Abstract

Background

Presently, where inactivity is an overarching problem globally, participation in recreational sports activities has become an important outlet for promoting healthy lifestyles. However, there is a lack of instruments to evaluate the commitment of recreational and leisure sports users. This study aimed to adapt, develop and provide preliminary validation of the Recreational Sport Commitment Questionnaire (RESQ).

Methods

The development and validation process in this study consisted of four stages: Stage One involved item pool generation, Stage Two focused on content and face validation, Stage Three encompassed data collection, and Stage Four included the assessment of the psychometric properties of the RESQ. This final stage was conducted through Exploratory Factor Analysis (EFA; n鈥�=鈥�275) and Confirmatory Factor Analysis (CFA; n鈥�=鈥�313) to evaluate convergent validity, discriminant validity, and composite reliability.

Results

The initial item pool generated 60 items, whereby 42 items were retained after the content validity evaluation process by the panel of experts. Next, EFA analysis suggested 31 items distributed across seven factors. Lastly, psychometric properties of RESQ with 26 items were established through CFA (GFI鈥�=鈥�0.91, CFI鈥�=鈥�0.97, RMSEA鈥�=鈥�0.04, TLI鈥�=鈥�0.96, X虏/df鈥�=鈥�1.52) and other psychometrics evidence.

Conclusion

Based on the collected psychometrics evidence, RESQ is a valid questionnaire to evaluate the commitment of recreational sports users. The newly developed questionnaire allows sports organizations to assess the factors influencing sport commitment among recreational sports users. Based on the results of RESQ, sport programmes can be tailored made to optimize sports engagement and promote continued commitment.

Introduction

Participation in recreational sport is essential for promoting a healthy lifestyle and reducing the risk of chronic conditions such as obesity, diabetes, and heart disease [1]. Engaging in sports activities supports personal well-being by offering enjoyment and relaxation while also encouraging social interactions, for instance, with neighbours and family members [2]. Additionally, sports help manage stress and provide avenues for personal growth [2,3,4]. Beyond the physical health advantages, recreational sports foster community involvement and provide a sense of purpose and fulfilment, which significantly contribute to overall well-being [3, 4].

Consequently, understanding sport commitment is important to gain insight into why people continue participating in recreational and leisure sports. Sport commitment is a psychological construct that encompasses the determination and desire to continue in sports participation [5, 6]. Plethora of studies found that factors such as enjoyment, social support and personal investment are important factors influencing sport commitment [6,7,8,9,10,11,12,13,14,15,16,17].

There are extensive studies in the context of competitive and organized sports. Furthermore, established measures have been developed, such as the Sport Commitment Model [5] and the Sport Commitment Questionnaire-2 (SCQ-2) [6]. The initial Sport Commitment Model (SCM) consisted five sources of sport commitment: sport enjoyment, involvement alternatives, personal investment, social constraints and involvement opportunities [5]. The SCM model underwent numerous tests and adjustments to explain why individuals participate and stay committed to sports. These include expanding the SCM model [6] and assessing its psychometric properties across countries [7, 8]. Overall, studies that employed SCM to explore sport commitment have shown evidence that sport enjoyment significantly influences commitment among individuals of all ages and populations [6,7,8,9,10,11,12]. Alternatively, various other factors have surfaced as key contributors to the commitment to sport. For example, competitive gymnasts, swimmers and athletes who return from injuries found that social support significantly contributed to their involvement and commitment to the sport [9, 13,14,15]. Studies also found that personal investment plays a crucial role in influencing athletes鈥� commitment to their sport [9, 14,15,16,17,18]. Besides, studies have shown that involvement opportunities directly impact sport commitment [6, 7, 14, 19, 20]. While the development of SCQ-2 was introduced to assess the commitment of elite athletes or competitive sport, relatively little attention has been given to commitment within recreational sport. Commitment in recreational sports settings differs from commitment towards competitive sports in several significant ways. The words and phrases in SCQ-2 are elite or professional sports-oriented; hence, they did not capture the essence of recreational activities, for example, 鈥淧eople who are important to me attend the majority of my competitions in this sport.鈥� 鈥淚 push myself to win every time I compete in this sport鈥�. In recreational sport, the focus is often on enjoyment, social interaction, and personal fulfilment rather than competition and performance outcomes [21]. Additionally, recreational sport participants may have diverse non-competitive motivations that are intrinsically motivated and goals for engaging in recreational sport activities, ranging from improving one鈥檚 health and fitness to seeking fun, leisure and relaxation. Numerous studies also reported additional factors that significantly influence individual engagement in sport [7, 22,23,24,25,26,27,28,29]. For instance, Berki [7] and Te Velde et al. [22] highlighted the importance of health benefits for general populations ranging from adolescents to adults. Other studies indicated the role of weight management across different age groups and sex in sport participation [25,26,27,28]. Furthermore, adolescents perceive appearance as an important factor influencing sport commitment [29]. Therefore, there is a need to develop a new instrument to examine the sport commitment among recreational sport users鈥� commitment.

