A classificatory system of anxiety-inducing situations in four team sports.

Dunn, John G.H. ; Nielsen, A. Brian

All human behavior is situated, and if behavior is to be fully understood, the features of the situations in which the behavior occurs must also be understood (Krahe, 1992; Van Heck, 1989). Despite the recognition of situational importance by researchers in social psychology (Cantor & Kihlstrom, 1987; Krahe, 1992), most research concerning anxiety in sport has concentrated upon the characteristics of the athletes, while paying little attention to the characteristics of the situations in which the athletes' behavior took place (Dunn & Nielsen, 1993; Hanin, 1989; Spielberger, 1989). This lack of attention to sport situational characteristics remains an obstacle to understanding how situations affect, or interact with, the athlete in competition, and inhibits the systematic investigation of the situations themselves.

The modern interactionist model of psychology proposes that behavior is a function of both the person and the situation (Endler & Magnusson, 1976). Although the effects of specific situational stimuli upon behavior have been examined by social psychologists in experimental settings, Edwards (1984) maintains that "much work is needed regarding the dimensions along which naturally occurring social situations can be analyzed and compared" (p. 147).

The situation plays an integral role in most theoretical models dealing with emotional responses. For example, anxiety is experienced by individuals who perceive a situation to be threatening to their psychological or physical well being (Lazarus & Averill, 1972; Martens, Vealey, & Burton, 1990; Smith & Lazarus, 1990; Spielberger, 1966). Elevated levels of anxiety can, in turn, have an adverse effect upon motor performance (Landers, 1980; Schmidt, 1988). Consequently, sport psychologists and coaches who assist in the enhancement of athletic performance must be aware of the objective competitive situations (Hackfort & Schwenkmezger, 1993; Martens et al., 1990) that are likely to be perceived of as threatening by the athletes.

Recent research has contributed to an understanding of the multidimensional nature of competitive anxiety and its effect upon performance (e.g., Jones, Swain, & Cale, 1990; Martens et al., 1990; Smith, Smoll, & Schutz, 1990). However, regardless of the exact nature of the anxiety experienced by an athlete (i.e., cognitive, somatic, or both), the characteristics of the situations with which the athletes interact still require identification (Spielberger, 1989).

In team sport environments, heightened competitive anxiety has been associated with specific situations where the potential for social evaluation is great (e.g., free throw in basketball, penalty kick in soccer: Scanlan, 1984). In these situations, the opportunity to diffuse performance responsibility among teammates is minimized (Martens et al., 1990) and the degree to which the "negative spotlight" (Fisher & Zwart, 1982, p. 145) focuses upon the athlete is maximized.

Situational circumstances have also been shown to differentially affect competitive state anxiety levels. Krane, Joyce, and Rafeld (1994) found that as the degree of criticality in a batting situation increased (e.g., bringing home a player on third base; batting in the late innings of a tied or one-run ball game) the levels of anxiety experienced by female intercollegiate softball players also increased. These findings reinforce calls to examine the characteristics of anxiety-inducing situations both within and across sports (Hackfort & Spielberger, 1989; Hanin, 1989). A greater research focus on the specific competitive situation in sport psychology has also been suggested by other researchers. For example, Krane et al., (1994) recommended that sport psychologists and coaches recognize the types of anxiety-inducing situations that occur during competition, while Wrisberg and Pein (1992) contended that defining the characteristics of anxiety-producing situations from different sports is of primary importance.

Various researchers have attempted to identify the psychological dimensions upon which threat and anxiety are based in specific competitive situations (e.g., ice hockey and soccer: Dunn & Nielsen, 1993; basketball: Fisher & Zwart, 1982; wrestling: Gould, Horn, & Spreeman, 1983). However, no studies have attempted to systematically identify, compare, and catalogue the objective features of these situations across sports. Thus possible between-sport comparisons of the types of situations that cause athletes anxiety have been very limited.

One of the first, and critical, steps toward understanding situational characteristics across environments is to describe and classify the situations in an orderly systematic manner (Endler, 1983; Krahe, 1992). Frederiksen (1972) proposed that in order to progress in understanding how situations influence behavior, we must develop a "systematic way of conceptualizing the domain of situations and situational variables" (p. 115). Thus, the development of situational classification systems or taxonomies is strongly recommended (Frederiksen, 1972; Krahe, 1992; Van Heck, 1989). Such classifications aid in the identification of similarities and differences across situations, thereby providing some objective structure to the seemingly complex environment of situational characteristics (Cvetkovich & Earle, 1985); however, few such schematic representations actually exist (Baumeister & Tice, 1985).

Given the necessity of understanding the role and characteristics of the situation in the competitive anxiety process, "classificatory systems" (Sokal, 1974, p. 1116) could provide a critical starting point for the improved understanding of anxiety-inducing sport situations. In addition, such a system would provide researchers with a catalogue of athlete-generated situations that could be used in the development of situational-anxiety inventories. Selecting situations in this manner helps the researcher sample stimuli (i.e., situations) that are relevant to the athletes being studied (Dunn & Nielsen, 1993).

