Structural charting and perceptions of momentum in intercollegiate volleyball.

Burke, Kevin L. ; Houseworth, Steve

Because momentum is an abstract phenomenon, a uniform definition has yet to be developed. Many researchers, however, have various descriptions of this elusive construct. Iso-Ahola and Mobily (1980) developed one of the earliest and most widely cited definitions of momentum which described momentum as "an added or gained psychological power which changes interpersonal perceptions and influences an individual's mental and physical performance." A definition of momentum which does use a bidirectional approach was proposed by Adler (1981). Adler's definition: "Psychological momentum is a bidirectional concept, affecting either the probability of winning or the probability of losing as a function of the outcome of the preceding event." In other words, successful and unsuccessful events lead to additional successful and unsuccessful events. Silva, Hardy, and Crace (1988), however, felt Adler neglected two other concepts which explain momentum, positive inhibition, and negative facilitation. Positive inhibition is considered a loss of momentum and increase in probable failure as a function of preceding success. An example of a football team losing after leading at halftime could possibly be explained by positive inhibition. Negative facilitation, on the other hand, is an increase in momentum and an increase in probable success due to preceding failure. The team which won after trailing at halftime could be accounted for by negative facilitation. Recently Taylor and Demick (1994) and Burke and Weinberg (see Burke, 1993) have offered definitions for momentum. Taylor and Demick defined momentum as a sequential negative or positive change in cognition, affect, and physiology caused by an event(s) which result in a change in competitive outcome. Burke and Weinberg defined momentum in two ways. Positive momentum is defined as a psychological state of mind affecting performance in a positive direction where most everything seems to "go right" for the performer(s). Negative momentum is defined as a psychological state of mind affecting performance in a negative direction where most everything seems to "go wrong" for the performer(s).

Early Success Models

Many researchers have worked from the premise that early points scored in a game or early wins in a multi-game match result in increased momentum, which leads ultimately to game or match victory. For instance, Iso-Ahola and Blanchard (1986) hypothesized that when two people are competing against one another, the individual who gains early momentum by winning the first game of a match is more likely to win the next game. Results from a racquetball tournament (Iso-Ahola & Blanchard, 1986) showed that 74.3% of second game winners won the first game of that match. Weinberg, Richardson, and Jackson (1981) also provided evidence for the early success model. Results of their study of over 2,000 tennis matches illustrated that 86% of male and 91% of female first set winners went on to win the tennis match. Others have demonstrated similar findings in tennis, racquetball, basketball, and football (Hardy and Silva, 1985; Iso-Ahola and Mobily, 1980; Ransom and Weinberg, 1985; Weinberg, Richardson, Jackson, and Yukelson, 1983; Winkleman, Weinberg, and Richardson, 1989). An attempt to account for other variables as they relate to momentum was made by Vallerand, Colacecchio, and Pelletier (1987). They devise an antecedent/consequence model which grouped factors that affect momentum in to three categories: 1) antecedents or events and conditions prior to competition; 2) perceptions and feelings of performers; 3) consequences of game events or "game play." Such a model is a step in the right direction, but future scientific investigations need to focus further on the myriad of factors that affect momentum.

Findings from a three year longitudinal investigation of men's and women's collegiate tennis players (Silva, Hardy, & Crace, 1988) were not as encouraging, though. Results indicated match victory could not be predicted by winning preceding sets when player's ability levels were evenly matched. Thus, momentum could not be inferred from early success. Hardy and Silva (1985), Silva and Hardy (1985), Adler, Richardson, and Jackson (1988), Richardson, Adler, and Jackson (1988), Weinberg and Jackson (1989) and Richardson, Adler, and Hankes (1986) have also found a minimal relationship between early success in tennis and winning the match. Gilovich, Vallone, and Tversky (1985) did not find support for the "hot hand" phenomenon in basketball. Miller and Weinberg (1991) used archival data to investigate momentum in volleyball and found the momentum team (scored three or more consecutive points) did not win the next point or game more than the team without momentum. Finally, a laboratory study (Silva, Cornelius, & Finch, 1992) found performance effects did not lead to momentum. These results suggest early success models cannot explain the driving forces that affect momentum. At best early success models infer that momentum exists. Early success models do not account for game event, psychological, physical, or environmental factors which influence the existence and nature of momentum.

