MOOD AND PERFORMANCE RELATIONSHIPS AMONG PLAYERS AT THE WORLD STUDENT GAMES BASKETBALL COMPETITION.

Lane, Andrew M. ; Chappell, Robert C.

The present study examined the relationship between pre-competition mood and performance at the World Student Games basketball competition. The purpose of the study was to compare mood and performance relationships using an idiographic (within-subject) design with mood and performance relationships using a cross-sectional (between-subject) design. Players from the United Kingdom basketball team (N = 10) completed the 24-item Profile of Mood States-Adolescents (Terry Lane, Lane, & Keohane, 1999,) 1 hour before competition for eight games. Participants completed a Performance Satisfaction questionnaire 1 hour after each game. Results indicated that the predictive power of mood was highly individualized. Within-subject analysis identified five players whose mood significantly related to performance. Data from these five players were grouped, and regression analysis on this group indicated that mood significantly predicted 40% of performance variance. By contrast, mood showed no association with performance in t he five other players. Multiple regression on data from all 10 players indicated mood accounted for only 9% of performance variance. We suggest that future research is needed to investigate factors such as personality and/or previous experiences that might account for individual differences in the strength of mood-performance relationships.

There has been a great deal of research interest in the influence of pre-competition mood on performance (see Beedie, Terry, & Lane, 2000; Hanin, 1997; Lane & Terry, 2000a; Prapavessis, 2000; Renger, 1993; Rowley, Etnier, Landers, & Kyllo, 1995; Terry, 1995). Recent research has suggested the nature of such relationships is highly individualized (see Hanin, 1997; Prapavessis, 2000). Research to investigate the predictive power of mood has typically used a cross-sectional design. This design precludes the examination of the extent to which such relationships vary between individuals. The present study investigated pre-competition mood among a sample of basketball players at the World Student Games. The purpose of the study was to compare mood and performance relationships using an idiographic (within-subject) design with mood and performance relationships using a cross-sectional (between-subject) design.

Research has typically used the Profile of Mood States (McNair, Lorr, & Droppleman, 1971, 1992) to assess mood. It has been suggested that successful sport performance is associated with above average scores (in comparison to normative data for a US student population) for Vigor coupled with below average scores for Anger, Confusion, Depression, Fatigue, and Tension. Morgan (1985) termed this the iceberg profile due to its shape when depicted graphically with the population average representing the water-line.

Research findings using a cross-sectional design to examine mood-performance relationships in basketball have been unclear. Craighead, Privette, Vallianos, and Bykrit (1986) found that mood could not discriminate between starters and non-starters over a season. Contrastingly, Newby and Simpson (1994) found a significant relationship between mood and performance. Newby and Simpson (1994) found that a player's mood at the start of the season predicted the number of minutes played during the season. More recently, in a study of seven professional basketball players over eight games, Hoffman, Bar-Eli, and Tenenbaum (1999) found that the mood-performance relationship was weak.

Recent research has suggested that the equivocality of mood-performance relationships may be a reflection of individual differences within the sample (see Hanin, 1997 for review). Terry (1992) reported scores on the POMS for two successful athletes. Athlete A reported mood scores that showed a classic iceberg profile, while Athlete B reported below average Confusion, Depression, and Anger with scores for Tension and Vigor on the 50th percentile. Terry (1995), in reviewing his work emphasized the point that a trend found from normative samples, such as an iceberg profile does not always apply to individuals.

Han in and Syria (1995, 1996) found that some athletes perceived pre-competition moods such as tension as facilitative of performance, while other athletes perceived tension as debilitative of performance. Hanin (1997) suggested that the extent to which athletes perceive mood as either facilitative or debilitative of performance is highly individualized. Thus, using a cross-sectional research design to examine such relationships is likely to produce a weak relationship.

