Attitude Towards Sport Practice: What Makes an Individual Continued Practice of Sport?

Sport practice is conditioned by different factors; previous studies have shown that the attitude towards physical activity and sports practice is largely conditioned by the individual. This study proposes to examine whether the classic model of the theory of planned behaviour reinforces the main motivations for people to adhere to a sport practice. An empirical model based on the classic model of the theory of planned behaviour was proposed, adding three motivational variables to sport practice. A sample was composed of 1008 valid questionnaires collected online in Spain and Colombia. The analysis of the data was performed using the partial least squares technique using the Smart PLS 3.0 statistical programme. The proposed model presents a statistical validity that proves that the practice of physical and sporting activity is determined by the individual’s evaluation of their abilities and perceptions and their relationship with the motivations generated by continuing the practice of physical or sports activity, consequently how adherence is generated.


Introduction
Sports practice and physical activities have been growing in modern societies. This increase is not only related to the social relationships that these activities offer but also to individual factors and personality (Apostolou and Lambrianou 2017).
Studies on behaviour in sports practice have been focused on the area of psychology and to the development of different models, theories, behaviour and motivations; however, there are recent developments in this field. Sporting behaviour already presents a solid description.
A description of the behaviour starts from the search for activities that leads to a positive result in a unique and coherent sense. This process leads to the perception of self-esteem that is the integration of intrinsically motivated and autonomous behaviours; however, excessive control or negative forces can frustrate this behaviour, evoke emotions, or negative feedback. This can be measured by the degree of motivation (Martin S. Hagger et al. 2003).
Studies on sports practice have focused on looking for differences between genders, ages, cultures, types of physical or sports activity, among others (Martin S Hagger et al. 2001;de Bruijn and van den Putte 2012;Debadeep Roy Chowdhury 2012;Beville et al. 2014;Molanorouzi et al. 2014;Heywood 2015;González-Serrano et al. 2017;Lee 2017;Aiken et al. 2018;Cepeda-Carrión and Cepeda-Carrion 2018;Haro-González et al. 2018;Puddle et al. 2018;Summers et al. 2018;Aleksovska-Velickovska et al. 2019;Deelen et al. 2019;Rodríguez Cañamero et al. 2019;Yavuz 2019;Zhou et al. 2019;Aoyagi et al. 2020;Sabiston et al. 2020). For example, regarding differences between age groups in the reasons for practising a physical or sporting activity, it has been found that young people are motivated by affiliation, mastery and enjoyment of the activity, while middle-aged adults are more motivated by psychological condition and other expectations (Molanorouzi et al. 2015). Regarding gender, men choose to do physical activity or exercise more than women for reasons of affiliation and ego (Apostolou and Lambrianou 2017). Regarding the typology in the practice of individual or group sports, no great differences are found, for example, in a study carried out in Korea (Song and Park 2015).
Few studies have attempted to integrate different models of analysis or new relationships, although the statistical validity of the influence of several variables on sports behaviour has been demonstrated (Silva Cortés et al. 2017;Sokolova and Kefi 2020). Especially, as some motivational variables, characteristic of sport is fundamental to individuals practising of physical and sports activities (Silva Cortés et al. 2017;Aleksovska-Velickovska et al. 2019;Aoyagi et al. 2020).
Regarding the sports sector of fitness and wellness services, there is a need to conduct studies that allow us to understand why the individual moves away from sports practice, which is not only based on extrinsic, social or service quality factors, which have been the subject of some studies especially in sports marketing (Lee 2017;Koronios and Kriemadis 2018;Rodríguez Cañamero et al. 2019).
Therefore analysis of the sports consumer is complex and requires close attention to their behaviour, especially due to social change in terms of the use of free time, being single, the personalization of services and other megatrends that do not follow classic patterns of behaviour (Aoyagi et al. 2020). It is for this reason that the main objective of this article is to examine whether the classic model of the theory of planned behaviour reinforces the main motivations for people to practise exercise, analysing what are the variables that affect the individual's perception of the self and the motivations for the practice and adherence to physical and sports exercise (Fig. 1).