Precompetitive Preparation Models

Another category of momentum research is a precompetitive preparation model. This model, which has not been studied as extensively as the early success model, purports that momentum is a result of psychological readiness to compete. Note for instance a project by Hoffman (1983) in which strength and endurance tasks (bar dips, sit ups, pull ups) were measured as means of comparing two groups. Results suggested that the experimental, or self-induced relaxation group performed significantly more repetitions on the strength and endurance tasks than did the control group.

Shelton (1978) found similar evidence for the precompetitive preparation model. Greater improvements in strength were found in the experimental group. Precompetitive preparation models of momentum, like early success models of momentum, only infer that momentum exists based upon performance outcome. Studies are needed to describe the nature and driving forces of momentum. The purpose of this study was to determine if the existence of momentum could be quantified through a system of structural charting and survey of players' perceptions of momentum.

Method

Subjects

The subjects of this study were female, collegiate volleyball players and coaches from four Midwestern universities. A total of 57 subjects participated in the study, 14 from Team A, 13 from Team B, 13 from Team C, and 17 from Team D. Forty-eight of the subjects were volleyball players while the remaining nine subjects were team coaches. One subject dropped out of the study due to an injury sustained during the tournament. Subjects were volleyball players on teams entered in a round-robin volleyball tournament where the amount of data collected would be significantly greater than during a dual team volleyball match. Also, the round-robin tournament would allow for multiple comparisons between teams of differing ability. All of the subjects completed the post-game momentum questionnaire. Only the survey results of those subjects who participated in the preceding volleyball match were used in the data analysis. Subjects' willingness to complete the post match inventory was a criterion for inclusion as a subject.

Instruments

Data were collected with two different instruments. The first instrument was a structural charting system. The charting system was designed on graph paper and used to chart volleyball game events in order to demonstrate game flow. For each game a line was drawn horizontally across the middle of the graph paper and acted as a baseline for game control. Starting on the baseline, the game flow line would move one graph position in favor of the team gaining possession (i.e., a side out) and two graph positions in favor of the team scoring a point. For the purposes of this study, the space above the baseline represented game control by team one while the space below the baseline represented game control by team two. The more the flow line was above or below the baseline reflected the greater number of points and game control by the team in that particular region of the graph (above or below the baseline). When game points were equal, the game flow line would lie close to the baseline. To describe how each sideout and point was achieved, game events were charted in abbreviated form next to the game flow line.

The second instrument used in this study was a questionnaire designed to gather perceived momentum or game flow perception data from the subjects. The items on the inventory were created with the aid of high school and collegiate volleyball players and coaches. The survey included questions regarding perceptions of game and match flow, and the antecedents and consequences of the flow. The inventory was previously tested and revised for this study (Houseworth & Stelplugh, 1989). Revisions included the addition of a Likert scale to quantify survey responses in order to obtain a composite momentum score.

Pilot Study

Both the structural charting and momentum inventory measuring tools were utilized in a pilot study with collegiate volleyball teams (Houseworth & Stelplugh, 1989). The pilot study served to clarify the wording accuracy of the momentum inventory, and to ensure that the characteristics of the survey were viewed as important game determinants to volleyball players. For the structural charting system of measurement, the pilot study attempted to ascertain whether the charting system was an effective and feasible representation of game flow. Results from the pilot study not only suggested that the structural charts and momentum inventory were accurate measuring tools for momentum, they also suggested the formation of a coincidental relationship (Houseworth & Stelplugh, 1989).

Procedures

Momentum. Each volleyball coach consented to his/her and the players' cooperation in this study before the round-robin tournament began. Each team competed in a best of three game match against the three other teams. Thus, each team played three matches. Game events were charted during the competition of each game. Accurate charting was ensured through inter-observer reliability in which two or more observers judged, clarified and confirmed game event play.