If mood and performance relationships are individualized, this presents a serious limitation for testing theoretical links using a cross-sectional design. For example. consider the following hypothetical results. A sample might comprise 33% of individuals showing tension was associated with debilitative performance (r = -.1). It also might comprise 33% of individuals in the sample showing tension was associated with facilitated performance (r = +.l), with the remaining 33% of the sample reporting zero for each tension item (r = 0). When analyzed using a cross-sectional research, the results would show that tension has no relationship (r = 0). Thus, to elucidate the influence of mood on performance, it is suggested that there is a need for idiographic research.

Several review articles have indicated that mood research has suffered from a number of methodological inconsistencies (Beedie et al., 2000; Lane & Terry, 2000a; Prapavessis, 2000; Terry, 1995). Indeed Beedie et al. argued that inconsistent, and sometimes poor methodologies were the main reasons why only 25 studies were included in their meta-analysis despite there being over 300 papers published in sport that have used the POMS (LeUnes & Burger, 1998; LeUnes, 2000). For research to test theoretical links, a consistent methodology should be used otherwise it is possible that different findings between studies could be ascribed to the varying methodologies used. An adoption of common methodologies would allow research to test the consistency of findings. It is suggested that through the revision of ideas that come about in the light of such comparisons, the mechanisms that underlie the influence of mood could be better understood.

In the present study, we use recommendations made from previous research to direct some of the methods used (Beedie et al., 2000; Lane & Terry, 2000a; Prapavessis, 2000; Terry, 1995). The first is that we used the 'right now' response set as the instructional set through which participants completed the POMS (or a version of it). Lane and Terry (2000a) argueq that mood is a transitory construct that varies as a function of a person-environment transaction. This conceptualization of mood is consistent with mood research in general psychology (Batson, Shaw, & Oleson, 1992; Ekman & Davidson, 1994; Morris, 1992; Parkinson. Totterdell, Briner, & Reynolds, 1996). Lane and Terry (2000a) argued that the 'right now' response set should be used for prediction purposes in order to reflect the mood at tile time of the competition

Second, we assessed performance using a self-referenced measure. Several researchers have argued that such an approach is necessary in order to assess the relatively subtle influence of mood on performance (Beedie et al., 2000; Lane & Terry, 2000a; Prapavessis, 2000; Terry, 1995). This point seems particularly important in team games where the outcome (win/loss/draw) is influenced by the performance of a number of individuals.

Collectively, there is a need for well-conducted research to investigate mood and performance relationships. In the present study, we examine the relationship between pre-competition mood and performance using an idiographic design, and compare mood-performance relationships with a cross-sectional design.

Method

Participants

Participants were male basketball players selected for United Kingdom university team at the 1997 World Student Games (N = 11, M = 21.36 years, SD = 1.50 years). Participants had been playing an average of 7 years (SD = 3.12 years). Selection to the National Student Team was done through a trial system. All participants were university students. One player only played in two games; therefore his data are excluded from the study.

Measures

Mood. Mood was measured using the 24-item Profile of Mood States-Adolescents (POMS-A: Terry, Lane, Lane, & Keohane, 1999). The POMS-A is a 24-item inventory that assesses the same constructs as the original POMS (McNair et al., 1971, 1992). Anger items include "Bad-tempered" and "Angry", Confusion items include "Mixed-up" and "Uncertain", Depression items include "Depressed" and "Downhearted", Fatigue items include "Worn out" and "Tired", Tension items include "Anxious" and "Panicky", and Vigor items include "Alert" and "Energetic". Items are rated on a 5-point scale anchored by "not at all" (0) to "extremely" (4).

Validation of the POMS-A involved 3,361 participants ranging in age from 12-39 years (Lane & Terry, 2000b; Terry et al., 1999). Confirmatory factor analysis supported the factorial validity of a 24-item six-factor model using both independent and multisample analyses. In addition, the POMS-A has demonstrated concurrent validity with correlations between POMS-A scores with previously validated inventories.