According to Scanlan et al. (2016) [6], sport commitment consists of two dimensions, i.e., enthusiastic commitment and constrained commitment. Enthusiastic commitment refers to a psychological construct that drives one鈥� desire to maintain participation in a sport over time [6]. This form of commitment is intrinsically motivated and reflects a genuine desire to engage in the sport rather than stemming from external obligations or pressures [6]. On the other hand, constrained commitment refers to a psychological construct reflecting a sense of obligation to continue participating in a sport over time. This type of commitment arises from perceived external pressures or responsibilities [6]. Examples of the sources of enthusiastic commitment are sport enjoyment and the desire to excel [6]. While examples of sources for constrained commitment are personal investment and social constraint [6].

Purpose of the study

The development and validation of the Recreational Sport Commitment Questionnaire (RESQ) plan to address several crucial needs in the context of recreational sports research. Firstly, there was a significant gap in the existing literature that is comprehensive and psychometrically sound measures specifically designed to assess the commitment to recreational sports activities [7]. Currently, available instruments that are related to commitment and motives in recreational/ leisure activities are the Physical Activity and Leisure Motivation Scale [30], Exercise Motivation Inventory [31], and Participation Motivation Questionnaire [32].

Though the Sport Commitment Questionnaire (SCQ-2) [6] is the closest relevant tool to measure sport commitment, however, SCQ-2 merely focuses on competitive and organized sports, neglecting the unique aspects of commitment in recreational settings [6,7,8]. Furthermore, Scanlan et al. [6] recommended that the questionnaire be utilized to assess the commitment among elite athletes. This is because items in the questionnaire were designed to be relevant to elite athletes [6]. Therefore, the development of the RESQ fills this gap by providing researchers and practitioners with a reliable tool to understand engagement and commitment in recreational and leisure sport.

Secondly, a validated measure of recreational and leisure sport commitment has the potential to advance our understanding of the factors that influence individuals鈥� sustained participation in recreational and leisure sports activities. Examining the factors associated with and predicting commitment to recreational sport allows researchers to uncover critical elements influencing commitment. This insight can help to design tailored treatments to increase people鈥檚 commitment to sport activities. These key insights can help to build more successful strategies for encouraging continuing engagement in recreational sport.

The RESQ is practically relevant for individual who engage in recreational sport settings. Coaches, teachers, physical education instructors, and program managers may utilise RESQ to assess participant sport engagement. RESQ could help identify persons at risk of disengagement or dropout and develop interventions to increase sport participation. Furthermore, using RESQ allows sport practitioners to monitor changes in participant commitment over time and assess the effectiveness of their programs.

Overall, this study emphasised the importance of understanding commitment from the context of recreational sport activities. Therefore, the development and validation of the RESQ fills the gap in the literature and serves as a valuable tool for increasing participation and promoting personal well-being.

Methods

This study employed a cross-sectional design. The development and validation process of the RESQ adhered closely to DeVellis鈥� guidelines [33], commencing with the generation of items, followed by expert evaluation and a validation study.

Ethical considerations

Before data collection, the research protocol underwent a rigorous review and was subsequently approved by the University Malaya Ethics Board (UM.TNC 2/UMREC). Prior to the study, all participants were given comprehensive details regarding the objectives and methods, as well as the possible risks and advantages of the research. Each participant was provided informed consent by signing consent forms for those present in person or utilizing electronic consent procedures for those participating online. Participation was voluntary, and they were made aware of their right to withdraw from the study at any point without facing any consequences.