Immediately following the completion of each match, the momentum questionnaires were administered to the players and coaches. Instructions to the subjects, required only after each teams' first match (due to the nature of the round robin tournament structure) were as follows: "Please answer the following items on this survey for the match you just played." There was no mention of the nature of the study.

Data Analysis. For each volleyball match, a multiple regression analysis was performed to examine the possibility of a coincidental relationship between each team's perceived momentum survey score and the structural charting game event and point/side out ratio scores. In addition, the multiple regression procedure was performed on the overall survey and charting scores from the entire tournament.

An analysis of variance of team by match inventory scores was also performed for each match on the momentum survey scores across teams in order to suggest that the differences in perceived momentum may be attributed to the relationship between the momentum inventory scores and game event and point side out ration scores. A Scheffe post-hoc analysis was used to determine where the inventory score differences occurred. The 95% level of confidence was selected, setting alpha = .05 for all analyses.

In addition to the statistical tests used to test the hypothesis, an analysis of the descriptive statistics of the momentum survey item responses was performed.

Results & Discussion

The intent of this study was to assess whether the phenomenon of momentum could be quantified through a system of surveying players' and coaches' perceptions of momentum and structural charting of game events. Momentum survey scores were used to verify whether players' could actually perceive momentum. Quantified survey scores were then correlated with structural chart momentum scores in order to investigate a possible coincidental relationship.

Multiple Regression Results

The momentum survey scores were predicted to correlate positively and significantly with the game event and point/sideout ratio scores. In every match the momentum survey score was significantly related to the structural charting momentum scores. In fact, the relationship of the momentum scoring methods was significant, F(2, 155) = 119.7, p [less than] .05 when the individual match momentum scores were correlated for the entire tournament.

The multiple R value for the correlation was .78 and the R squared value was .61. A correlation of the momentum inventory and point/side out scores produced a Pearson's r of .80 for match one, .75 for match two and .63 for match 3. The Pearson's r for the relationship of the momentum inventory scores and the game event scores was .87 for match one, .79 for match two and .74 for match three. Last, a zero order correlation between game event and point/sideout ratio scores for the entire tournament produced a significant positive relationship and a Pearson's r of .48 and an [R.sup.2] of .22. Table 1 summarizes the multiple regression analysis results of the momentum inventory and game event scores and point/sideout ratios for each match and for the entire tournament.

Analysis of Variance Results

Results of the analysis of variance (team by match momentum score) illustrate significant differences between team momentum scores at .05 alpha for every match played. A complete description of the analysis of variance results can be found in Table 2 and Figure 1.

Descriptive Data Results

Question seven of the survey asked subjects to rate the extent to which a number of external factors influenced match flow. "Warm up prior to the match" was the most widely cited factor affecting match flow. "Officials' calls" were the least influential. Seventy-eight percent of the responses indicated the pregame warm up at least "somewhat" influenced match flow compared to only 14% indicating officials' "somewhat" influenced match flow. However, 68.5% of the responses indicated the officials had "little or no effect" on match flow. A complete breakdown of athletes' and coaches' responses to how external factors influenced match flow is presented in Table 3.

[TABULAR DATA FOR TABLE 1 OMITTED]

[TABULAR DATA FOR TABLE 2 OMITTED]

[TABULAR DATA FOR TABLE 3 OMITTED]

The momentum questionnaire also required the participants to answer questions regarding the existence and flow of momentum. Results show that 96% of responses indicated that there was either a positive or negative groove during the match. However, only 53.1% of the responses indicated that a big play or series of events either helped or hurt their team or opponent. Of the responses which indicated the influence of a big play or specific series of events, 94.2% suggested that the big play affected their teams' consequent performance. Sixty-three percent (63.3%) indicated that the big play influenced the teams' confidence level.

Table 4 summarizes the results of survey questions regarding game flow perception.