Scores on the POMS-A are transformed into standard T score format (M = 50; SD = 10) from normative data from athletes reported by Lane and Terry (1999). It should be noted that normative mood data on athletes differs from the original normative data on students (McNair et al., 1971). Using the original POMS, Terry and Lane (2000) showed that an iceberg profile was the normal mood for athletes. Terry et al. (1999) reported the same trend for POMS-A data for differences between athletes and students in a British population. The implication of this is that T-scores of 50 on all six POMS-A subscales are indicative of mood scores previously described as an iceberg profile on the original POMS.

The POMS-A was chosen as the measure of mood for three reasons. First, it was validated on samples of athletes whereas other mood scales, such as the original POMS (McNair et al., 1971) were developed for student or psychiatric populations. Second, it was validated for use with a British population, hence an appropriate group to the sample used in the present study. The original POMS has a North American orientation with items such as 'Blue' and 'Bushed' which are expressions not commonly used in the United Kingdom. Third, as mood was assessed shortly before competition, brevity was an important consideration. Given that completion time is a function of the number of items and the difficulty of items, the POMS-A was judged to be the most appropriate available measure.

Measures of Performance. In accordance with recent recommendations. performance was assessed using a self-referenced measure (see Beedie et al., 2000; Prapavessis. 2000). In the present study, each participant rated his performance in competition in response to two items: "How do you feel about your performance in the last game" and "I-low did your performance relate to your pre-game expectations? Items were assessed on a nine-point scale anchored by "extremely dissatisfied" (1) and "extremely pleased" (9), and "extremely poor" (1) and "extremely well" (9). Internal consistency coefficients indicated an internally consistent scale (alpha = .87).

Procedure

The team manager who also is the second author on this paper administered the questionnaires. Players were informed that they were part of a research program to examine players' thoughts and feelings before and after competition. As a strategy to promote honesty, players were informed that they did not have to attach their name to the questionnaire. It was emphasized that the players' answers would be treated in strict confidence. All players attached their name to the questionnaires. Participants completed the POMS-A (Terry et al., 1999) approximately 1 hour before competition and completed the self-referenced performance measure approximately I hour after competition.

The questionnaires were completed before and after eight games. The first two were preliminary games, the next four were group games, and the final two were second round games having qualified for the next round of the tournament.

Data Analysis

Mood and performance relationships were examined using an idiographic design (within subject) and a cross-sectional (between subject) design. For each design, two types of statistical tests were used. First, correlation was used to investigate mood and performance relationships for each individual over the eight games played. Cross-sectional relationships were examined by correlating mood and performance scores for each game. In accordance with suggestions from recent research (Schutz & Gessaroli, 1993), greater emphasis was placed on the size of the correlation rather than the significance. A limitation with reliance on statistical significance alone to interpret the meaningfulness of findings is that sample size has a considerable influence on the probability value. This means that the higher the sample size, the smaller the correlation has to be for it to be significant. In the present study, data from ten players (idiographic design) and eight games (cross-sectional design) resulted in low statistical po wer. Further, as data were collected at the World Student Games it was difficult to collect more data as the team used in the present study played only eight games. We decided not to collect more data, as the level of competition would be different from World Championship level. Terry (1995) suggested that researchers should investigate mood and performance in homogeneous samples. We suggest that a change in the level of competition might have compromised the relative homogeneity of the data.

Vincent (1995) reported that as a general rule, an r value of .5 to .7 is considered low, .7 to .8 moderate, and .9 or higher to be high. Using an alpha level of .05, in the idiographic analysis (df = 7), the r value had to be greater than .67 to be significant. For the cross-sectional analysis, (df = 9), the r value had to be greater than .57 to be significant at the .05 level. Therefore, we extended consideration for significance to the [less than].10 level to be consistent with the likely effect size. We acknowledge some researchers might question the approach of conducting a number of univariate analyses as this inflates the risk of a Type 1 error. We also accept that some researchers might question the validity of our findings. It is argued that lower alpha levels would not necessarily increase the validity of findings. It is suggested that researchers test the reproducibility of findings from the present study to a different sample.

