The Sport Grid-Revised as a measure of felt arousal and cognitive anxiety.

Ward, D. Gant ; Cox, Richard H.

Research suggests that the interaction between increased arousal and accompanying psychological mood have a combined effect upon behavior and performance (Edwards & Hardy, 1996; Hardy, 1996b; Hardy & Parfitt, 1991; Janelle, Singer, & Williams, 1999; Thelwell & Maynard, 1998; Woodman, Albinson, & Hardy, 1997). Most theories of behavior, however, fail to demonstrate exactly how changes in psychological state (e.g., cognitive anxiety) and changes in physiological arousal can interactively cause a change in skilled athletic performance. An exception to this general observation is the catastrophe model of anxiety and sports performance.

Hardy and associates (Fazey & Hardy, 1988; Hardy, 1990; Hardy & Parfitt, 1991) hypothesized a three-dimensional model that illustrates the interactive effects of arousal and cognitive anxiety on performance. Cognitive anxiety has been conceptualized as "negative expectations and cognitive concerns about oneself, the situation at hand and potential consequences" (Morris, Davis, & Hutchings, 1981, p. 541). Hardy et al.'s model, technically a cusp catastrophe model, has the ability to describe how performance can be affected by continuous changes in two independent variables--cognitive anxiety and physiological arousal (Hardy, 1996a).

The basic postulations of the catastrophe model include: (a) at a low level of arousal, an increase in cognitive anxiety will result in improved performance (Jones & Cale, 1989; Parfitt, 1988; Parfitt & Hardy, 1987); (b) at a high level of arousal, an increase in cognitive anxiety will result in decreased performance (Burton, 1988; Gould, Petlichkoff, & Weinberg, 1984; Morris et al., 1981); and (c) at a high level of cognitive anxiety, low, medium, and high levels of physiological arousal have differential effects upon performance (Edwards & Hardy, 1996; Hardy & Parfitt, 1991; Hardy, Parfitt, & Pates, 1994; Thelwell & Maynard, 1998).

Although the model would seem to have heuristic appeal, it has received only modest attention. It has been proposed that the model's complexity and the difficulty in testing it are reasons that few researchers have devoted time to its further investigation (Hardy, 1996a). We propose that improvements in instrumentation could facilitate the testing, and thereby the understanding, of the catastrophe model. Instrumentation used in the investigation must allow for the independent constructs of arousal and cognitive anxiety to vary freely, thereby allowing an examination of the entire predicted performance surface.

For variations in arousal, cognitive anxiety, and performance to be observed, and subsequently studied, multiple measures of these constructs over the course of a performance competition are necessary. However, coaches and athletes are often reluctant to participate in arousal-anxiety-performance studies that necessitate repeated measurement of physiological and psychological factors immediately before and between performances. This is understandable considering that lengthy questionnaires and physiological apparatuses can be burden-some distractions to performing athletes. The best instruments for studying the arousal-anxiety-performance relationship over the course of an actual competition will require as little of the athlete's time and attention as possible.

Another point to be considered in examining the arousal-anxiety-performance relationship is the fact that physiological and psychological states are constantly in flux. It is for this reason that physiological and psychological measures should be obtained as close to actual performance as possible. The nearer to actual performance that a measure of psychological and physiological states is obtained, the more confident one can be that the data are representative of the athlete's true psychological and physiological states immediately before and/or during performances.

The Competitive State Anxiety Inventory-2 ([CSAI-2]; Martens, Burton, Vealey, Bump, & Smith, 1990) is currently the most frequently used measure of state anxiety in sport, and measures the constructs of cognitive anxiety, somatic anxiety, and self-confidence. Rather than conceptualizing anxiety as a unidimensional construct, current thought describes the anxiety response as being composed of both psychological (cognitive) and physiological (somatic) components. Specifically, somatic anxiety has been defined as "one's perceptions of the physiological-affective elements of the anxiety experience, that is, ... unpleasant feeling states such as nervousness and tension" (Morris et al., 1981).