Table 4

Distribution of Survey Items Regarding Perception of Momentum

Question & Number of Responses Response
Response Categories Responses Percentages

Groove
Yes 143 96.0%
No 6 4.0%
Total 149 100.0%

Big plays
Yes 86 53.1%
No 76 46.9%
Total 162 100.0%

Team Play
Much Worse 2 2.3%
A Little Worse 9 10.5%
About the Same 5 5.8%
A Little Better 45 52.3%
Much Better 25 29.1%
Total 86 100.0%

Turning Point
Lost Confidence 9 5.8%
No Effect 29 18.7%
Gained Confedence 41 26.5%
No Turning Point 76 49.0%
Total 155 100.0%

Discussion of Results

Multiple Regression Results

The multiple R values for the momentum inventory scores and structural chart scores were not only significant, but also similarly positive and consistent across all matches. In addition, the [R.sup.2] values were equally similar and meaningful across matches. The consistency of multiple regression results across matches indicated the relationship between the momentum inventory and structural chart scoring methods were constant and real. The results suggested what the players perceived as "game flow," "groove" or momentum actually existed.

However, more important than the significant relationships found between the momentum scoring methods was the strength of the variance found for each match. The lowest [R.sup.2] among all of the multiple regression results was .61, which indicated for each match at least 61% of the variance in the momentum inventory scores could be explained by the changes surrounding the game events and point/sideout ratio scores. These results are meaningful because they demonstrated further the strong relationship between perceived and charted momentum.

The multiple regression results also suggested the two momentum structural charting methods were strong predictors of momentum regardless if the inventory scores are correlated with the combined structural chart scores using a multiple correlation or correlated independently using a zero order correlation. It is not true, however, that these findings indicate that the point/sideout and game events scoring methods are one and the same. That is, the two scoring methods are assessing momentum in the same fashion and therefore, do not increase the strength of the multiple R value. Note that a zero order correlation between the game event scores and point/sideout ratios was R = .47 with an [R.sup.2] = .22. These findings indicate that the two structural chart scoring methods are independent of one another. An examination of the scoring schemes suggest they are assessing different aspects of game flow via game events. While the game event score is a function of the way in which points were attained, the point/sideout ratio is more of an absolute value of game control.

Because the findings suggest momentum is affected by game events (i.e., spike, dig, etc.) and ratio of points to side outs, perhaps now coaches could use the structural charts as an evaluation of his/her team's momentum during and between games and matches, and then make the necessary chances to influence his/her team's play. In addition, it may be possible for coaches to aim for a certain game event score or point/ sideout score which had proven to be valid predictor of game victory for their teams in the past.

Analysis of Variance Results

The results of the team by match momentum score analysis of variance further supports the momentum phenomenon. In every match there was a significant difference between team's momentum inventory scores, suggesting the subjects were perceiving momentum, according to team play based upon game events. Of course, the degree to which they perceived momentum depended on the actual events of the game. Since the game events show a significant relationship with perceived momentum, the logical conclusion is the game events were affecting the differences in players' perceived momentum. For example, consider match one in which Team A's momentum survey score was significantly different from Team B's. In this case, the conclusion to make is such differences can be attributed to game play events, and the game events score for both teams will also be quite different.

A lack of significant differences between team inventory scores would not disconfirm the hypothesis. In fact, data from this study indicate that structural game event scores and momentum inventory scores coincide. Thus, teams with similar momentum inventory scores would also have similar momentum structural chart scores. For instance, in a close match where both teams are playing well, not only would momentum inventory scores be similar, but the game event and point/sideout scores would be similar as well.

It is important to remember the survey is sensitive to performance and the resulting momentum, and not to match outcome. In other words, survey scores are a function of performance, not a function of whether a team wins or loses. Therefore, it is not unlikely for a talented team which played poorly, but still won the match, to have a similar inventory score to a less talented team which played superbly but still lost the match.

The inclusion of the analysis of variance in this study suggested if significant differences in perceived momentum occurred, it was most likely an effect of the game events. The results of the analysis of variance suggested momentum exists, can be sensed by the subjects, and can be charted structurally.

Descriptive Data Results

Examination of the descriptive data from the survey and structural charts both refute and confirm some of the early definitions and theories about momentum. For instance, an analysis of the survey item regarding game and match flow demonstrated athletes and coaches perceived momentum to be not only positive, but negative as well. Many participants felt the momentum shifted up and down throughout the match. In addition, the results of the survey illustrated the subjects can both gain and lose confidence after a big play or series of events. Such evidence is contrary to Iso-Ahola and Mobily's (1980) unidirectional definition of momentum, but supports Adler's (1981) and Burke and Weinberg's (1993), Houseworth and Burke's (1990) bidirectional approach to momentum.

