Fuelling the passion: Psychological needs and behavioural regulations as antecedents of passion for football

ABSTRACT

The purpose of the present study was to examine the motivational antecedents of passion for sport by testing a model where players’ satisfaction of the three basic psychological needs was expected to predict their harmonious and obsessive passions through the mediation of autonomous and controlled motivations. Four-hundred eighty-seven elite U18 male football players (M_age = 17.43, SD = 0.71) completed measures of psychological need satisfaction, behavioural regulations and passion for sport. Measurement models were defined using exploratory structural equation models. The results support the model where autonomous and controlled regulations partially mediate the relationship between psychological needs and passion. In this model, the need for relatedness positively and directly predicts harmonious passion and indirectly predicts obsessive passion, which supports its key role in discriminating between the two types of passion. Moreover, autonomy and competence satisfaction present direct and indirect effects on both types of passions. Finally, autonomous regulations positively predict obsessive and harmonious passion, although the strength of the relationship was stronger for the latter. In sum, our results highlight that (a) satisfaction of each need is crucial to promote harmonious passion and (b) both autonomous and controlled regulations contribute to the understanding of how passion is developed in football.

KEYWORDS:

• Motivation
• mediation
• need satisfaction
• passion
• sport

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental data

Supplemental data for this article can be accessed online https://doi.org/10.1080/02640414.2020.1840056.

Notes

¹ Recent research on SDT has provided value for the use of bifactor ESEM and bifactor confirmatory analysis (bifactor CFA) models to test the factorial structure of basic needs satisfaction (Garn et al., 2019; Sánchez-Oliva et al., 2017) and behavioural regulations (e.g., Howard et al., 2018; Litalien et al., 2017). Therefore, in addition to ESEM and first-order CFA models, we also conducted bifactor ESEM and bifactor CFA analyses for these questionnaires. We did not test bifactor CFA and bifactor ESEM models for passion. Results of these analyses are available in Tables B, C, D, E and F of the Supplementary Materials.

² Regarding the selection of the best measurement model for each instrument, we considered different criteria, apart from the fit indexes of each model: (a) the interpretability of the factors obtained (i.e., items significantly loading on its intended factors), (b) the absence of cross-loadings >.30 (or at least the presence of higher loadings on the intended factors than the estimated cross-loadings on non-intended factors), (c) the discriminant validity of each factor (i.e., the absence of high correlations between factors) and (d) the absence of problems of collinearity when the measurement model was included within the structural model.

³ The addition of correlated uniquenesses to improve measurement model fit was only considered when sufficient rationale supported their inclusion. In our case, such rationale came from the content and wording of the items. First, we decided to include the CU involving items 1 (“I decided which activities I practiced”) and 2 (“I had a say on what skills I worked on.”) of autonomy satisfaction. Although the meaning of these two items is not the same, they do share more information between them than with the rest of the items of autonomy satisfaction (i.e., “It was my choice to play football”, “I felt the freedom to do some things my own way” and “I had some choice in what I did”). From our point of view, items 1 and 2 are more specifically related to practice and the other three are more general, more related to the overall experience of playing football. Similarly, there was strong rationale to add a correlated uniqueness in BRSQ (involving items 1 and 3 of identified regulation). The wording of these items is “because the benefits of football are important to me” (item 1) and “because I value the benefits of football” (item 3). Previous studies (e.g., Viladrich et al., 2013) included in the former some examples (e.g., “developing as a player, getting fit, playing with my teammates”), but it was not the case in our study. We believed the meaning and wording of these two items was very similar, which was also supported by the modification indices. Finally, as in previous studies (e.g., Marsh et al., 2013), we included two correlated uniqueness in the measurement model of passion: one between HP items 1 (“Football is in harmony with the other activities in my life”) and 6 (“Football is in harmony with other things that are part of me”), and another between OP items 3 (“Football is the only thing that really turns me on”) and 4 (“If I could, I would only play football”).