The current study was undertaken as a first step to developing a classificatory system, wherein anxiety-inducing game situations from a variety of similarly structured team sports could be identified, categorized, and analyzed according to their most salient features. This procedure is in accordance with the recommendations of Rotter (1955), who proposed that situations be identified and analyzed on the basis of their objective characteristics prior to being analyzed on the basis of their psychological characteristics (Pervin, 1978).

The major problem of classifying situations on the basis of the psychological meaning which the perceiver attaches to the situation, is that the same behavioral settings (i.e., objective situations) are frequently perceived differently by different individuals (Dunn, 1994; Pargman, 1993). In other words, two people may experience different forms of anxiety (e.g., cognitive or somatic) in the same situation but for different reasons and to different degrees. Using objective criterion to describe the situations gives the researcher a starting point for examining the person-situation interaction and provides directions as to where and when these unpleasant interactions may potentially occur. The nature of the anxious response (whether cognitive or somatic) can then receive the appropriate follow-up attention. Therefore, while an "objective classificatory system" of situations may not be sufficient, in and of itself, to let us understand the complex person-situation interaction, it does provide researchers with a necessary starting point from which to examine the situational factors influencing competitive state anxiety.

The two purposes of the study were: (a) to identify competitive game situations perceived by athletes (competing in similarly structured team sports) as being typical and anxiety-inducing, and (b) to develop a situational classificatory system that permitted the investigation into whether athletes from different team sports experienced anxiety in similar types of situations.

Methods

Subjects

A total of 185 athletes (117 males and 68 females) from nine high performance university, college, and regional teams in four sports (basketball, field hockey, ice hockey, soccer) served as subjects in the study (see Table 1). The sports were chosen because it was felt that they displayed certain parallel structural features associated with continuous flow invasion sports (cf., Dunn & Nielsen 1993; Thorpe, Bunker, & Almond, 1986). The term "high performance" was used to describe the collective performances of the teams studied. Five of the six university teams finished their seasons ranked in the national top 10, while all three non-university teams (college basketball, provincial field hockey, and AAA Midget ice hockey) reached their respective national finals. The mean age of the entire sample was 20.5 years (SD = 2.69) and subjects reported an average of 9.3 years (SD = 4.02) playing experience in their affiliated organized sport.

[TABULAR DATA FOR TABLE 1 OMITTED]

Instrument

A two-part questionnaire was developed with part 1 requesting demographic and playing information (e.g., age, experience, playing positions). Part 2 requested that the subjects identify and describe up to five situations in their sport that are typically encountered and which produce anxiety during competition. Athletes were also requested to provide a brief explanation of why anxiety was experienced; this helped the researchers ensure that the situations cited were indeed anxiety-producing, and not merely situations associated with general excitement or arousal.

To ensure that the athletes understood the concept of anxiety, verbal and written instructions were provided. Specifically, anxiety was defined as a response to a psychological or physical danger (Martens et al., 1990) which causes the individual to experience unpleasant negative emotions or distress (Everly & Rosenfeld, 1981). It was made clear to the athletes that they were to cite only those situations that were associated with negative emotions. Situations in which athletes experienced positive emotions related to general increased arousal levels (e.g., excitement, looking forward to the challenge) were to be avoided. This operational definition of anxiety was considered to be "user friendly" for the athletes in the study, and followed the recommendations of Hackfort and Spielberger (1989), who argue that researchers should use definitions of anxiety that prove useful in solving the practical problems they are investigating. Upon review of the situations cited by the athletes, the definition was considered successful in eliciting the types of responses that reflected negative emotions.

Although it was recognized that anxiety can be considered as a multidimensional construct (Gould & Krane, 1992), the purpose of this exploratory study was to identify and classify the specific anxiety-inducing situations based upon their objective features, rather than to determine the nature of the anxiety experienced (i.e., cognitive or somatic) or have athletes make such discriminations. To classify the situations according to whether cognitive or somatic anxiety was experienced was not considered feasible because, as previously discussed, different individuals can perceive and react differently to the same situation (Mischel, 1979). In other words, the predominant anxiety response for one athlete to a certain situation may be cognitive in nature, whereas another athlete may experience a predominantly somatic response to the same situation. Under these circumstances, the same situation would be located in two separate classificatory systems. Therefore, no attempt was made in this study to determine or classify the cognitive versus somatic disposition of athletes' anxiety; an idiographic approach seems most suitable to examine this phenomenon. Finally, the confidentiality of responses from the coach was affirmed both verbally and on the instrument.

Procedures and Analysis

The questionnaire was administered to the athletes by the investigators at times convenient to the teams and coaches (e.g., team meeting, after practice, etc.) during the latter half of the regular seasons. Just prior to questionnaire completion, subjects were informed of the general nature of the study and encouraged to cite only those anxiety-eliciting situations most typically encountered in their athletic experience. Thus, the use of rare and unique situations (e.g., final moments of the championship game) was discouraged.