Sampling methods

The sampling for this study involved a few steps. In order to enhance and evaluate the RESQ for content validity (questionnaire items), a panel of experts in sport psychology and related fields was purposively selected. It is commonly agreed upon that these experts are the individuals who are experts in the relevant areas [34]. Polit and Beck suggested [35] that a minimum of six experts is vital for the content validity evaluation of items. Therefore, during the developmental stage, seven experts were chosen to examine the content validity of the questionnaire. Prior to reviewing the items, the objectives and rationale of RESQ were conveyed to expert members.

Subsequently, to recruit a sufficient sample size for further SEM analysis, G Power version 3.1 was employed, which suggested 166 respondents. However, a minimum of 200 respondents was suggested [36] to utilize the SEM analysis method.

This study employed purposive sampling to ensure adequate representation of sex, age groups, and location. Inclusion criteria were actively engaged in their chosen sport (participating at least once per week), having a minimum of one year of experience, being above 18 years of age, and being able to provide informed consent. In addition, those who participated in competitive sports were excluded to ensure that the targeted population were recreational-oriented. Individuals with low English comprehension were excluded from the study. This study consisted of 275 and 313 respondents for EFA and CFA analysis respectively after eliminating the outliers.

Stage one: items generation

The Recreational Sport Commitment Questionnaire (RESQ) was developed by thoroughly by examining the literature on sport commitment, encompassing competitive and recreational sports settings. The researchers utilized established questionnaires such as the Sport Commitment Questionnaire [5, 6], Exercise Motivation Inventory [31, 37] and the PALMS [30, 38] to identify suitable items that would effectively measure the cognitive, affective, and behavioural aspects of commitment to recreational sport. These three questionnaires were included for item generation due to their robust theoretical foundations and comprehensive coverage of key dimensions of sport commitment. For example, Sport Commitment Questionnaire-2 provides a well-validated psychometric property for understanding factors contributing to sport commitment. While Exercise Motivation Inventory captures a wide range of motivational factors crucial for measuring commitment鈥檚 behavioural aspects. Finally, PALMS focuses on psychological motivation, which addresses the affective components of physical activity. By integrating items from these validated frameworks, the questionnaire ensures a thorough and reliable assessment of complex commitment dimensions in recreational sport.

Initially, the proposed RESQ consisted of eight constructs with 60 items, namely sport enjoyment (5 items), personal investment (11 items), social support (8 items), social constraint (4 items), stress management (4 items), desire to excel (12 items), appearance (6 items), positive health (10 items).

A 5-point Likert scale was selected during the development of RESQ due to several advantages. First, the method presented an ideal combination of clarity and detailed input for the respondents [39]. Second, the 5-point Likert scale can detect even the most nuanced differences in responses and allow researchers to interpret results easily [40]. Furthermore, a neutral midpoint on the scale allows respondents to express neutrality or uncertainty, thereby improving the overall accuracy of the data gathered [41]. Finally, the 5-point Likert scale is a well-established global rating system allowing for more straightforward comparisons and evaluation of outcomes among diverse studies [42].

Stage two: content and face validity

Content validity

The initial selection of items was evaluated by a panel of seven experts representing related disciplines such as sport psychology, sport sociology, physical education, sport coach and language experts. Expert evaluations of items were based on item relevance, clarity, and representativeness using a four-point scale, as shown in Table听1. The responses from the experts were then computed for CVI (I-CVI) and K* score analysis, which aimed to evaluate each item鈥檚 relevance, clarity and representation. The I-CVI score for each item was determined by calculating the proportion of experts assigned a rating of 3 or 4. CVI values ranged from 0 to 1, with higher values indicating a stronger content validity. The minimum acceptable I-CVI value is 0.80, based on a suggestion by Lynn (1986) [43] and McHugh (2012) [44]. The interpretation for 魏* score strictly follows the suggestion from literature whereby items with lower scores were eliminated [43,44,45].

Table 1 4-point likert scale for each evaluation criterion

Following the criteria suggested by these studies [46,47,48], 42 items met the requirements for the REL criteria, 45 items for CLA, and 48 items for Rep, falling within the range of 0.80 to 1.00. These items were deemed suitable (I-CVI鈥夆墺鈥�0.80) for inclusion in the RESQ听(Table 2). The 魏* score calculations were then computed for all the items (Table 3).