Results from this study also support the early success and precompetitive preparation theories of momentum. In every match the team that won the first game eventually won the match. Such findings are parallel to those of Iso-Ahola and Blanchard (1986), Weinberg, Richardson, and Jackson (1981), Hardy and Silva (1985), Iso-Ahola and Mobily (1980), Ransom and Weinberg (1985), Weinberg, Richardson, Jackson, and Yukelson (1983), and Winkleman, Weinberg, and Richardson (1989). It should be noted, however, every match only lasted three games. In other words, in each match the victorious team won the match in three straight games. Such domination of one team over the other may have possibly reduced the validity of the early success model in this study because when teams are not evenly matched, momentum and other factors may not be able to overcome the influence that team talent has on the outcome of the match (Silva, Hardy, & Crace, 1988).

The precompetitive preparation model of momentum gained support from the results of the survey questions which required subjects to rate how various factors influenced match momentum.

The precompetitive factors of "warm up prior to match" and "practices prior to match" were perceived as influential factors affecting match flow. Shelton (1978) and Hoffinan (1983) showed similar findings in their studies regarding readiness state factors of momentum. The emphasis of readiness factors by the subjects also supported Houseworth and Burke's claim (1990) that momentum may begin prior to performance. While traces of support for the precompetitive preparation and early success models of momentum were found in this study, it is important to remember these models can only infer the existence of momentum.

A closer look at the factors which affect momentum demonstrated the participants felt external factors like the crowd, player substitutions, and officials had little influence on match flow. These findings should please volleyball coaches because the results suggest athletes are taking internal responsibility for performance and flow rather than dismissing it to external factors.

As already noted, the main objective of this study was to establish momentum as a verifiable phenomenon by associating structural game events with momentum, as perceived by volleyball athletes and coaches. The data indicated this association is real and may be verified. For example, responses from the momentum inventory overwhelmingly indicated the subjects felt the existence of momentum during each match.

Specifically, 96% of the participants admitted to their team playing well and in a positive groove or not playing well and in a negative groove (question # 1 of the survey). Other questions regarding a rating of game and match flow (questions 2 & 3 on the survey) further showed participants felt the existence of momentum.

An interesting finding from the descriptive data was only 53% of the respondents felt a big play or specific series of events influenced momentum. Consequently, while most of the subjects were able to sense the existence of momentum, no single game event or series of events were viewed as extremely influential on game flow. However, recall each match in this study lasted only three games. Perhaps in each match one team so completely dominated the other team from start to finish momentum was not perceived to have been affected by any one particular big play or series of events. Momentum for this explanation builds if data are analyzed from those respondents who said there was a big play or series of events that hurt or helped their teams. It seems that if a big play or series of events was perceived to exist, performance and confidence was affected as a result (items 5 & 6).

Future studies of momentum in volleyball (and other sports) are necessary before any valuable conclusion may be drawn. These studies should examine spectators, coaches, and players of all ages, skill levels, and gender to get a better grasp on this elusive phenomenon.

References

Adler, P. (1981). Momentum: A theory of social action. Beverly Hills, CA: Sage.

Adler, W., Richardson, P., & Jackson, A. (1988). Psychological momentum in tennis: Predicting match victory for professional female players. Paper presented at the annual meeting of the Association for the Advancement of Applied Sport Psychology, Nashua, NH.

Burke, K.L. (1993, October). Psychological momentum: Are we on a roll yet? In J. Taylor (Chair), Keeping the mo' in momentum research: The past, present, and future of momentum in sports. Symposium conducted at the eighth annual conference of the Association for the Advancement of Applied Sport Psychology, Montreal.

Burke, K.L., Weinberg, R. S., & Weingand, D. (1992). Perceptual agreement on the origin, end, and cause of psychological momentum in basketball. Paper presented at the annual meeting of the Association for the Advancement of Applied Sport Psychology, Colorado Springs, CO.