To illustrate what might constitute "typical" situations, five hypothetical examples were verbally provided from a game other than the athletes' respective sport. These examples were also intended to illustrate the great diversity of possible responses and reflected varied situations concerning physical characteristics (e.g., injury), responsibility (e.g., coverage), opponents (e.g., skill levels), coaching (e.g., personnel decisions) and officiating. It was emphasized that these were examples only and were provided in order to stimulate, rather than restrict, any responses an athlete may wish to offer. The examples were quite specific and easily understood. Athletes were again reminded of the confidentiality of all responses from the coach.

The data analysis was conducted using an inductive reasoning technique commonly referred to as "clustering" (Krippendorff, 1980; Miles & Huberman, 1984). The inductive process is described by Patton (1980) as one which permits the creation of clusters of common themes to emerge from the data. This is in contrast to a deductive methodology which requires the data to be fitted into pre-existing categories - an approach which may be more appropriate once a classificatory system exists. Scanlan, Stein, and Ravizza (1989) refer to the process as an "inductive content analysis" and liken it to a conceptual factor analytic technique. The data analysis procedure employed in the present study was very similar to the process adopted by Scanlan et al. in their investigation of enjoyment sources for elite figure skaters.

In accordance with the clustering protocol suggested by Krippendorff (1980), an attempt was made to develop mutually exclusive categories whereby the differences in the situational characteristics of each cluster were clearly identifiable. All responses were initially grouped by sport and according to their most prominent key features (e.g., scoring opportunities, substitutions, officiating decisions, etc.), This procedure led to the creation of preliminary categories of responses which clustered together around common salient themes. The poorest fitting initial categories were then re-examined, resulting in them being redefined, collapsed, or altered in some way so that ill-fitting responses were allocated to more appropriate categories or were used to generate new independent categories. Following a similar methodology used by Scanlan et al. (1989), "consensus validation" (p. 72) was employed by the researchers to determine the final location of each situation within the classificatory system. That is, when the researchers did not readily agree upon the characteristics of a theme or upon the classification of a situation, they each argued their own perspective until arriving at a mutually acceptable solution. This process reduces the effect of the potential bias of each analyst (Scanlan et al., 1989). All responses were eventually assigned to an appropriate category through the repetitive use of this "convergence" (Patton, 1980) procedure.

Guba (1978) suggested that the degree to which the resulting classificatory system reflects a reliable accommodation of responses should be examined. Two individuals who had no previous access to the data were therefore asked to classify one randomly selected subset of original responses (n = 98) according to the categories provided. The degree to which the coders agreed with the original classification of the responses was determined using Kappa coefficient (Cohen, 1960).

Finally, upon completion of the classification process for each sport, an effort was made to categorize responses across sports to determine the extent to which a system of generic categories was viable. This process involved an examination of the salient features of each set of responses for each of the team/sport settings.

Results and Discussion

The inductive classification procedures with responses across all sports led to the emergence of four general superordinate level 1 categories labelled Ongoing Game Situations, Game/Score/Time Criticality Situations, Coach Related Situations, and Miscellaneous Situations. These categories constitute the most generalizable level of the classificatory system and encompass a wide variety of situations with similar underlying characteristics. These categories were then subdivided into 16 more specific subordinate level 2 categories (e.g., Offensive Situations, Defensive Situations, Stoppage Situations, etc.). The subordinate categories encompass a narrower range of common situational features. A review of Figure 1 will help to illustrate the types of responses that were received and clarify their classification.

Kappa coefficients for the two independent coders versus the original response classification were .97 for the four superordinate categories and .91 for the 16 subordinate categories. These coefficients indicate excellent inter-rater agreement at both levels (Bakeman & Gottman, 1986). Of the few classification discrepancies, most were due to the unclear nature of some of the responses as opposed to a lack of comprehension or unsuitability of the classificatory system itself. Due to the confidential nature of the results, follow-up interviews could not be conducted to clarify those responses where discrepancies in rater interpretation occurred.

Superordinate Categories

The Ongoing Game Situations (OGS) category included those situations that occurred while the ball/puck was still in play. In other words, the ball or puck was considered "live" and could be accessed by either team; thus, stoppage situations (e.g., free throw, penalty kick) were excluded from this category. The second superordinate category accommodated situations of Game/Score/Time Criticality (GSTC). The primary focus of the GSTC category was on conditions of score, time, or game importance. Included within the category were situations in which the athletes cited the main reason for experiencing anxiety was due to temporal factors or score considerations and constraints. Coach Related Situations (CRS) was the third major category and reflected situations that were considered to be under the direct or indirect influence of the coach. The fourth, and final, main category was labelled Miscellaneous Situations (MS) and [TABULAR DATA FOR TABLE 2 OMITTED] accounted for responses that were discernible in content but which did not fit under the three previous categories. The major categories and their associated subcategories are shown in Table 2 along with the frequencies and proportions of responses from the entire sample accommodated by each. The classification of the responses according to specific sports is displayed in Table 3. The total number of responses received from basketball, field hockey, ice hockey, and soccer were 249, 132, 244, and 196 respectively.