In the present study, we used the meta-analysis results of Beedie et al. (2000) as a guide to establish the likely strength of the association between mood and performance, and to set an alpha level accordingly. Beedie et al. (2000) reported the following effect sizes (ES) for groups that could include basketball: Open skilled sports (ES = 0.39), team sports (ES = .30), and long duration sports (ES = 0.26). Thomas and Nelson (1996) argued that an effect size of [greater than] 0.8 is large, around 0.5 is moderate, and [less than]0.2 is small. Thus, Beedie et al. found that the likely effect of mood on performance was moderate to small.

The second method used to analyze data was multiple regression. To investigate mood and performance using a cross-sectional design, data were merged into a single file. This yielded 76 cases of data, which satisfies the case number to the number of independent variables needed for multiple regression. In the present study, the ratio was 7.6:1, hence greater than the 5:1 minimum ratio recommended by Tabachnick and Fidell (1996).

Violation of the case number to the number of independent variable precluded investigating idiographic multiple regression analyses to predict performance for each individual. However, the research strategy was to use correlation results to identify, and then group players into one of two groups. Group I comprised individuals who showed significant mood-performance correlations (Significant relationships group). Group 2 comprised individuals who showed no significant mood-performance relationship (Non-significant relationships group). Multiple regression was used to investigate the strength of mood and performance relationships in the two groups.

Results

Descriptive statistics for each individual are contained in Table 1. Mood and performance relationships for each individual are contained in Table 2. As Table 2 indicates there were individual variations in the strength of relationships for all six mood dimensions. Of 60 correlations between mood and performance conducted, there were eight significant relationships. Pre-competition Anger was associated with debilitated performance in one player (player 8), with relationships near significance in another two players (players 4 & 6). Four players (players 1, 3, 9, & 10) scored zero for each Anger-item before all eight games (see Table 1) and therefore it was not possible to test the Anger-performance relationship. Depression was associated with debilitated performance in two players (player 5 & 7), and close to significance in one other (player 6). Tension was associated with debilitated performance in three players (players 1, 5, & 9), and close to significance in one player (player 8, see Table 1). Vigor was associated with facilitated performance in one player (player 6). Confusion and Fatigue were not significantly associated with performance in any of the ten players.

Mood and performance relationships for each of the eight games are contained in Table 3. As Table 3 indicates, Anger was associated with debilitated performance in Game 4. Confusion was associated with debilitated performance in Game 6. Fatigue also was associated with debilitative performance in Game 6. Vigor was associated with facilitated performance in Games 6 and 8. Depression and Tension did not significantly correlate with performance in any of the eight games.

Multiple regression to predict performance from mood indicated that only 9% of the variance was explained (Adj [R.sup.2] = .09, p [less than] .05). Anger (Beta = -.25. p [less than] .05) was the only significant predictor of performance, indicating that low anger scores tended to be associated with perceptions of successful performance.

To assess the extent to which mood and performance relationships were individualized two further multiple regression analyses were conducted. Using correlation results in Table 2, data were grouped according to whether an individual reported a significant mood and performance relationship. This led to a significant mood-performance relationships group (players 1, 5, 7, 8, and 9), with the remaining players in the non-significant mood-performance relationships group (players 2, 3, 4, 6, and 10). Player six was a marginal case as his correlation results showed significance at the p [less than].10 level. We conservatively placed player 6 in the non-significant relationships group. This resulted in 45 cases of data in Group 1 and a case to independent variables ratio of 4.5:1 and 35 cases of data in Group 2, and a ratio of 3.5:1.

In the significant relationships group, multiple regression results indicated that mood significantly predicted 40% (Multiple R = .69, Ad] [R.sup.2] = 4O, p [less than].01) of the variance in performance. Significant results for independent mood dimensions indicated that as performance was viewed negatively, Anger (Beta = -.41, p [less than].01) and Confusion (Beta = -.39, p [less than].05) tended to increase. By contrast in the non-significant group, multiple regression accounted for 0% of the variance. MANOVA to compare differences in the intensity of mood between these groups indicated no significant difference (Wilks' [lambda.sub.6,67] = .89, p [greater than].05).