Gilovich, T., Vallone, R., & Tversky, A. (1985). The hot hand in basketball: On the misperception of random sequences. Cognitive Psychology, 17, 295-314.

Hardy, C. J., & Silva, J. M. (1985). The mediational role of psychological momentum in performance outcome. In J.R. Thomas (Ed.), Psychology of motor behavior and sport, (p. 106). Gulf Park, MS: NASPSA.

Hoffman, A. J. (1983). Effects of psychological momentum on the physiology and cognition among American athletes. International Journal of Sport Psychology, 14, 41-53.

Houseworth, S., & Burke, K. L. (1999). Issues in understanding and researching psychological momentum. Paper presented at the annual meeting of the Association for the Advancement of Applied Sport Psychology, San Antonio, TX.

Houseworth, S., & Stelplugh, C. (1989). Tracking the elusive "Big MO": A structural charting and interview model. Unpublished manuscript.

Iso-Ahola, S. E., & Mobily, K. (1980). "Psychological Momentum": A phenomenon and an empirical (unobtrusive) validation of its influence in a competitive sport tournament. Psychological Reports, 46, 391-401.

Iso-Ahola, S. E., & Blanchard, W. J. (1986). Psychological momentum and competitive sport performance: A field study. Perceptual and Motor Skills, 2, 763-768.

Miller, S., & Weinberg, R. (1991). Perceptions of psychological momentum and their relationship to performance. The Sport Psychologist, 5, 211-222.

Ransom, K., & Weinberg, R. (1985). Effect of situational criticality on performance of elite male and female tennis players. Journal of Sport Behavior, 8, (3) 144-148.

Richardson, P. A., Adler, W., & Hankes, K. D. (1986). Game, set, match: Psychological momentum in tennis. The Sport Psychologist, 2, 69-76.

Richardson, P. A., Adler, W., & Jackson, A. J. (1988). Gaining psychological momentum: The serve as a determinant of match success for professional tennis players. Paper presented at the annual meeting of the Association for the Advancement of Applied Sport Psychology. Nashua, NH.

Shelton, T. O. (1978). The content and effect of psyching up strategies in weight lifters. Cognitive Therapy and Research, 2 (3), 275-284.

Silva, J. M., Cornelius, A. E., & Finch, L. M. (1992). Psychological momentum and skill performance: A laboratory study. Journal of Sport & Exercise Psychology, 14 23, 119-133.

Silva, J. M., & Hardy, C. J. (1985). Psychological momentum and performance outcome in college tennis. In J.R. Thomas (Ed.), Psychology of motor behavior and sport, (p. 105). Gulf Park, MS.

Silva, J. M. & Hardy, C. J., & Crace, R. (1988). Analysis of psychological momentum on male and female tennis players revisited. Journal of Sport & Exercise Psychology, 10, 346-354.

Taylor, J., & Demick, A. (1994). A multidimensional model of momentum in sports. Journal of Applied Sport Psychology, 6 (1), 51-70.

Vallerand, R.J., Colaceccllio, P. G. & Pelletier, L. G. (1982). Psychological momentum and performance inferences: A preliminary test of the antecedents-consequences psychological momentum model. Journal of Sport Psychology, 10 (1), 92-108.

Weinberg, R., & Jackson, A. (1989). Analysis of psychological momentum in intercollegiate tennis. Journal of Sport Behavior, 12 (3), 167-179.

Weinberg, R., Richardson, P. A., & Jackson, A. J. (1981). Effect of situation criticality of tennis performance of males and females. International Journal of Sport Psychology, 12, 253-259.

Weinberg, R., Richardson, P., Jackson, A., & Yukelson, D. (1983). Coming from behind to win: Sex differences in interacting sport teams. International Journal of Sport Psychology, 14, 79-84

Winkelmann, A., Weinberg, R., & Richardson, P. (1989). Psychological momentum in collegiate football: A preliminary investigation. Paper presented at the Association for the Advancement of Applied Sport Psychology Annual Conference, 1989, Seattle, WA.