[TABULAR DATA FOR TABLE 3 OMITTED]

The results of the classification process did support a viable system of categories to which the sport specific responses conformed. Many of the actual athlete responses initially appeared to be sport specific and unique, yet, as the essential situational features were examined, these responses contained the same conceptual characteristics as responses across sports. For example, the soccer response, "Unexpectedly getting the ball in the penalty box from a rebound, and having an open shot at goal" anticipates a good scoring opportunity in the same way as does the hockey response, "Picking up a loose puck at center ice and getting a breakaway with only the goaltender to beat." Thus, there emerged the overriding types of situations that manifest themselves differently in different sports, but which retain the generic commonality of their category.

The differential "loadings" of sports in the various categories (see Table 3) was not surprising, and reflected the differences in the nature of the sports themselves. That is, each sport environment may reflect a typical pattern of responses but these are expected to contrast with the different natures of the games. Therefore, the proportions of responses in each category can be expected to vary from soccer to hockey or from basketball to field hockey. These differences at least partly reflect the differences in the sports themselves. This was not considered problematic because the purpose of this cross sport classificatory system was to accommodate and classify all available stimuli based upon their salient characteristics and at the same time permit the typical sport-specific response patterns to contrast with each other. The inclusion of the four sports in the present system makes comparisons and further investigation of these differences possible.

Subordinate Categories

Ongoing Game Situations. Within the OGS superordinate category, four subordinate categories emerged from the data: offensive situations, defensive situations, situations where the offensive and defensive nature was unspecified, and injury related situations. The subordinate offensive and defensive categories encompassed situations during uninterrupted play that were clearly defined by the athlete as being of an offensive or defensive nature. Generally, offensive situations occurred when the athlete was in possession of the ball or puck. However, some instances such as "clearing" in the defensive end were deemed to be defensive situations despite the athlete's momentary possession of the object.

Basketball was the only sport studied to show substantially fewer responses in the defensive subcategory as compared to the offensive subcategory. This finding may be partly due to the high scoring nature of the game whereby good scoring opportunities are frequently presented to both teams. This is in contrast to the small number of truly good scoring opportunities which arise in sports such as field hockey and soccer where such chances are quite rare. In other words, a mistake in a defensive situation in basketball resulting in points for the opposition may not carry the same critical importance or cause the same degree of anxiety as a mistake resulting in a goal to the opposition in the other sports, because many opportunities to make up the score deficit in basketball are likely to occur. When the offensive or defensive nature of the situation was not specified (e.g., "One on ones. Scared I might fail.") such responses were placed in the nature unspecified subcategory. Therefore, although its inclusion in the model was necessary, such responses were infrequent.

The fourth subcategory encompassed within the major OGS category was labelled injury situations and included all situations where the potential for physical harm was cited. Injury-related responses were cited by athletes in all four sports, however, the sport of ice hockey supplied half (n = 17) of those responses. This finding is not surprising due to the frequency and nature of the physical contact experienced in that sport. However, of the 244 ice hockey responses obtained across all categories, potential injury situations still only accounted for 7% of this total. Although the potential for physical injury is a recognized threat for ice hockey players in certain situations (Dunn & Nielsen, 1993), these findings suggest that they certainly do not constitute the primary source of anxiety in the sport. It is also recognized that certain situations may have the potential to cause personal humiliation in addition to injury. The primary reasons for anxiety in these situations require further investigation.

Game/Score/Time Criticality. The GSTC superordinate category contained three subordinate level 2 categories: stoppage situations, within-game flow situations, and pregame situations. The level 2 subcategory containing the third highest frequency of total responses (n = 136) in the entire classificatory system was labelled stoppages. Stoppage situations included the free throw (basketball), penalty kick (soccer), face-off (ice hockey) and penalty corner (field hockey). Almost one third (n = 34) of the total stoppage responses were cited by basketball athletes. With the exception of four responses indicating tactical or time based stoppages (level 3 subcategories contained within the stoppage category: e.g., tactical time outs), the remaining 30 cited the free-throw situation. However, it should again be noted that the majority of these athletes indicated that the free throw was especially anxiety-inducing when encountered at perceived critical times in the match.

Probably the most prominent feature of the stoppage responses was that over 90% were anticipatory in nature. That is, the anxiety was experienced prior to taking the free throw as opposed to being experienced after missing a free throw. Very few athletes reported experiencing the anxiety as a result of having failed to successfully execute the desired performance. Fisher and Zwart (1982) have previously shown that collegiate basketball players do differentiate between anxiety in situations where failure has occurred and where the outcome of the event is still uncertain. The significance of these findings will be discussed in more detail later.

The within-game flow subcategory refers to general score or time-based situations within which the athlete experienced the anxiety. Examples included "Trailing in a game by a close margin in the closing seconds," "Defending a close lead in the closing minutes," or generally "Playing during important games."