Discussion

The present study examined mood and performance relationships in basketball at the World Student Games. Results show tentative support for the hypothesis that the predictive capability of mood would be highly individualized. The cross-sectional analysis of mood and performance relationships indicated that only 9% of the variance in performance was explained by mood. Further analysis indicated that the sample comprised five individuals in whom variations in mood were associated with variations in performance, and five individuals in who mood was independent of performance. Multiple regression indicated that mood predicted 40% (Group 1) of performance variance in the significant group, and 0% of performance variance in the non-significant group (Group 2). Dividing the sample into two groups served to illustrate that mood relates to performance in some individuals and not in others. These results indicate that there is a great deal of variation in mood and performance relationships even among players from the s ame team in the same sport.

Findings of the present study are consistent with the theoretical proposals of Hanin (1997) suggesting that mood-performance relationships are highly individualized. However, in contrast to Hanin (1997), who proposed that researchers should identify zones of optimal functioning for mood-performance relationships, the present study identified individuals whose performance is more mood-dependent than others.

A direct comparison between findings of the present study with previous studies that has examined mood and basketball performance relationships is difficult. It is difficult because of the varying methodologies used (Craighead et al., 1986; Newby & Simpson, 1994). As each study used the POMS model of mood and a performance measure, some researchers might argue that this is a trivial point. We suggest that an inappropriate methodology can compromise the testing of theoretical links between mood and performance. For example, in the studies by Craighead et al. (1986) and Newby and Simpson (1994), mood was assessed at the start of the season, and this measure was used to predict performance over an entire season. This conceptualization of mood is more in keeping with a trait approach than a state approach, and mood is clearly a state construct. Trait theory assumes that the stable underlying construct will predict behavior when behavior is averaged over time (see Matthews & Deary, 1999). Further, trait theory pr oposes that a trait should show a relatively weak relationship with a single measure of behavior. There is a great deal of theoretical and empirical support demonstrating that mood is a transitory construct (Batson et al., 1992; Ekman & Davidson, 1994; Morris, 1992; Parkinson et al., 1996). Consistent with this conceptualization, mood is likely to be influenced by factors within the environment, and as a consequence, mood is likely to relate to the behavior in that situation only.

At least two factors might influence mood and performance relationships. It is suggested that personality type might be a factor that moderates the predictive power of mood. There is a great deal of research suggesting that mood and personality is related in the general psychology literature (Costa & McCrae, 1980; McFatter, 1994; Meyer & Shack, 1989). A general consensus among research findings is that extroversion is related to positive mood and neuroticism is related to negative mood. Although application of these findings to sport seems plausible at first, the proposal that neuroticism is related to moods such as anger and depression does not help explain findings suggesting that anger is associated with facilitated performance in some individuals but is associated with debilitative performance in others (see Hanin, 1997; Lane & Terry, 2000a). Further, there has been very little research to examine the influence of personality on mood changes in sport. Prapavessis and Grove (1994) used a battery of person ality tests and found that personality did not predict changes in mood, hence did not support findings from general psychology. Given the exploratory nature of Prapavessis and Grove's study, there is need for further research in this area.

A second possible explanation to account for individual differences in the mood-performance relationship is through the proposal that individuals learn to develop optimum performance states. In two articles, Hanton and Jones (1999a, 1999b) reported that athletes learn to view pre-competition anxiety as a necessary precursor to a good performance. However, the small sample used in these studies make generalizations of these findings to a larger population difficult (a similar argument applies to findings from the present study). Therefore, it is not known if it is possible for all individuals to learn to view anxiety as facilitative of performance, or that the findings of Hanton and Jones (1999b) might be explained by personality characteristics of the athletes in their sample. We suggest that future research should attempt to identify personality factors associated with mood changes in sport, and the impact of mood on performance.

