Autonomy-Supportive Coaching and Athletic Performance

Introduction

Not unlike the world of business, many researchers have evaluated the world of sports as a way to analyze whether certain coaching styles are beneficial or detrimental to team performance. The question is a multi-faceted one, and analyzing it in a controlled environment can be difficult. Sample sizes and "apples to apples" comparisons present real challenges. Even so, there are many meaningful points of analysis that can and should be undertaken to gauge the efficacy of a team subjected to the autonomy-supportive coaching approach as opposed to other methods.

The subject of this report is an analysis of whether autonomy-supportive coaching is effective in boosting the performance of the athletes subjected to it. Not unlike business and other leadership contexts, this paper assesses whether a team's coach can improve performance by using the autonomy-supportive method. Team dynamics in sports differ from those in a business environment, and sports teams vary further depending on the game being played — the number of players on the field, the overall strategy required, the degree of physical aggression involved, and so forth.

The purpose of this research is to drill down specifically on sports team dynamics and assess whether the autonomy-supportive method is worthwhile and effective. This question matters because certain coaches — such as Nick Saban in college football and Billy Donovan in college basketball — consistently compete for championships, while other programs remain perennial underachievers. In professional and high-profile college sports, outcomes often come down to talent levels and financial resources. However, examining amateur or high school sports provides a cleaner environment that is far easier to analyze in a controlled and concise fashion.

The guiding argument of this report is whether a team-based environment can produce consistently strong results when certain tactics and habits are adopted by both the athletes and the coach. To that end, the author evaluated players engaged in basketball, hockey, soccer, lacrosse, and football. Performance serves as the dependent variable, while autonomy-supportive coaching serves as the independent variable. The same sports environments are examined both with and without the independent variable — that is, both with and without the coaching method being assessed — so that the difference between the two conditions can be analyzed.

The methodology of this research study can be summarized across several dimensions. All studies are either purely quantitative, purely qualitative, or a blend of the two. This study is a blend; however, quantitative statistics and reports form the primary basis of the results and analysis. That quantitative analysis is then compared and contrasted with commonly accepted, peer-reviewed research findings in the field.

As noted in the introduction, the dependent variable is performance outcomes measured both with and without the independent variable — the autonomy-supportive coaching style and approach. Coach-led environments were evaluated both with and without the autonomy-supportive method applied. The perceptions, opinions, and outcomes from both conditions were compared and analyzed using SPSS and other assessment tools.

The quantitative data were collected through a focus group and a general survey. The survey instrument, release form, and related materials are displayed in the appendices of this report. As noted in the variables section, the data were compiled and analyzed using SPSS. The breakdown of participants by class standing (e.g., freshman, sophomore, etc.) and by sport played is reported in the results. The results include frequencies, percentages, minimum and maximum values, standard deviations, and kurtosis measures, all of which are reported and explained below.

To state the primary finding clearly: no correlation was found between the use of the autonomy-supportive coaching style and subsequent performance outcomes. The Cronbach's alpha reliability of the Autonomy-Supportive Coaching Questionnaire (ASCQ) results was 0.67, and the Test of Performance Strategies (TOPS) score was 0.48 — both indicating fairly low reliability. The full breakdown of participants by class standing is as follows:

Class Rank
Freshman: 13%
Sophomore: 24%
Junior: 28%
Senior: 35%

The breakdown of participants by sport played is as follows:

Sport Played
Basketball: 10%
Tennis: 33%
Hockey: 11%
Soccer: 46%

Regarding gender, there was a 66-to-34 split in favor of men over women. Hockey is typically a sport dominated by women (if not played exclusively by women in many contexts), while the other three sports are generally available to both men and women. The age of the players ranged from 18 to 23, with a standard deviation of 1.22 and an overall mean age of 20.32. In terms of years of experience in their sport, the range was 9 to 18 years, with a standard deviation of 1.71 and a mean of 13.44 years. The kurtosis standard error scores for both age and years in sport were 0.478. The kurtosis statistic for age was -0.368, and for years in sport it was 0.159.

The overall correlation statistics between coaching style and performance showed a two-tailed significance value of 0.746 and a Pearson Correlation of -0.033. As noted above, no correlation was found between the use of the autonomy-supportive coaching style and overall performance, compared with the use of no particular method or an alternative method entirely.