Athletes from all four sports provided responses that contributed to the within-game flow subcategory, but about two thirds (n = 42) of these responses were cited by ice hockey players: half of whom identified overtime play. The athletes indicated that the main reason for the anxiety was that one mistake in overtime can have critical consequences resulting in the immediate loss of the game. Due to the rule structure in the three other sports examined, overtime is seldom, if ever, encountered during the regular season and therefore was probably not considered typical.

Coach Related Situations. With almost one fifth (n = 148) of all responses being classified within the CRS category, it is clear that the behaviors of the coach play an important part in producing athlete anxiety. The three subordinate level 2 categories contained within the CRS category included personnel decisions, tactical decision conflicts, and performance feedback situations.

The subcategory labelled personnel decisions affecting the individual represented situations whereby coaching decisions affected an athlete's playing time. The responses concentrated predominantly on substitution decisions (both into, and out of, the game), pregame team selections and the lack of playing time athletes were experiencing.

A much smaller number of coach-related responses (n = 21) focused on tactical decision conflicts. These were situations that made the athletes anxious over issues concerning team strategy (e.g., feeling that the team was using an overly complicated powerplay formation) and assigned playing duties external to the context of substitutions (e.g., being asked by the coach to play an unfamiliar and uncomfortable role for the team).

Performance feedback (in the form of verbal and nonverbal messages) was the third CRS subcategory and was further divided into two level 3 subdivisions - situations where coach feedback was anticipated and situations when anxiety was experienced as an actual consequence of receiving the feedback. Only five other level 2 subordinate categories in the entire classificatory system contained a greater number of responses than the performance feedback category (n = 42), demonstrating that the feedback provided by the coach can be an important contributing factor to causing athlete anxiety. The emergence of a performance feedback category supports Smoll and Smith's (1989) model of leadership behavior which indicates that coaching gestures such as verbal and nonverbal feedback influence the emotional reactions of athletes. Indeed, Kenow and Williams (1992) confirmed that certain coaching behaviors have a strong influence on causing athletes to experience anxiety.

Miscellaneous Situations. Four level 2 subordinate categories were encompassed Within the major MS category: audience factors, officiating factors, teammate behaviors, and opponent behaviors. The audience subcategory contained those situations in which the athletes cited the spectators as being the primary cause of anxiety. Over half of the audience-related responses were given by ice hockey players. All of the "audience" responses (n = 18) identified by the hockey players were cited by the younger midget hockey players (mean age = 16.7 years). This finding is consistent with those of Ewing, Feltz, Shultz, and Albrecht (1988) who showed that social approval is a very strong motivator among elite midget level players. In addition, the subjects in this study were very aware of the possibility that professional team scouts were observing and assessing their performances during competition. This situation is culturally related, and the proportion of responses within this category would probably differ depending on the cultural context of the sport (e.g., elite youth soccer in Europe, or collegiate basketball in the U.S.A.).

Officiating decisions were cited a total of 41 times as being a source of high anxiety. Although almost three quarters of the responses cited "poor" officiating decisions, the remaining 30% did not directly identify the official as the source of the problem. That is, some responses identified the problem as the athlete's own undisciplined actions which resulted in a penalty that could hurt the team (e.g., two minute minor penalty in ice hockey resulting in short handed play, or receiving a foul in basketball which leads to personal foul trouble). The remaining two important level 2 subcategories dealt with teammate behaviors (e.g., teammates not trying hard or giving up) and opponent characteristics or behaviors (e.g., athletes playing against a stronger opponent and worrying about being unable to cope or getting embarrassed).

Implications

It is evident that the responses obtained in this study share many similar features both within and across team sports. This suggests that the anxiety-inducing situations that occur typically in one specific sport have, as their counterparts, parallel typical situations in other team sports.

The situational features that emerged from the major categories and subcategories appear to be closely related to research topics that have been investigated in a number of anxiety-related studies. For example, Folkins (1970) suggested that periods of anticipation (20-60 seconds) prior to encountering a threatening situation are accompanied by maximal elevations in state anxiety. Many of the stoppage situations cited here (e.g., free throw, face-off, penalty stroke) require the athlete to wait for a time period similar to those described by Folkins (1970). Not disregarding the fact that all these examples were situations of added responsibility, it remains that over 90% of stoppage related responses indicated that the anxiety experienced was due to anticipating some event. Future research into the amount of time athletes must wait prior to performing skills following a stoppage in a critical situation would provide valuable information for coaches who so frequently call tactical time outs prior to an opponent's execution of these closed skills.

To help athletes deal with stoppage-type situations, applied sport psychologists have been examining preperformance routines (e.g., prior to taking a free throw) that assist athletes in maintaining control (of their focus and anxiety levels) in these types of situations (Cohn, 1990; Wrisberg & Pein, 1992). The classificatory system should help sport psychologists identify the types of situations across sports that may be most suitable to receive preperformance-routine training.