Table 2 Summary of the number of items based on I-CVI scores

Based on Table听3, 42 items for Relevance, 56 items for Clarity and 47 items for Representative were rated as outstanding. Based on Polit and Beck鈥檚 recommendations, only the items that achieve 鈥淥utstanding鈥� shall be included in the instrument [35]. Therefore, the final 42 items constitute the initial RESQ.

Table 3 Item status based on the 魏* score

Face validity

The initial version of the RESQ was distributed to a group of ten participants engaged in recreational sports to evaluate for face validity. Participants were asked to provide feedback regarding choice of words, layout, and choices in the RESQ survey. These steps were carried out to determine any potential issue in comprehension or interpretation of RESQ questions. The collected responses were then systematized into the table, and the proportion of agreement was calculated. The overall percentage for 42 items was 99.6%. Table听4 showed the face validity of the RESQ in which all items were understood and readable by the participants.

Table 4 The face validity of RESQ by ten random participants

Stage three: administration of RESQ

Pilot study

Prior to validation study, a pilot test was conducted to ensure the reliability and validity of the questionnaire. Literature suggested that pilot test sample sizes range from 12 to 100. Consequently, approximately 100 questionnaires were distributed via WhatsApp, Google Forms, and face-to-face invitations, yielding 84 completed responses. The data were analyzed using SPSS version 25.0 for reliability testing. After screening for outliers and inappropriate responses, 17 respondents were eliminated due to straight-lining (5), non-normal outliers (6), and non-responses (6), resulting in 67 valid respondents for the pilot study. The results of the pilot study are shown in Table听5. The results of the alpha coefficient for the pilot study are shown in Table听5 below, with the Cronbach alpha coefficient ranging from 0.71 to 0.92, indicating that all items in the initial Recreational Sport Commitment Questionnaire (RESQ) had good internal consistency.

Table 5 Alpha coefficient range for RESQ

Validation study (administration of RESQ)

The latest version of the RESQ was used and tested on a large group of recreational sports users to assess its reliability and validity psychometrically. The RESQ questionnaire was disseminated either online (e-survey) or face-to-face (paper-and-pencil method). The details collected include demographics (i.e., age, sex, type of recreation/sport activity) which is shown in Table听6. This process involved two distinct phases: recruiting participants for Exploratory Factor Analysis (n鈥�=鈥�275) and Confirmatory Factor Analysis (n鈥�=鈥�313).

Table 6 Demographic information of the participants

Stage four: data analysis

The construct validity of the RESQ was assessed by performing both Exploratory Factor Analysis听(EFA) and Confirmatory Factor Analysis (CFA). EFA was employed at the initial stage to investigate the underlying factor structure of the RESQ and refine the assessment of items during scale development [49, 50]. At this stage, Kaiser-Meyer-Olkin (KMO), which included the standards/requirement for KMO and Barlett鈥檚 measure and Bartlett鈥檚 sphericity test, were examined using factor extraction method using principal axis factoring (PAF) and Promax (Oblique method).

Subsequently, confirmatory factor analysis (CFA) was carried out to examine the observed variables (questionnaire items) as indicators of latent constructs (seven factors in RESQ) as postulated in the theoretical framework. Hair et al. (2010) [51] recommended at least four fit indexes need to be reported to determine the good fit of a measurement model, while Kline (2016) [52] suggested reporting chi-square, RMSEA, CFI, and SRMR. Through CFA, the goodness of fit, which are GFI, Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR), was carefully examined. The recommended threshold values for GFI, CFI, and TLI are 0.90, while RMSEA and SRMR are below 0.08 [53].

Lastly, convergent validity and discriminant validity were calculated. To further assess the convergent validity of the RESQ, we calculated both the Average Variance Extracted (AVE) and the Composite Reliability (CR). The acceptable threshold values for AVE were 0.50 or higher, and CR was set at 0.70 or higher, respectively [54, 55]. The discriminant validity, which aimed to establish the evidence of no association between the constructs, was calculated using the Heterotrait-monotrait ratio of correlations (HTMT). The acceptable HTMT of the measured model should be below 0.90 [56].