Findings from the present study have implications for sport psychology practitioners. Previously, negative mood states above the 50th percentile have been interpreted as potential indicators of poor performance. In the present study, results show that there was no significant difference in the intensity of mood between individuals who show a significant moodperformance relationship and individuals who do not. Consequently, it is not possible to identify performance threatening moods from the intensity of the mood alone. A high negative mood score could be threatening to performance in individuals whose mood has been found to consistently relate to performance. By contrast, a similar mood score from an individual whose mood does not consistently relate with performance would indicate that there is no need to implement an intervention strategy. However, before sport psychologists use findings from the present study to guide their work, we suggest that there is need to verify findings from the present study in another sample.

In conclusion, the present study found that mood-performance relationships were highly individualized with significant relationships being found in five players. When analyzed collectively using a cross-sectional design, the predictive power of mood on basketball performance was weak. Findings showing that mood and performance relationships vary between individuals question the utility of mood as a predictive construct. We suggest that there is a need for future research to examine the interaction between personality and situational variables in predicting mood changes. It also is suggested that there is need to examine if these variables have a moderating effect on the predictive power of mood.

References

Batson, C. D., Shaw, L. L., & Oleson, K. C. (1992). Differentiating affect, mood, and emotion: Toward functionally based conceptual distinctions. In M. S. Clark (Ed.), Emotion (pp. 294-326). Newbury Park, CA: Sage.

Beedie, C. J., Terry, P. C., & Lane, A. M. (2000). The Profile of Mood States and athletic performance: Two meta-analyses. Journal of Applied Sport Psychology 12, 49-68.

Biddle, S. J. H., & Hill, A. B. (1991). Relationships between attributions and emotions in a laboratory-based sporting contest. Journal of Sports Sciences, 10. 65-75.

Costa, P. T., & McCrae, R. R. (1980). Influence of extroversion and neuroticism on subjective well-being: Happy and unhappy people. Journal of Personality and Social Psychology 38, 668-678.

Craighead, D. J., Privette, G., Vallianos F., & Byrkrit D. (1986). Personality characteristic of basketball players, starters, and non-starters. International Journal of Sport Psychology, 17, 110-119.

Ekman, P., & Davidson, R. J. (1994). The nature of emotion. Oxford: Oxford University Press.

Hanin, Y. L. (1997). Emotions and athletic performance: Individual zones of optimal functioning model. European yearbook of sport psychology, 1, 29-72.

Hanin, Y., & Syrja, P. (1995). Performance affect in junior ice hockey players: an application of the individual zones of optimal functioning model. The Sport Psychologist, 2, 169-187.

Hanin, Y., & Syrja, P. (1996). Predicted, actual, and recalled affect in Olympic-level soccer players: Idiographic assessments on individualized scales. Journal of Sport & Exercise Psychology, 18,325-335.

Hanton, S., & Jones, G. (1999a). The acquisition and development of cognitive skills and strategies: I. Making the butterflies fly in formation. The Sport Psychologist. 13, 1-21.

Hanton, S., & Jones, G. (1999b). The effects of a multimodal intervention program on performers: II. Training the butterflies to fly in formation. The Sport Psychologist, 13, 22-41.

Hoffman, J. R., Bar-Eli, M., & Tenenbaum, G. (1999). An examination of mood changes and performance in a professional basketball team. Journal of Sports Medicine and Physical Fitness, 39, 74-79.

Lane, A. M., & Terry, P. C. (2000a). The nature of mood: Development of a conceptual model with a focus on depression. Journal of Applied Sport Psychology 12, 16-33.

Lane, A. M., & Terry, P. C. (2000b). Construct validity of the Profile of Mood States-A for use with adult athletes. (Manuscript in preparation).

LeUnes, A., & Burger, J. (1998). Bibliography on the Profile of Mood States in sport and exercise, 1971-1995. Journal of Sport Behavior, 21, 53-70.

LeUnes, A. (2000). Updated bibliography on the Profile of Mood States in sport and exercise psychology research. Journal of Applied Sport Psychology 12, 110-113.

Matthews, G., & Deary, I. J. (1999). Personality traits. London: Wiley.