Frijda (1985) has also referred to the importance of the "time perspective" of the situation when examining anxiety. Time perspectives refer to the amount of time which individuals have to evaluate the situation in which they find themselves. Having identified stoppages within a systematic framework of situation characteristics, future research can begin to examine whether the nature of competitive anxiety experienced during stoppages (i.e., cognitive and somatic anxiety) is different from that experienced in other types of situations.

Elevated pregame state anxiety in athletes of varying ages, experience levels, and sex has been confirmed by numerous researchers (e.g., Jones, Swain, & Cale, 1991; Sanderson & Ashton, 1981). The anxiety most likely occurs because during this period the athlete has a considerable amount of time to anticipate the potential scenarios surrounding personal performance or match outcome. The pregame and stoppage sub-categories appear to reflect the time perspective situational characteristics proposed by Frijda (1985).

The degree to which athletes can control the situations they encounter appears to be another underlying feature of many cross-sport situations. For example, situations from the superordinate coach and miscellaneous categories appear to be almost entirely out of the athlete's control. The construct of controllability has been identified by numerous researchers as a factor that can potentially affect situational threat and anxiety (e.g., Dunn & Nielsen, 1993; Martens et al., 1990; Paterson & Neufeld, 1987). It may be that as the degree to which the athlete lacks control of the situation increases, the level of anxiety increases correspondingly. Further investigation into the actual levels of competitive state anxiety experienced in the aforementioned situations is required to investigate this relationship.

Another common theme associated with the objective characteristics of many situations is the degree to which the situation is open or closed (i.e., the existence of available ways to avoid confronting the threatening situation: Frijda, 1985). As the situation offers fewer decision choice alternatives, the situation becomes more closed. The offensive, defensive, and stoppage situation examples in Figure 1 clearly provide very few feasible options for potential escape. Apparently, as the situation becomes more closed, the likelihood of experiencing anxiety is increased. Closed situations may therefore be labelled as "situations of added responsibility" since the athlete must assume virtually all responsibility for the outcome. These situations also provide increased opportunity for those watching the athlete to judge the performance, suggesting that some kind of evaluative apprehension may trigger the anxious response (Scanlan, 1984). The classificatory system again helps researchers identify the underlying situational factors that are common across sports.

Finally, the GSTC category reflects those situations where the consequences of failure were evaluated as having greater importance than if the same situation had occurred at another less critical period in the competition. The emergence of this category comes as no surprise, since many sport psychology researchers have used "situation criticality" as an independent variable to determine its effect upon state anxiety, perceived psychological crisis, and performance behavior (e.g., baseball: Lowe, 1973; basketball: Bar-Eli & Tenenbaum, 1989; Wrisberg & Pein, 1992; softball: Krane et al., 1994).

In summary, the purpose of this study was to identify and classify situations associated with causing anxiety for athletes in a variety of similarly structured team sports. The results support the development of a classificatory system with which the characteristics of anxiety-provoking situations may be better understood. The generic cross-sport solution resulting from the inductive analysis appears to accommodate the wide variety of anxiety-inducing situations which athletes, across similarly structured team sports, typically encounter. Although a number of between-sport differences were observed surrounding the percentage of responses in various categories, it is recognized that results could have been affected by the idiosyncratic nature of each team. If the findings of this study are to be generalized to other athletes participating in similar and different sports, the classificatory system needs to be tested with other teams to discover if similar response patterns would occur. In addition, the effect that athletes' age, gender, and experience levels may have upon the types of responses provided needs further investigation.

If the role of the situation is to be more fully examined in future sport psychology research, the situations used in these studies need to be athlete-generated and representative of those encountered within the environments being studied (Fisher & Zwart, 1982; Krahe, 1986). Ultimately, the classificatory system developed in this study may provide a basis upon which such situated athlete behavior in team sports can be examined.

References

Bakeman, R., & Gottman, J.M. (1986). Observing interaction: An introduction to sequential analysis. New York: Cambridge University Press.

Bar-Eli, M, & Tenenbaum, G. (1989). Game standings and psychological crisis in sport: Theory and research. Canadian Journal of Sports Science, 14, 31-37.

Baumeister, R.F., & Tice, D.M. (1985). Toward a theory of situational structure. Environment and Behavior, 17, 147-192.

Cantor, N., & Kihlstrom, J.F. (1987). Personality and social intelligence. Englewood Cliffs, NJ: Prentice Hall.

Cohn, P.J. (1990). Preperformance routines in sport: Theoretical support and practical applications. The Sport Psychologist, 4, 301-312.

Cohen, J. (1960), A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37-46.

Cvetkovich, G., & Earle, T.C. (1985). Classifying hazardous events. Journal of Environmental Psychology, 5, 5-35.

Dunn, J.G.H. (1994). Toward the combined use of nomothetic and idiographic methodologies in sport psychology: An empirical example. The Sport Psychologist, 8, 376-392.

Dunn, J.G.H., & Nielsen, A.B. (1993). A between-sport comparison of situational threat perceptions in ice hockey and soccer. Journal of Sport & Exercise Psychology, 15, 449-465.

