Invariance Testing of the Adult-Oriented Sport Coaching Survey Across Masters Athletes’ Age, Gender, Competition Level, and Sport

ABSTRACT

The Adult-Oriented Sport Coaching Survey (AOSCS) is a valid and reliable measure of coaches’ and Masters athletes’ perspectives of how often adult-oriented coaching practices are used. However, Masters athletes’ heterogenous traits have been acknowledged as barriers to generalizing research findings on coaching behaviors. Therefore, this study aimed to conduct invariance testing of the AOSCS across groups of Masters athletes based on age, gender, competition level, and sport grouping variables. A sample of 616 Masters athletes (61.9% female, 37.5% male; M_age = 54.47 years, SD = 10.82) completed the AOSCS-A (athlete version) and demographic questions. The results indicated the AOSCS-A demonstrates configural, metric, scalar, and strict invariance across Masters athletes that differed on age, gender, competition level, and sport. This evidence advances the AOSCS-A as an assessment tool by ensuring confidence in the measurement and interpretation of adult-oriented coaching practices reported by Masters athletes, irrespective of age, gender, competition level, and sport.

KEYWORDS:

• Masters sport
• measurement equivalence
• survey development
• coaching adult sport
• assessment

Acknowledgments

Acknowledgment is given to Christine E. Pacewicz for their assistance with our data analyses and Mplus coding.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/1091367X.2023.2203137.

Notes

¹ Motz et al. (2022) used a full factorial structural equation model to examine the cross-sectional associations between MAs’ perceptions of adult-oriented coaching practices and various qualities of their relationship with their coach. Motz et al. (2023) examined changes over time, using latent variable path analyses and difference scores (absolute difference in athletes’ scores over time), in MAs’ perceptions of adult-oriented coaching practices in relation to basic psychological needs satisfaction and thwarting, and in relation to changes in the coach-athlete relationship. Neither study examined invariance – the explicit aim of the present study.

² The respective age to qualify as a MA was vetted in Motz et al. (2023) on a case-by-case basis, based on the primary sport for which the participant reported engagement and cross-referencing this with the minimal age for Masters competition in that sport. Thus, though 14 athletes in the present sample were under 35 years of age, they were legitimately MAs in their respective sports.

³ Notably, for age, gender, and competition level, the Δχ² value was significant when comparing the scalar to metric models but was non-significant in the previous step (i.e., comparing metric to configural models). As an optional step, we attempted to examine partial metric invariance for these groups using the forward method suggested by Jung and Yoon (2016). All partial scalar invariance tests indicated model misspecification for the partial invariance models, thus we could not make inferences about partial invariance based on our age, gender, and competition level comparison groups.

Additional information

Funding

This work was supported by the Social Sciences and Humanities Research Council of Canada under Insight Grant # 227348 held by B.C, B.Y., and S.R