Koehn, Stefan (2007) Propensity and attainment of flow state. PhD thesis, Victoria University.

Abstract

In this thesis, I investigated the influence of personality and situational variables on the experience of flow in order to enhance flow state in tennis competition. Based on propositions of the sport-specific flow model (Kimiecik & Stein, 1992), I conducted three interconnected studies. In Study 1, I examined the relationship between personality variables and flow. In Study 2, I tested the effect of the interaction between two key personality variables, trait sport confidence and action control, and key situational variables, self- and externally-paced tasks, on flow state and performance. Finally, in Study 3, I investigated the efficacy of an imagery intervention designed to enhance confidence and action control to increase flow state and self-paced and externally-paced performance in tennis competitions. The purpose of Study 1 was to investigate the influence of personality variables on dispositional flow and state flow in junior tennis players. I entered personality variables, which demonstrated moderate correlations with flow, into regression equations. Except for the Tellegen Absorption Scale (TAS), I entered the Action Control Scale-Sport (ACS-S), the Sport Imagery Questionnaire (SIQ), and the Trait Sport Confidence Inventory (TSCI) as predictor variables into stepwise multiple regression analyses with the Dispositional Flow Scale-2 (DFS- 2; N = 271) and the Flow State Scale-2 (FSS-2; N = 134), respectively, as criterion variables. The results showed that trait sport confidence was the strongest predictor of dispositional flow, accounting for 32.83% of the variance, and action control was the strongest predictor of state flow, explaining 15.52% of the iii variance. On a DFS-2 subscale level, confidence was the main predictor for challenge-skills balance and sense of control, whereas imagery use was the main predictor for clear goals, unambiguous feedback, concentration on the task at hand, and autotelic experience. In the FSS-2 regression analyses, action control was the strongest predictor for most of the entered criterion variables of state flow subscales, namely clear goals, unambiguous feedback, and sense of control. The purpose of Study 2 was to test the Kimiecik and Stein’s (1992) hypothesis that person and situation factors interplay in the generation of flow state. Based on the findings in the previous study, I chose examine interaction and main effects between two key personality characteristics, namely trait sport confidence and action control, and situational variables, such as a self-paced service task and an externally-paced groundstroke task, on flow state and performance in tennis. Following service and groundstroke performance, the participants, junior tennis players (N = 60) between 12 to 18 years, completed the FSS-2. Based on a median split on the TSCI, I assigned participants to groups of high or low confidence. I carried out a two-way repeated-measures ANOVA on flow state with high and low confidence as levels of the independent group factor and self-paced and externally-paced tasks as levels of the repeated measures factor. The results showed a significant main effect between groups of high and low confidence and flow, F(1, 58) = 6.82, p < .05, ç² = .11. The interaction for flow state was not significant, but revealed a moderate effect size, F(1, 58) = 2.64, ns, ç² = .04. I carried out similar ANOVAs on performance showing a significant main effect for performance. Participants demonstrated a greater accuracy in the groundstroke task than in the service task, showing a large effect size, F(1, 58) = 12.74, p < .001, ç² = .18. Analyses of interaction effects between high and low confidence and self- and externally-paced tasks on performance outcome showed a moderate effect size, but was not significant, F(1, 58) = 2.97, ns, ç² = .05. Following the same procedure for action control, I used a median split to divide participants into groups of action orientation and state orientation. There were no significant main or interaction effects between action- and state-oriented groups and flow. With regard to performance, a significant main effect was found for task type, with participants scoring higher on the groundstroke than the service task, and performance outcome, F(1, 58) = 12.13, p < .001, ç² = .17, indicating a large effect size. The purpose of Study 3 was to examine the effect of an imagery intervention on flow state and performance in tennis competition. The study included an A-B design with a baseline and post-intervention phase to evaluate the efficacy of imagery, using a standardised imagery script. I measured flow state and performance over a range of official ranking-list tournaments. I developed the imagery script based on findings of Study 1, taking into account correlational results between personality variables of action control, imagery use, and trait sport confidence and dimensions of flow. The script consisted of three parts, starting with a relaxation component, then imagery on self-paced performance of first and second serves, and, finally, imagery in externally-paced performance situations, including forehand and backhand groundstrokes. For the intervention, four male junior tennis players between 13 and 15 years of age worked with the imagery script three times a week for four consecutive weeks. Participants were of an advanced skill level, being ranked between 203 and 244 in the Australian Junior Ranking List at the beginning of the study. After the four-week intervention phase, all participants demonstrated an increase in service and groundstroke performance winners. In addition, participants increased their ranking-list position from beginning to end of the study between 24 and 145 positions. Visual inspection of the data revealed that three participants increased in state flow intensity across phases. In a social validation interview, which I conducted at the end of the study, three participants confirmed an increase in flow and confidence level after the intervention. Overall, results confirmed several propositions of Kimiecik and Stein’s (1992) sport-specific flow model. Firstly, dispositional personality variables, action control, imagery use, and trait sport confidence demonstrated a moderate relationship with flow. Secondly, significant and near-significant main and interaction effects were evident between situational and personal variables on the experience of flow state. Thirdly, an imagery intervention showed an increase in flow and performance. With regard to future research, I recommend the use of the flow model, as proposed by Kimiecik and Stein (1992), to further assess the influence of personality and situation characteristics and their interaction on flow. In addition, more studies on the flow-performance relationship would be fruitful to enhance theoretical understanding and to inform applied work.

Search Google Scholar

Repository staff login