Edwards, J.M. (1984). Situational determinants of behavior; In N.S. Endler & J.M. Hunt (Eds.), Personality and the behavioral disorders (Vol. 1, 2nd ed., pp. 147-182). New York: Wiley.

Endler, N.S. (1983). Interactionism. A personality model but not yet a theory. In M.M. Page (Ed.), Nebraska symposium on motivation (1982): Personality-Current theory and research (pp. 155-200). Lincoln: University of Nebraska.

Endler, N.S., & Magnusson, D. (1976). Toward an interactional psychology of personality. Psychological Bulletin, 83, 956-974.

Everly G.S. & Rosenfeld, R. (1981). The nature and treatment of the stress process: A practical guide for clinicians. New York: Plenum.

Ewing, M.E., Feltz, D.L., Shultz, T.D., & Albrecht, R.R. (1988). Psychological characteristics of competitive young hockey players. In E.W. Brown & C.F. Branta (Eds.), Competitive sports for children and youth (pp. 49-61). Champaign, IL: Human Kinetics.

Fisher, A.C., & Zwart, E.F. (1982). Psychological analysis of athletes' anxiety responses. Journal of Sport Psychology, 4, 139-158.

Folkins, C.H. (1970). Temporal factors and the cognitive mediators of stress reaction. Journal of Personality and Social Psychology, 14, 173-184.

Frederiksen, N. (1972). Toward a taxonomy of situations. American Psychologist, 27, 114-123.

Frijda, N.H. (1985). Toward a model of emotion. In C.D. Spielberger, I.G. Sarason, & P.B. Defares (Eds.), Stress and anxiety (Vol. 9, pp. 3-16). New York: Hemisphere.

Gould, D., Horn, T., & Spreeman, J. (1983). Sources of stress in junior elite wrestlers. Journal of Sport Psychology, 5, 159-171.

Gould, D., & Krane, V. (1992). The arousal-athletic performance relationship: Current status and future directions. In T.S. Horn (Ed.), Advances in sport psychology (pp. 119-141). Champaign IL: Human Kinetics.

Guba, E.G. (1978). Toward a methodology of naturalistic inquiry in educational evaluation. CSE Monographs Series in Evaluation No. 8, Los Angeles: Center for the Study of Evaluation, University of California, Los Angeles.

Hackfort, D., & Schwenkmezger, P. (1993). Anxiety. In R.N. Singer, M. Murphey, & L.K. Tennant (Eds.). Handbook of research on sport psychology (pp. 328-364). New York: Macmillan.

Hackfort, D., & Spielberger, C.D. (1989). Sport-related anxiety: Current trends in theory and research. In D. Hackfort & C.D. Spielberger (Eds.), Anxiety in sports: An international perspective (pp. 261-267). New York: Hemisphere.

Hanin, Y.L. (1989). Interpersonal and intragroup anxiety in sports. In D. Hackfort & C.D. Spielberger (Eds.), Anxiety in sports: An international perspective (pp. 19-28). New York: Hemisphere.

Jones, G., Swain, A., & Cale, A. (1990). Antecedents of multidimensional competitive state anxiety and self-confidence in elite intercollegiate middle-distance runners. The Sport Psychologist, 4, 107-118.

Jones, G., Swain, A., & Cale, A. (1991). Gender differences in precompetition temporal patterning and antecedents of anxiety and self confidence. Journal of Sport & Exercise Psychology, 13, 1-15.

Kenow, L. J., & Williams, J.M. (1992). Relationship between anxiety, self-confidence, and evaluation of coaching behaviors. The Sport Psychologist, 6, 344-357.

Krahe, B. (1986). Similar perceptions, similar reactions: An idiographic approach to cross situational coherence. Journal of Research in Personality, 20, 349-361.

Krahe, B. (1992). Personality and social psychology: Towards a synthesis. Newbury Park, CA: Sage.

Krane, V., Joyce, D., & Rafeld, J. (1994). Competitive anxiety, situation criticality, and softball performance. The Sport Psychologist, 8, 58-72.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Beverly Hills, CA: Sage.

Landers, D.M. (1980). The arousal-performance relationship revisited. Research Quarterly, 51, 77-90.

Lazarus, R.S., & Averill, J.R. (1972). Emotion and cognition: With special reference to anxiety. In C.D. Spielberger & I.G. Sarason (Eds.), Stress and anxiety (Vol. 1, pp 121-128). New York: Hemisphere.

Lowe, R. (1973). Stress, arousal, and task performance of Little League baseball players. Unpublished doctoral dissertation, University of Illinois, Champaign.

Martens, R., Vealey, R.S., & Burton, D. (1990). Competitive anxiety in sport. Champaign, IL: Human Kinetics.

Miles, M.B. & Huberman, A.M. (1984). Qualitative data analysis. Beverly Hills, CA: Sage.

Mischel, W. (1979). On the interface of cognition and personality. American Psychologist, 34, 740-754.