While the survey results are not consistent with what the author expected, it is reasonable to suggest that further research conducted with more homogenous athlete populations and more tightly controlled conditions could yield different findings, and a meaningful relationship between autonomy-supportive coaching and performance outcomes may yet be established. One clear limitation of this study was the wide dispersion of sports included. Analyzing each sport separately, rather than pooling them together, would be a wiser approach. The sports examined differ substantially in terms of the number of players on the field, the overall size of the roster, the typical gender distribution (for example, women rarely play hockey at the levels sampled), and the degree to which individual performance can be distinguished from team performance.

Tennis in particular presents unique challenges for this kind of analysis. It is typically played by a maximum of four people at one time (two teams of two in doubles), and often played one-on-one with no team whatsoever. This stands in stark contrast to basketball, where up to ten players (two teams of five) are on the court simultaneously, or hockey and soccer, where roster sizes and playing configurations vary by level of play and gender.

Method

Gender represents another significant limitation. The survey and focus group pool was skewed 2:1 in favor of men, which is disappointing given that all but one of the sports involved — hockey — is played widely by both men and women. Even within the same sport, the rules can differ by gender: male professional tennis players compete in best-of-five-set matches, while women typically play best-of-three. More broadly, men and women tend to respond differently to the same stimuli on a general level, and sports contexts are no different, as summarized in studies such as Habif et al. (2001) and others. Conroy and Coatsworth (2007) examined boys and girls in a youth sport context and found no discernible difference in outcomes between the genders; however, their subjects were younger than those in the present study. As noted by Klomsten et al. (2004), the physical and psychological differences between men and women become more pronounced beginning in adolescence, meaning that lumping adult men and women together in a single study introduces a meaningful confound.

Another factor that could not be controlled given the structure of this study is the variability among coaches. Every coach is different in how they frame situations, deliver feedback, and implement a given coaching philosophy, as noted by Stevens (2012). Restricting a future study to a single sport, a single coach, and a single gender would eliminate most of the "apples to apples" comparison problems — but it would also produce only one outcome to analyze, and the internal dynamics of that one team could introduce their own complications. For example, if one player dominates ball-handling on a basketball or soccer team, very little may change in measurable performance terms regardless of what the coaching variable does, making the effect of the intervention difficult to detect.

Measuring the effects of coaching style on team dynamics is inherently difficult to quantify — much like measuring leadership effectiveness in a business setting — but a sufficiently large and well-controlled sample could overcome many of these obstacles. At the same time, a very large and diverse sample introduces its own complications. A more focused approach — restricting the sample to men only or women only (or at minimum reporting results separately by gender), keeping the sample as homogenous as possible beyond the variable under examination, and controlling for the presence of dominant individual performers — would produce more interpretable and reliable findings. Insights from Hodge and Lonsdale (2011) on the role of coaching style in shaping athlete motivation further underscore the importance of controlling for these variables in future designs.

In the end, this research was a valuable learning experience. Although the findings were not what the author anticipated, the true relationship between autonomy-supportive coaching and athletic performance may still be out there — and may well differ from what this particular study found. Further research on this topic is warranted, and future studies should exercise much tighter control over variables, sample composition, and sport selection than was achieved here. Despite its limitations, this study may serve as a useful indicator of the methodological considerations that future researchers will need to address. Controlling the sample more carefully and separating results by meaningful subgroups — sport, gender, and coaching context — should yield a much more definitive outcome.

I am conducting a survey about autonomy-supportive coaching strategies in team sports. This survey is being administered as part of a required research assignment. Your responses will be kept anonymous and confidential. By responding to this survey, you are consenting to voluntarily participate. I am grateful for your participation, and it is because of people like you that I am able to conduct this study. Thank you.

Athlete Survey

Please answer the following questions.

1. What year are you in college?
a) Freshman
b) Sophomore
c) Junior
d) Senior

Results

2. What is your gender?
a) Male
b) Female

3. What is your age?

4. What sport do you play?
a) Basketball
b) Tennis
c) Hockey
d) Soccer
e) Football

5. How long have you been playing your current sport?

The following questions relate to team communication in sports. Each coach has a different coaching style, and no one style is necessarily better than another. Think about your most current coaching and team experience. Please indicate how much you agree or disagree with each statement using a scale from 1 (Strongly Disagree) to 7 (Strongly Agree). Please answer honestly based on how you feel about your own team sport experience.

1. My coaches offer me choices about what we do in practice. (AI 1) — 1 to 7

2. My coaches praise me for the things I choose to do in practice. (AB1) — 1 to 7

3. My coaches ask for the team's opinion about what we should do in practice. (AI 2) — 1 to 7

4. My coaches praise me for the things I do in practice. (AB2) — 1 to 7

5. My coaches ask for my opinion about what we should do in practice. (AI3) — 1 to 7