This study, however, shows that experience quality for migrants is more closely associated with the intangible items (e.g., staff reliability, peer relationship, and interaction), while much literature puts emphasis on tangible items such as physical facility, equipment, accessibility, and price. In addition, the dimension of interaction program provides a unique finding compared with previous literature; it is also different from other researches mainly because it was developed from the perspective of the specific demographic segment of migrants.
Interaction with staff and peers and interacting program are key experience factors for migrant members. The interaction happens immediately when migrant members come to the sport clubs, because most service experience involves the interaction between the provider and the members, or among members (Zeithaml, Parasuraman, & Berry, 1985). The results of this study indicate that migrant members consider interaction the most significant experience at sport clubs. This finding suggests that migrant members perceive attending sport clubs as opportunities for interpersonal relationships. Hence, sport clubs can be venues for enhancing social networks through sharing of information and experience, which contribute to interactive and cultural integration of migrants into the host society.
Consequently, this study provides expanded theoretical and practical understanding of experience quality by developing a conceptual framework and measurement instrument for the sport club setting. The findings also suggest insights for further exploration of the factors and attributes in the context of sport and leisure.
Implications
The SCEQM scale can provide practitioners with a valid and reliable instrument for a better understanding of experience quality in a sport club from migrants’ perspective. Along with the generic dimensions (e.g., staff, sport program, and facility), the particular dimensions (e.g., peer and interaction program) allow practitioners to identify and solve quality problems and build further sophisticated management strategies on the basis of the feedback from members. Furthermore, monitoring the SCEQM helps managers and trainers determine what specific characteristics need to be strengthened, augmented, or even deemphasized in the process of service provision. The efforts to improve experience quality can lead to club members’ satisfaction and loyalty. In addition, while the proposed model focuses on a specific setting-migrants’ participation in sports clubs-it can be also applied to other contexts of sports participation where interaction is highly emphasized (e.g., sport centers, fitness centers, and other recreational venues).
Limitations and future research
The present study is not without its limitations. Although the majority of evidence from this study supports the validity and reliability of the SCEQM scale, it is important to reemphasize that validity and reliability are considered ongoing processes (Devellis, 2016). It is possible, despite the findings in constructing a comprehensive measure of the SCEQM, that not every dimension of the SCEQM is included in this research. Hence, we encourage future research to identify more relevant dimensions and attributes of the sport club experience. Further evidence is needed to demonstrate the structure factor and/or criterion validity in future studies. More data will ensure the stability of the proposed model and generalizability of the findings.
The present research allows follow-up researches to conduct cross-cultural comparisons. As the samples employed in the research include only migrants in the western part of Germany, the proposed model needs to be applied to a wider range of sport clubs in the rest of Germany as well as overseas, to highlight social integration and sports as a means of integration. Until its reliabilities and validities are confirmed for other populations, the SCEQM cannot be generalized across cultures or countries. Future research is encouraged to investigate the temporal stability of the SCEQM over time.
Given that the present research is an attempt to develop the constructs of the SCEQM, investigating the relationships between the SCEQM and other outcome variables can be productive. For example, experience satisfaction and behavioral intensions (e.g., willingness to revisit and/or recommend sport clubs to others) will be highly correlated with the SCEQM. Future research can provide empirical evidence that the SCEQM has an impact on the outcome variables.
1In Germany, there are approximately 91,000 sport clubs that work as non-profit organizations where an average 420 members participate in sport activities. Half of the clubs are multi-sport clubs providing various types of sports (Nichols, Wicker, Cuskelly, & Breuer, 2015). 41,800 sport clubs (46.3%) have their own facility. 55,200 clubs (61.2%) use public sport facilities, of which 51.2% pay usage fee. Half of all sports clubs charge a maximum monthly membership fee of € 2.50 for children, € 3 for adolescents and € 6.30 for adults (Breuer & Feiler, 2017).