Results

Exploratory Factor Analysis

Prior performing EFA, the data was screened to assess outliers and normality. The outliers were detected using Mahalanobis test and skewness and kurtosis were checked. Following the examination of multivariate outliers, 30 outliers were identified and excluded from the study as shown in Table听7. Subsequent tests for normality indicated that skewness (ranging from 鈭掆��1.26 to 0.88) and kurtosis (ranging from 鈭掆��1.62 to 1.93) were within acceptable limits, as shown in Table听8.

Table 7 Value of Mahalanobis distance for the multivariate outlier (n鈥�=鈥�30)

Table 8 Analysis of normality for EFA

Subsequently, the number of factor(s) was determined using parallel analysis due to its robustness in accurately identifying the number of meaningful factors [57]. The measurement used was comparing eigenvalues from EFA with eigenvalues of random data. If the eigenvalues of EFA are larger than the eigenvalues of random data, the factor was retained. As the result, seven factors were extracted (Table 9).

Table 9 Factor extraction using parallel analysis and total variance explained

The EFA was performed using principal axis factoring (PAF), and Promax (Oblique method) was set to perform an EFA on the initial set of items. The Kaiser-Meyer-Olkin (KMO) value should be above 0.80, indicating adequate sampling. Bartlett鈥檚 test of sphericity assesses whether the correlation matrix is not an identity matrix; it should be significant (p鈥�<鈥�0.05) to confirm that factor analysis can be performed. The KMO value with 42 items was 0.89, and Barlett鈥檚 test of sphericity with p鈥�=鈥�0.00. Therefore, the model was deemed to possess sufficient validity. According to Costello and Osborne (2005) [49], five items with loading less than 0.32 and cross-loadings were eliminated [49]. In addition, six items with cross-loading were deleted based on Costello and Osborne鈥檚 suggestion [49]. As a result, seven factors were established with the remaining 31 items. The factors which determined sport commitment were the desire to excel (7 items), appearance (4 items), social support (5 items), personal investment (3 items), sport enjoyment (5 items), social constraint (4 items) and positive health (3 items). The final loading of items is shown in Table听10 below. The 31 items of RESQ were then distributed to the targeted group before performing CFA.

Table 10 The final model for EFA

Confirmatory Factor Analysis

Prior to CFA, the normality of the data was checked. The data was examined using Mardia鈥檚 coefficient to measure skewness and kurtosis. The result of skewness (ranged from 鈭掆��1.54 to 0.48) and kurtosis (ranged from 鈭掆��1.35 to 1.59) indicate that the data was distributed normally (critical ration value is between 鈭掆��1.96 to +鈥�1.96) (Table 11).

Table 11 Analysis of normality for CFA

Then, CFA was performed to validate RESQ and second-order factor analysis was proposed for recreation sport commitment model as a multidimensional construct, comprising both enthusiastic commitment (EC) and constrained commitment (CC). The rationale for using a second-order factor model was driven by the need to represent the hierarchical nature of sport commitment. Specifically, sport commitment was conceptualized as a broader construct influenced by two higher-order dimensions (enthusiastic and constrained commitment) [6, 7], which in turn are each driven by multiple underlying factors. As shown in Fig.听1, the initial model did not show a good model fit in which the GFI did not meet the threshold value. Therefore, a series of modifications were performed, including deletion of items (for factor loading below 0.60 and residual covariance more than 2) and covariance between the items, which was suggested by the AMOS. The final model, or Model 2 (Fig.听2), showed a good model fit for RESQ.

Fig. 1

Initial model of CFA for RESQ

Fig. 2

Final model of CFA for RESQ

The final model showed good model fit with fit data of GFI鈥�=鈥�0.91, CFI鈥�=鈥�0.97, TLI鈥�=鈥�0.97, RMSEA鈥�=鈥�0.04, ChiSq/df鈥�=鈥�1.52 and SRMR鈥�=鈥�0.05.

Discriminant validity, convergent validity and reliability

Table 12 Discriminant validity, convergent validity (AVE) and reliability (CR) of RESQ

Based on Table听12, the value of HTMT for each factor was below 0.9. Therefore, the discriminant validity was established to indicate that the factors were unrelated [56]. On the other hand, AVE constructs in RESQ were more than 0.5, which provides evidence for convergent validity as the corresponding latent variable explained more than half of the variance in the belonging indicators [58]. Lastly, the reliability of the items in all the factors was established as all the CR values had met the cutoff value of more than 0.7.