McFatter, R. M. (1994). Interactions in predicting mood from extroversion and neuroticism. Journal of Personality and Social Psychology 66, 570-578.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1971). Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Services.

McNair, D. M., Lorr, M., & Droppleman, L. F. (1992). Revised Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Services.

Meyer, G. J., & Shack, J. R. (1989). Structural convergence of mood and personality: Evidence for old and new dimensions. Journal of Personality and Social Psychology 57, 691-706.

Morgan, W. P. (1985). Selected psychological factors limiting performance: A mental health model. In D. H. Clarke & H. M. Eckert (Eds.), Limits of human performance (pp. 70-80). Champaign, IL: Human Kinetics.

Morris, W. N. (1992). A functional analysis of the role of mood in affective systems. In M. S. Clarke (Ed.), Emotion. (pp. 257-293). Newbury Park: Sage.

Newby, R. W., & Simpson, 5. (1994). Basketball performance as a function of scores on Profile of Mood States. Perceptual and Motor Skills, 78, 1994, 1142.

Parkinson, B., Totterdell, R, Briner, R. B., & Reynolds, S. (1996). Changing moods: The psychology of mood and mood regulation. London: Longman.

Prapavessis, H. (2000). The POMS and sports performance: A review. Journal of Applied Sport Psychology 12, 34-48.

Prapavessis, H., & Grove, R. J. (1994b). Personality variables as antecedents of precompetitive mood state temporal patterning. International Journal of Sport Psychology, 22, 347-365.

Renger, R. (1993). A review of the Profile of Mood States (POMS) in the prediction of athletic success. Journal of Applied Sport Psychology 5, 78-84.

Rowley, A. J., Landers, D. M., Kyllo, L. B., & Etnier, J. L. (1995). Does the Iceberg Profile discriminate between successful and less successful athletes? A meta-analysis. Journal of Sport and Exercise Psychology 16, 1 85-199.

Schutz, R. W., & Gessaroli, M. E. (1993). Use, misuse, and disuse of statistics in psychology research. In R. N. Singer, M. Murphy, & L. K. Tennant (Eds.), Handbook of research in sport psychology (pp. 901-921). McMillan, NY.

Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics. New York, NY: Harper and Row.

Terry, P. C. (1992, July). Mental training and psychological support for the British bobsleigh team, 1989-1994. Paper presented at the Olympic Scientific Congress, Malaga, Spain.

Terry, P. C. (1995). The efficacy of mood state profiling with elite performers. A review and synthesis. The Sport Psychologist, 9, 309-324.

Terry, P. C., & Lane, A. M. (2000). Normative values for the Profile of Mood States for use with athletic samples. Journal of Applied Sport Psychology 12, 93-109

Terry, P. C., Lane, A. M., Lane, H. J., & Keohane, L. (1999). Development and validation of a mood measure for adolescents: POMS-A. Journal of Sports Sciences, 17, 861-872.

Thomas, J. R., & Nelson, J. K. (1996). Research methods in physical activity. Human Kinetics: Champaign, IL.

Vincent, W. J. (1995). Statistics in kinesiology. Champaign, IL: Human Kinetics.

Authors' Acknowledgments

The authors would like to thank Christopher Beedie and Matthew Stevens for their useful comments made in the preparation of this study.

