Abigail Farr, Jonathan Mietchen, Bryce Tobin, Trammell Cox, and Michael Millard with Dr. Benjamin Ogles, Psychology

This study sought to investigate the relationship between motives for participating in a marathonand pre-race mood and their separate and combined influence on marathon running performance. Our goal was to determine whether or not performance may be dependent upon specific motivations (motivational clusters as determined by previous cluster analyses of marathon samples completing the motivations questionnaire) and pre-race mood among marathon runners. Our hypothesis was that there is interaction between a marathoner’s affective mood state and their specific motivation for running. Furthermore, it may be possible to discover optimal mood states for individuals with specific motivations

Participants were be asked to complete an online survey, the Motivations of Marathoners Scale (MOMS) and other basic information about past running history, prior to the day of the race. At the race, participants were given the 20 item Positive and Negative Affect Scale (PANAS) to complete. After the conclusion of the race, on their own time, participants took a Qualtric’s survey online. This survey contained questions concerning demographic information as well as participants actual and expected race performance. Using previous cluster analyses of the MOMS, the runners were divided in to cluster groups. Cluster groups were then compared in terms of mood and performance. As has been mentioned, a weakness of this study can be found in the administration of the PANAS. Many individuals took the PANAS pre-race, which was desired. Others took the PANAS post-race via Qualtrics survey software. Thirty-one individuals took the PANAS pre-race and post-race. To ensure validity of the measure, a Pearson Bivariate Correlation was conducted. The pre-race and post-race positive scale correlation was 0.90 (p<0.01) and the pre-race and post-race negative scale correlation was 0.85 (p<.01). This gives support to the notion that although the PANAS was administered twice, it was still a reliable measure and the results may still be valid.

After conducting a Pearson Bivariate Correlation of affect (positive and negative) and the nine scales of the MOMS, there were a number of significant correlations. Positive affect was correlated with the self-esteem motivational scale (r=0.33, p<0.01), health orientation motivational scale (r=0.38, p<0.01), personal goal achievement motivational scale (r=0.39, p<0.01), coping motivational scale (r=0.36, p<0.01), and life meaning motivational scale (r=0.32, p<0.01). Negative affect was correlated with the self-esteem motivational scale (r=0.30, p<0.05), and the recognition motivational scale (r=0.27, p<0.05). After conducting a cluster analysis of the MOMS, five clusters were identified, just as they had been in the original MOMS study. Clusters three and five however had a very small number of individuals (n=4, n=8) and thus they were combined with other clusters. After this combination, there were a total of three clusters remaining. There were no significant differences between cluster types in regard to positive and negative affect. This is to say that no individual cluster was significantly more or less positive or more or less negative than any other cluster group.

In the attempt to predict a marathoner’s finish time, there were two variables that could predict after we had controlled for race distance (marathon or half-marathon), and sex. The personal goal achievement scale on the MOMS was a significant predictor of finish time (p<0.05). This motivational scale consists of items regarding improving running speed, competing with oneself, pushing oneself, beating a certain time, and trying to run faster (Ogles & Masters, 2003). This predictor did however become insignificant when miles run per week during training was taken into account. This variable then became the significant predictor of finish time (p<0.05) and the p value of personal goal achievement scale was no longer significant (p>0.05). This could be due to the covariance between the personal goal achievement motivational scale and miles run per week during training. After conducting a Pearson Bivariate Correlation, the two variables did covary significantly (r=0.28, p<0.05).

Participants with positive pre-race affect were correlated with self-esteem, health orientation, personal goal achievement, coping, and life meaning. Participants with negative pre-race affect were correlated with self-esteem and recognition. Despite several pre-race questionnaires being distributed after the race, reliability and validity remained intact. Personal goal achievement could predict race time alone but when miles run during training was taken into account it was the only predictor of race time. However, both co-vary so as an individuals personal goal achievement increased, it is likely that their miles run per week during training also increased. There were no differences between cluster groups in affect and mood could not predict race time for individual clusters.