Sources of enthusiastic commitment and constrained commitment

The result of the CFA showed that sport enjoyment (p鈥�=鈥�0.00, 尾鈥�=鈥�0.66), desire to excel (p鈥�=鈥�0.00, 尾鈥�=鈥�0.93), and social support (p鈥�=鈥�0.00, 尾鈥�=鈥�0.87) were strongly associated with enthusiastic commitment (EC). On the other hand, social constraint (p鈥�=鈥�0.00, 尾鈥�=鈥�0.45), positive health (p鈥�=鈥�0.01, 尾鈥�=鈥�0.30), and appearance (p鈥�=鈥�0.00, 尾鈥�=鈥�0.41) were moderately associated with constrained commitment (CC). However, this is different from the finding of Scanlan et al. (2016) [6], whereby it was suggested that personal investment (p鈥�=鈥�0.06, 尾鈥�=鈥�0.16) was not associated with constrained commitment. This could be due to the several subscales that were included in the CFA analysis, several of which were related to constrained commitment. Given prior evidence, future research should explore the relationship between personal investment and constrained commitment across different demographic groups.

Discussion

This study aimed to develop an instrument to evaluate sport commitment for recreational and leisure sport users. The initial Sport Commitment Questionnaire-2 was created to assess elite athletes, given that studies [6, 7, 59,60,61,62] have indicated the significant relevance of the constructs of the sport commitment for elite adult athletes. Therefore, it is essential to develop an instrument to measure the commitment of sport for recreational sports users, which the RESQ in fulfilling the need to promote sport for all. Initially, 60 items were generated for the RESQ item鈥檚 pool, representing the eight factors contributing to sport commitment for recreational sports users. However, 18 items were eliminated during the experts鈥� evaluation stage. The remaining 42 items constitute eight factors (i.e., sport enjoyment鈥�=鈥�5 items, desire to excel鈥�=鈥�7 items, positive health鈥�=鈥�6 items, appearance鈥�=鈥�5 items, social support鈥�=鈥�6 items, stress management鈥�=鈥�4 items, social constraint鈥�=鈥�4 items, and personal investment鈥�=鈥�5 items).

Subsequently, the Exploratory Factor Analysis (EFA) analysis identified seven factors that collectively explain factors that contribute towards sport commitment. Each factor represents a unique aspect of the construct under investigation, covering the different dimensions contributing to the overall concept of sport commitment. Based on EFA analysis, a total of 31 items with loadings more than 0.32 and did not cross-load were retained based on the criteria suggested by Costello and Osborne [49].

Consequently, the 31-item RESQ was further tested using CFA to allow the test of the relationship between observed variables and their underlying latent constructs. The modifications, such as covarying and deleting items, finally achieved acceptable fit indices for the 26-item RESQ. The RESQ also demonstrated good internal consistency, convergent validity, and discriminant validity. Overall, the confirmatory factor analysis (CFA) results provide strong evidence for the validity and reliability of the RESQ, supporting its use in future research and practice.

The result shows evidence that the psychometric properties of the RESQ are relevant to evaluating the sports commitment among recreational sports users. This indicates that factors such as sport enjoyment, personal investment, social support and positive health play an important role in influencing individuals鈥� commitment to sports activities [6,7,8, 12, 17].

However, our findings also reveal distinct constructs that differentiate the RESQ from other measures, particularly the Sport Commitment Questionnaire 2 (SCQ-2). By emphasizing factors such as positive health and appearance, the RESQ expands upon the existing frameworks for understanding sport commitment. A key difference between the RESQ and the SCQ-2 lies in their factor structures. The SCQ-2 primarily focuses on sport commitment for elite athletes. In contrast, the RESQ encompasses broader dimensions relevant to recreational participants, such as enjoyment and appearance or positive health. This difference provides important insights towards the understanding of sport commitment, indicating that recreational participants stay committed to recreational sport activities due to distinct reason as compared to elite athletes.