Theoretical Framework
Different behavioural theories have been adopted in behavioural research focused on physical and sports activity. The theory of self-efficacy, according to Bandura, reflects an evaluation of the individual's confidence in the performance of a given behaviour, being, therefore, an important predictor of physical exercise and can also allow an evaluation of the perception of confidence and skills in the achievement of the individual's objectives (Bandura 1977;Martin S Hagger et al. 2001). Following this, the theory of planned behaviour arises (TPB) (Ajzen 1991), on the basis that physical activity or sport is conditioned by the intention or attitude, which is a reflection of the motivations to do so, previous experiences, the subjective norm and the perceived behaviour control (PBC) (de Bruijn and van den Putte 2012). Likewise, the main predictor of physical or sports activity is the individual's intention to exercise, which is positively related to the degree of identification of his or her personality towards sports practice and that allows a higher level of adherence long term (Beville et al. 2014;Song and Park 2015).
Under the theory of perceived behavioural control (TPBC), behaviour can be explained more fully; this defines the degree to which an individual believes they have control over physical activity or exercise (Bozionelos and Bennett 1999), the intrinsic and extrinsic variables that influence behaviour are better defined, or even aligned to explain sports behaviour (Bozionelos and Bennett 2009). Ajzen (1991) points out that the perception of behaviour control is determined by external variables, the opportunity for the person to practise physical-sporting exercise, and by variables of an internal nature, such as perceptions of skill and effectiveness to carry out that action (selfregulation, self-efficacy and perceived behavioural control (PBC)). According to Brooks et al. (2017), the theory of behaviour planning (TBP) fails to explain by itself how its constructs lead to an attitude towards sports practice, supported by suggestions that this theory should be a flexible theoretical framework that can incorporate other constructs. From self-determination theory (SDT) (Deci and Ryan 1985;Deci and Ryan 2000), motivations for behaviour are examined, and different scales are derived to measure different types of motivations for the practice of physical and sports activity: the Behavioural Regulation in Exercise Questionnaire (BREQ) (Mullan et al. 1997), (BREQ2) (Markland and Tobin 2004), (BREQ3) (Wilson et al. 2007); the Sport Motivation Scale (SMS) (Vallerand et al. 1992), (SMS-6) (Mallett et al. 2007), (SMS-II) (Pelletier et al. 2013); the Behavioural Regulation in Sport Questionnaire (BRSQ) (Lonsdale et al. 2008); and the most recent complete scale to measure motivation in sports practice is called the Physical Activity and Leisure Motivation Scale (PALMS) (Molanorouzi et al. 2014), which includes eight motives: mastery, enjoyment, psychological condition, physical condition, appearance, others' expectations, affiliation, and competition /ego, and is the most explanatory such us mastery (the search for improvement and excellence of the practice), psychological condition (positive perception of sports practice in improving the mental and emotional condition) and physical condition (perception of improvement in the physical condition due to sports) (Molanorouzi et al. 2015).
Therefore, the predictive variables of behaviour are the TPB variables, while the motivational variables may better explain the adherence to sports practice. Bearing in mind that TPB variables are intrinsic to models of behaviour for the practice of sports, but they need to affect the main motivational variables to generate sports behaviour.
A study by Mickelsson explains this very well giving the example of a gym (Mickelsson 2017), where this should be understood as a system of people, tools and activities. Consumers are committed to this system to bring about physical and mental changes in themselves, that is, the results of practising the exercise motivating factors and finally behaviour. They are the factors that generate the emotions to follow the sport practice; that is to say, it is not the knowledge that generates the sports practice as a direct factor, but rather measurements of achievement are required.
However, Mickelsson (2017) highlights that a high percentage of people tend to behave according to their habits; that is, they do not necessarily reflect on the achievement of their sporting activities.
Therefore, this study aims to verify whether the variables intrinsic to the behaviour of an individual precede the motivations, which ultimately lead to the attitude towards exercise or physical activity, especially its long-term adherence. The following hypotheses are proposed: H1: Perceived behavioural control (PBC) of the individual influences the motivations to practise sport and physical activities. H2: The self-efficacy of the individual influences the motivations to practise sports and physical activities. H3: The self-regulation of the individual influences the motivations to practise sports and physical activities.
H4, H5 and H6: the individual's main motivations to practise sport are the generators of their attitude for their adherence and practice in the long term.

Methodology
The construction of the model was based on the classical variables of the Bandura theory, self-regulation and self-efficacy (Rodrigues et al. 2020); the theory of planned behaviour (TPB), attitude and perceived behavioural control (PBC) (M. S. Hagger et al. 2002), the variable social influence was excluded; and this model was completed with the variables that are most representative of the different motivational sports theories, mastery, physical condition and perceived enjoyment (Molanorouzi et al. 2014;Brooks et al. 2017;Aleksovska-Velickovska et al. 2019). A psychometric test was applied following the process for transcultural translations and adaptations proposed by Wild et al. (2005). A pre-test of the measurement tool was conducted (see Annex 1, Table 7).

Sample
The sample collection process was determined in a non-probabilistic manner to obtain the largest number of completed questionnaires. For this purpose, respondents were motivated to participate in the study by being given a purchase voucher for a high prestige sports store. Data were collected according to the ethical and legal aspects informed to the participants. For both countries, the questionnaire was distributed in electronic form in the period between June and December 2018; in Spain a total of 459 questionnaires were completed and for Colombia a total of 549 questionnaires. The final sample was 1008 valid questionnaires between the two countries. The sample highlights that a high percentage of participants belong to the millennial generation and that women participated in a greater proportion than men; however, no differences were found that biased or moderated the data with any of the demographic variables ( Table 1).