 Descriptive Statistics for Pre-
 competition Mood
Player/ Anger Confusion Depression Fatigue
Game M SD M SD M SD M
Player 1 44.52 0.00 44.94 4.18 45.70 2.74 43.01
Player 2 46.94 3.16 45.65 3.991 47.31 4.33 44.05
Player 3 44.52 0.00 44.45 2.78 44.73 0.00 43.79
Player 4 45.14 1.64 43.46 0.00 44.73 0.000 43.91
Player 5 47.01 4.24 44.03 1.49 45.28 1.47 43.46
Player 6 45.07 1.54 43.46 0.00 45.21 1.37 45.75
Player 7 45.39 1.94 45.83 3.52 45.50 1.73 44.89
Player 8 45.61 3.07 43.46 0.00 44.73 0.00 41.83
Player 9 44.50 0.00 43.46 0.00 44.73 0.00 42.23
Player 10 44.52 0.00 45.43 2.17 46.02 3.17 53.46
Game 1 46.10 2.19 45.25 3.68 45.79 2.51 52.70
Game 2 45.00 1.45 4.34 1.74 46.45 2.82 51.72
Game 3 46.10 3.52 43.82 1.19 44.73 0.00 43.86
Game 4 45.00 1.45 43.46 0.00 44.73 0.00 39.87
Player/ Tension Vigor
Game SD M SD M SD
Player 1 4.74 44.07 4.99 49.07 11.25
Player 2 4.43 43.81 4.17 39.04 6.01
Player 3 6.86 42.89 3.35 54.81 4.96
Player 4 5.92 43.06 3.08 55.41 3.12
Player 5 4.59 40.80 4.71 59.90 5.82
Player 6 7.00 45.65 4.38 43.03 11.43
Player 7 6.50 42.33 2.64 40.92 3.82
Player 8 2.33 38.94 1.12 65.37 2.02
Player 9 2.78 44.86 4.13 55.71 9.44
Player 10 16.19 43.81 6.52 55.01 9.83
Game 1 7.74 45.14 4.98 51.24 9.11
Game 2 9.51 43.81 2.73 48.43 10.28
Game 3 4.86 41.99 4.98 51.68 10.44
Game 4 0.00 42.40 3.07 56.22 8.71
Game 5 46.26 3.67 43.46 0.00 45.12 1.23 39.87 0.00 45.17
Game 6 44.96 1.38 45.43 3.35 44.73 0.00 42.07 3.32 40.44
Game 7 45.61 3.07 43.46 0.00 45.70 2.74 140.66 1.45 41.31
Game 8 44.52 0.00 45.43 4.21 47.15 4.60 43.40 3.11 42.89
All Games 45.49 2.42 44.35 2.46 45.49 2.28 44.45 6.87 42.95
Game 5 4.81 55.41 11.27
Game 6 2.66 53.48 10.55
Game 7 3.13 50.28 12.87
Game 8 5.58 44.84 13.90
All Games 4.31 51.63 10.88
 Idiographic Analysis of the Relationships
 between Pre-competition Mood and
 Performance
Player Anger Confusion Depression Fatigue
 performance performance performance performance
1 NT -.40 -.40 .33
2 .30 -.27 -.05 -.34
3 NT .23 NT -.31
4 -.53 [**] NT NT -.24
5 -.38 .43 -.79 [*] -.11
6 -.50 [**] NT -.50 [**] -.19
7 -.04 .37 -.82 [*] -.12
8 -.68 [*] NT NT -.32
9 NT NT NT -.58 [**]
10 NT .29 -.42 .13
Player Tension Vigor
 performance performance
1 -.72 [*] .03
2 .14 .44
3 -.33 .15
4 -.30 .05
5 -.90 [*] .24
6 -.33 .55 [**]
7 .21 .00
8 -.51 [**] .09
9 -.84 [*] .45
10 .35 .05
(*.)p [less than].05
(**.)p [less than].10
NT = not tested
 Cross-sectional Analysis of Mood
 and Performance Relationships in
 Eight Games
Game Anger Confusion Depression Fatigue Tension
 performance performance performance performance performance
1 .21 .06 -.32 -.11 -.17
2 -.07 -.03 .06 .29 -.19
3 -.41 .23 NT .37 .40
4 -.75 [*] NT NT NT .42
5 -.31 NT -.35 NT -.20
6 -.44- .61 [*] NT -.75 [*] .12
7 -.09 NT -.09 -.09 .15
8 NT -.18 -.37 .40 -.34
Game Vigor
 performance
1 -.48
2 .16
3 .27
4 .47
5 .02
6 .79 [*]
7 -.39
8 .66 [**]
(*.)p [less than].05
(**.)p [less than].10
NT = Not tested