References
Methods and goals in human behavior genetics (, ) (1965) New York, NY: Academic Press Cattell, R. B. (1965). Methodological and conceptual advances in evaluating hereditary and environmental influences and their interaction. In S. G. Vandenberg (Ed.), Methods and goals in human behavior genetics (pp. 95-140). New York, NY: Academic Press. , Methodological and conceptual advances in evaluating hereditary and environmental influences and their interaction, In, pp. 95-140
Flow and the foundations of positive psychology (, et al., , ) (2014) Dordrecht: Springer Csikszentmihalyi, M., & Larson, R. (2014). Validity and reliability of the experience-sampling method. In M. Csikszentmihalyi (Ed.), Flow and the foundations of positive psychology (pp. 35-54). Dordrecht: Springer. , Validity and reliability of the experience-sampling method, In, pp. 35-54
Der nationale integrationsplan: Neue wege - neue chancen. [The national integration plan: New ways - new opportunities] (German Federal Government) (2007) German Federal Government (2007). Der nationale integrationsplan: Neue wege - neue chancen. [The national integration plan: New ways - new opportunities] Retrieved from https://www.bundesregierung.de/Content/DE/Archiv16/Artikel/2007/07/Anlage/2007-08-30-nationaler-integrationsplan.pdf?__blob=publicationFile&v=1 , Retrieved from https://www.bundesregierung.de/Content/DE/Archiv16/Artikel/2007/07/Anlage/2007-08-30-nationaler-integrationsplan.pdf?__blob=publicationFile&v=1
Integration in Germany: Support networking dialogue (German Federal Office for Migration and Refugees) (2007) German Federal Office for Migration and Refugees (2007). Integration in Germany: Support networking dialogue. Retrieved from http://www.bamf.de/SharedDocs/Anlagen/EN/Publikationen/Broschueren/02-integration-in-deutschland-e.pdf?__blob=publicationFile , Retrieved from http://www.bamf.de/SharedDocs/Anlagen/EN/Publikationen/Broschueren/02-integration-in-deutschland-e.pdf?__blob=publicationFile
Integration durch sport: Ein programm stellt sich vor. [Integration through sport: A program is presented] (German Olympic Sports Federation) (2012) German Olympic Sports Federation (2012). Integration durch sport: Ein programm stellt sich vor. [Integration through sport: A program is presented] Retrieved from https://www.dosb.de/fileadmin/fm-osb/arbeitsfelder/ids/images/2012/Downloadversion_dosb_integration_scr.pdf , Retrieved from https://www.dosb.de/fileadmin/fm-osb/arbeitsfelder/ids/images/2012/Downloadversion_dosb_integration_scr.pdf
Integration durch sport: Programmkonzeption (German Olympic Sports Federation) (2014) German Olympic Sports Federation (2014). Integration durch sport: Programmkonzeption. Retrieved from http://www.integration-durch-sport.de/fileadmin/fm-dosb/arbeitsfelder/ids/images/2014/Programmkonzeption_3_Aufl_2014.pdf , Retrieved from http://www.integration-durch-sport.de/fileadmin/fm-dosb/arbeitsfelder/ids/images/2014/Programmkonzeption_3_Aufl_2014.pdf
Informationsdienst “Integration durch Sport”. [Information service “Integration through Sport”] (German Olympic Sports Federation) (2017) German Olympic Sports Federation (2017). Informationsdienst “Integration durch Sport”. [Information service “Integration through Sport”] Retrieved from http://ids.dosb.de/de/integration-durch-sport/service/informationsdienst/ , Retrieved from http://ids.dosb.de/de/integration-durch-sport/service/informationsdienst/
The integration of immigrants and legal paths to mobility to the EU: Some surprising (and encouraging) facts (, , et al., ) (2017) Guild, E., Carrera, S., & Luk, N. C. (2017). The integration of immigrants and legal paths to mobility to the EU: Some surprising (and encouraging) facts. Retrieved from https://www.ceps.eu/system/files/LegalMobilityPaths.pdf , Retrieved from https://www.ceps.eu/system/files/LegalMobilityPaths.pdf
The integration of immigrants in European societies: National differences and trends of convergence (Vol. 7) (, et al., ) (2003) Hamburg: Lucius and Lucius Heckmann, F., & Schnapper, D. (Eds.). (2003). The integration of immigrants in European societies: National differences and trends of convergence (Vol. 7). Hamburg: Lucius and Lucius
Handbook of sport psychology (, et al., , , et al., ) (2007) Indianapolis, IN: Wiley Shields, D.L.L., & Bredemeier, B.J.L. (2007). Advances in sport morality research. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of sport psychology (pp. 662-684). Indianapolis, IN: Wiley. , Advances in sport morality research, In, pp. 662-684
International migration report 2015 (United Nations (UN)) (2015) United Nations (UN, 2015). International migration report 2015. Retrieved from http://www.un.org/en/development/desa/population/migration/publications/migrationreport/docs/MigrationReport2015.pdf , Retrieved from http://www.un.org/en/development/desa/population/migration/publications/migrationreport/docs/MigrationReport2015.pdf