Paterson, R.J., & Neufeld, R.W.J. (1987). Clear danger: Situational determinants of the appraisal of threat. Psychological Bulletin, 101, 404-416.

Patton, M.Q. (1980). Qualitative evaluation methods. Beverly Hills, CA: Sage.

Pargman, D. (1993). Individual differences: Cognitive and perceptual styles. In R.N. Singer, M. Murphey, & L.K. Tennant (Eds.), Handbook of research on sport psychology (pp. 379-401). New York: Macmillan.

Pervin, L.A. (1978). Definitions, measurements, and classifications of stimuli, situations, and environments. Human Ecology, 6, 71-105.

Rotter, J.B. (1955). The role of the psychological situation in determining the direction of human behavior. In MR. Jones (Ed.), Nebraska symposium on motivation (pp. 245-268). Nebraska: University of Nebraska.

Sanderson, F.H., & Ashton, M.K. (1981). Analysis of anxiety levels before and after competition. International Journal of Sport Psychology, 12, 23-28.

Scanlan, T.K. (1984). Competitive stress and the child athlete. In J.M. Silva & R.S. Weinberg (Eds.), Psychological foundations of sport (pp. 118-129). Champaign, IL: Human Kinetics.

Scanlan, T.K., Stein, G.L., & Ravizza, K. (1989). An in-depth study of former elite figure skaters: II. Sources of enjoyment. Journal of Sport & Exercise Psychology, 11, 65-83.

Schmidt, R.A. (1988). Motor control and learning (2nd ed.). Champaign, IL: Human Kinetics.

Smith, C.A., & Lazarus, R.S. (1990). Emotion and adaptation. In L.A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 609-637). New York: Guildford.

Smith, R.E., Smoll, F.L., & Schutz, R.W. (1990). Measurement and correlates of sport-specific cognitive and somatic trait anxiety: The Sport Anxiety Scale. Anxiety Research, 2, 263-280.

Smoll, F.L., & Smith, R.E. (1989). Leadership behaviors in sport: A theoretical model and research paradigm. Journal of Applied Social Psychology, 19, 1522-1551.

Sokal, R.R. (1974). Classification: Purposes, principles, progress, prospects. Science, 185, 1115-1123.

Spielberger, C.D. (1966). Theory and research on anxiety. In C.D. Spielberger (Ed.), Anxiety and behavior (pp. 3-20). New York: Academic.

Spielberger, C.D. (1989). Stress and anxiety in sports. In D. Hackfort & C.D. Spielberger (Eds.), Anxiety in sports: An international perspective (pp. 3-17). New York: Hemisphere.

Thorpe, R., Bunker, D., & Almond, L. (1986). Rethinking games teaching. Northants, England: Nene Litho.

Van Heck, GL. (1989). Situation concepts: Definitions and classification. In P.J. Hettema (Ed.), Personality and environment: Assessment of human adaptation (pp. 53-69). Chichester, England: Wiley.

Wrisberg, C.A, & Pein, R.L. (1992). The preshot interval and free throw shooting accuracy: An exploratory investigation. The Sport Psychologist, 6, 14-23.

RELATED ARTICLE: Figure 1. Examples of actual responses and their classification.

Superordinate Categories and Subordinate Categories (with sport specific examples of typical response quotes)

I. Ongoing Game Play Situations

A. Offensive (Soccer: "I get a breakaway on the goalie...The player should always score.")

B. Defensive (Ice Hockey: "Defending 1 v 1 in open ice. If I get beat, they get breakaway.")

C. Offensive/Defensive Nature Unspecified (Field Hockey: "1 on 1...you're on your own...don't want to look bad.")

D. Injury (Ice Hockey: "Going into corner [to get puck] with back towards opponent. Have seen other players get hit from behind and suffer severe injuries.")

II. Game/Score/Time Criticality Situations

A. Stoppage (Basketball: "free throws in tight game; better not miss or we'll lose game.")

B. Pregame (Soccer: "Before game I worry about now knowing how other team plays.")

C. Within-Game Flow (Soccer: "Being behind in the closing minutes of an important game.")

III. Coach Related Situations

A. Personnel Decisions (Basketball: "Lack of playing time for reasons other than fouls. I want to play, and have worked hard to do so. I feel I deserve more playing time.")

B. Tactical Decision Conflicts (Ice Hockey: "Coach putting out certain players for a big play. I feel they are the wrong players for the situation.")

C. Performance Feedback (Ice Hockey: "Coach giving me *** for a bad play. Don't like being singled out for a mistake. I know I did it; he doesn't have to highlight it.")

IV. Miscellaneous Situations

A. Audience (Soccer: "Parents, friends, & lots of fans at game. Pressure not to look bad.")

B. Officiating (Field Hockey: "Poor umpiring. You rely on umpires to call game properly.")

C. Teammates (Basketball: "Non-intense performances from my teammates.)

D. Opponents (Field Hockey: "Playing against good player. Worry if I can cope.")