Analysis of the Data and Results
The method of analysis used was the partial least squares (PLS) technique, given that the model is exploratory and the relationships between the variables have not been supported in previous studies (Hair et al. 2014).

Validation of the Measurement Model
The first step in the analysis was to examine the convergent and discriminant validity of all the variables to determine the reliability of the measures of the items. Regarding convergent validity, all the results had correlation loads higher than 0.505 with significance levels lower than p value 0.001, meeting this test (Hair et al. 2014) ( Table 2). The other measures of convergent validity were also accepted ( Table 3). The Dillon-Goldstein coefficient of reliability value was greater than 0.70 for all variables (Gefen et al. 2000). Cronbach's alpha test presented values well above 0.70 (Churchill and Iacobucci 2004), and the analysis of variance showed values greater than 0.50 (Henseler et al. 2014).
The divergent validity was verified by means of two tests, the first one by comparing the AVE value of the variables with the correlation of the constructs with respect to each variable elevated to the square, being inferior to the square root of the AVE, demonstrating that each variable is related more strongly to its items than to the other variables (Fornell and Larcker 1981) (Table 4). The second Henseler-Ringle test shows that the values are below 0.90, acceptable therefore to validate the entire tool in terms of divergent validity (Henseler et al. 2014) ( Table 5).

Validation of the Structural Model
For the validation of the structural model, the predictive capacity of the model was first examined; the value of R2 for the dependent variables was greater than 0.1, especially for attitude towards sports practice (R2 = 0.503) which allows to positively verify that the model establishes the causality between its variables. To establish the statistical significance of the model, the technique of resampling or bootstrapping was applied, taking 5000 sub-samples in order to give greater statistical rigour (Hair et al. 2014) (Table 6).

Discussion
The results of the study at a general level without differentiating sports show an interesting cause-effect relationship between classical factors of human behaviour that were only examined directly on attitude; however, in this study, it was shown that these intrinsic variables generate other more specific motivating variables that enhance the attitude towards practising sport or physical activity, that is, to adhere to sports practice.  Specifically, hypothesis H1 was statistically validated with especially a higher value of effect of the perceived behavioural control (PBC) variable on physical condition H1b: B = 0.179; that is, those who have a high degree of control of physical exercise will tend to consider that sport generates physical and health improvements. Similarly for H2, the effect of the "self-efficacy" variables being greater on the motivation "mastery" H2a: B = 0.375, those who have greater confidence in sports practice will tend to consider that sport generates perfection in its achievement. H3 is supported regarding the impact of "self-regulation" on motivations, especially with "mastery" H3a: B = 0.375 and "perceived enjoyment" H3c: B = 0.366, whereby those who plan and control their physical activity tend to consider that sport generates enjoyment and improvement in the achievement of their objectives (Table 6) (Fig. 2).
The results indicate, as suggested (Funk 2017), that sports behaviour factors are the sports user (needs); the sports context (experience of use); and sports organization (business objectives); being, therefore, a very complex model that cannot be minimized or generalized in only one of the theories, this is where personalization in the offer of sports services and support are fundamental especially for achieving adherence to the practice of physical and sporting exercise in the long term (Joachim et al. 2020).

Conclusions, Limitations and Future Lines of Research
The theoretical implications of this article lie in the exploration of different relationships between the variables of the behavioural and motivational models that have been analysed separately and that the findings indicate that sports behaviour is complex and requires that many variables be taken into account, both intrinsic as well as extrinsic.
The managerial implication of this study requires that the management and planning of offers of services for exercise and sports should take into account all the aspects that contribute to the behaviour of the sports consumer.
Regarding intrinsic factors, sports centres can accommodate users with high PBC variables, self-efficacy and self-regulation, for example, by offering services with broad hourly coverage, good access to facilities and sports practice, flexible rates and motivational factors such as company, personalized training, the establishment of achievements, prizes and bonuses for the achievement of objectives, a variety of activities, and support services are necessary for users to continue their sports routine.
Here it is clear that marketing actions focused on promotions and attracting new users may not be effective even in the medium term, since, if the essential reasons for membership are not generated, the desertion rate will be high.
The limitations of this study are considerable, principally, that it is an initial exploration that requires further verification; likewise, it is necessary to study other populations and cultures and integrate other variables that were not taken into account, such as social influence, among others.

Compliance with Ethical Standards
Conflict of Interest The authors declare that they have conflicts of interest.
Informed Consent All individuals have individual rights that are not to be infringed.
Consent to Participate Informed consent was obtained from all individual participants included in the study.