Notably, stress management items were excluded from the final version of the RESQ. The decision to omit stress management items stemmed from EFA analysis, indicating that these constructs did not load strongly with the experiences of the target population. One possibility is stress management may not be a primary motivator for recreational participation. This finding aligns with previous literature suggesting that recreational sport users often prioritize enjoyment as important factors in sport commitment [12].

By adopting a second-order factor analysis in this study, it provides a more robust framework for modelling the latent structure of sport commitment, facilitating a deeper understanding of the interrelationships between enthusiastic and constrained commitment and their respective influencing factors. This model helps clarify how the different sources contribute to two distinct types of commitment (EC and CC). This study confirmed that sport enjoyment, desire to excel, and social support are significantly associated with enthusiastic commitment, aligning with previous findings [6, 7]. Conversely, constrained commitment is primarily driven by positive health, appearance, and social constraints [6,7,8, 30]. This suggests that individuals may feel a sense of obligation or limited dedication driven by external factors related to physical health and aesthetic appeal. These results imply that when commitment is influenced by concerns about positive health and appearance, it may not be fully intrinsic or enduring. However, contradicting to the previous studies [6, 7] where personal investment predicts constrained commitment, the findings of this study revealed that personal investment did not significantly predict constrained commitment could be due to one鈥檚 time, energy, and resources devoted to sports participation do not necessarily lead to feelings of obligation or externally driven commitment. Therefore, due to the past evidence, future studies could explore the relationship between constrained commitment and personal investment.

Practically, the development of the RESQ offers valuable insights for recreational sport users by capturing the factors influencing their commitment. This instrument helps recreational users understand their commitment factors and allows sport organizations to tailor programs that reinforce these motivations, thereby enhancing long-term engagement in recreational sports activities. For example, organizations could emphasize enjoyable experiences and health benefits to maintain commitment to sports, potentially enhancing a more active lifestyle. The application of RESQ could possibly extend beyond adult populations to adolescents. The RESQ could serve as an effective instrument to assess commitment factors relevant to this age group. Therefore, examining the RESQ鈥檚 applicability across age groups is essential in future studies.

Limitation and future studies

While the results of this study offer a valid instrument to evaluate sport commitment, it is important to acknowledge the limitations of this study. Firstly, the findings may not be universal across all sports. Therefore, future studies should consider using the RESQ to evaluate specific sports, allowing for a more inclusive understanding of sport commitment within each discipline.

Second, the present study focused on the preliminary validation of the RESQ and did not assess its predictive validity or sensitivity to change over time. Future longitudinal studies could provide the predictive utility of the RESQ in predicting changes in recreational sport participation and explore its responsiveness to interventions aimed at promoting commitment and engagement.

The current study recruited participants 18 years and older as the target population. Although the RESQ was designed for adults, it may also be extended to younger populations or adolescents. Future studies could adapt and validate the RESQ specifically for adolescents to understand the nuances of their sport commitment. These studies could provide educators, coaches, and policymakers with critical insights to develop interventions that foster a commitment to sports and promote holistic well-being among adolescents. By examining the RESQ鈥檚 applicability across age groups, future research can further establish its role in promoting consistent sport engagement and healthy lifestyles.

Testing measurement invariance of RESQ could support the generalizability of RESQ, for example, across sex, age groups and multiculturalism. This is because the measurement invariance test would indicate whether the RESQ provides the same measurement in different groups, ensuring that the observed effects are not the consequence of any measurement bias. However, due to the need for a larger sample size to assess measurement invariance across sex, age and multiculturalism, we could not perform every invariance test in this study. Future analyses conducted on larger samples could determine if RESQ retains its validity across different demographic groups or broader recreational sport populations.

Conclusion

In summary, the development and preliminary validation of the Recreational Sport Commitment Questionnaire (RESQ) offers a reliable and valid instrument for researchers and practitioners to evaluate users鈥� commitment toward recreational and leisure sport, aiding in the comprehension of factors that affect involvement and continuity in sport engagement. By enhancing researchers鈥� comprehension of recreational sport commitment, the RESQ could potentially guide the design of strategies to encourage continued participation in recreational or leisure sport activities, ultimately improving individuals鈥� health and overall well-being (Table 13).

Table 13 Recreational Sport Commitment Questionnaire (RESQ)