The fact that it takes between three and 10 minutes to complete the CSAI-2 often makes it impractical and less valid for use in an actual competitive environment where distractions and delays between assessment and performance are problematic. In attempts to overcome this obstacle and those mentioned previously, several brief instruments have been developed. Two of these instruments are short forms of the CSAI-2.

The Mental Readiness Form (MRF) was developed by Murphy, Greenspan, Jowdy, and Tammen (1989) to provide a less intrusive measure of the three CSAI-2 subscales. The MRF consists of three items (i.e., single-item subscales for cognitive and somatic anxiety and self-confidence) and can be completed in just a few seconds by making a mark on a ruled line corresponding to how one is thinking, physically feeling, and feeling confident. In 1994 Krane developed the MRF-Likert and MRF-3, effectively replacing the visual analog scale with an 1l-point Likert scale (Krane, 1994).

A second brief instrument is the Anxiety Rating Scale (ARS) (Cox, Russell, & Robb, 1998). This scale also measures cognitive anxiety, somatic anxiety, and self-confidence before and between performances. It is composed of three seven-point Likert items (one-item subscales) taken directly from the CSAI-2. A revision to the ARS was later made in order to improve athletes' interpretation of the items (Cox, Robb, & Russel, 2000).

The major limitations of the MRF and ARS in testing catastrophe models is their inability to measure the construct of arousal. Though somatic anxiety (measured by the MRF and ARS) is related to arousal, the distinction between the two constructs is critical. Hardy and associates have made considerable effort to impress upon the field of sport psychology the important distinction between physiological arousal and somatic anxiety (Hardy, 1996a, 1996b; Hardy & Parfitt, 1991 ; Hardy & Whitehead, 1984; Hardy et al., 1994). The logical argument was perhaps best presented by Hardy et al. (1994) when they stated:

There are important differences regarding the means by which physiological arousal and somatic anxiety might exert an influence upon performance. Theoretically, physiological arousal could cause 'direct hit' effects upon performance by changing the availability of certain resources to performers (Hockey & Hamilton, 1983; Humphreys & Revelle, 1984; Parfitt, Jones, & Hardy, 1990). Furthermore, a number of researchers have shown that physiological arousal which is associated with anxiety continues to fluctuate during performance (Baddeley & Idzikowski, 1983; Cox, Hallam, O'Connor, & Rachman, 1983). Conversely, somatic anxiety is hypothesized to influence performance only if subjects become preoccupied with a negative interpretation of their symptoms (Burton, 1988; Martens et al., 1990); it has also been hypothesized to dissipate once performance commences (Martens et al. 1990). (p. 328)

Further evidence for the independence of somatic anxiety and physiological arousal is the small to moderate correlations typically found between the two constructs (Hardy & Whitehead, 1984). Clearly, using the construct of somatic anxiety as a substitute for physiological arousal is problematic.

A brief instrument that is capable of measuring arousal is the Sport Grid (Raedeke & Stein, 1994; Raedeke, Stein, & Schmidt, 1993). This instrument allows for nearly unobtrusive measures of arousal and psychological mood to be obtained immediately prior to, and between, performances. Because the Sport Grid offers a measure of arousal rather than somatic anxiety, the Sport Grid is the most suited brief instrument for testing catastrophe models.

The Sport Grid accomplishes its measure of arousal by utilizing the concept of felt arousal. Raedeke and Stein (1994) defined felt arousal as simply "how aroused or activated a person felt, independent of whether the feeling associated with arousal was positive or negative" (p. 364). Although research exists which supports the importance of the directional component (i.e., the degree to which an athlete believes the intensity of symptoms will help or hinder his or her performance) in competitive anxiety's influence on performance (Jones & Hanton, 1995; Jones & Swain, 1992; Jones, Hanton, & Swain, 1994; Jones, Swain, & Hardy, 1993; Swain & Jones, 1993), catastrophe models are concerned with the intensity of arousal and cognitive anxiety. Therefore, it is the intensity of arousal and cognitive anxiety symptoms which are of importance to this study.

Asking participants to indicate their arousal level through self-report for the purposes of research is not a new concept. In fact, early investigations in physiological research informs us that self-report measures of arousal (felt arousal) are more representative of general activation than physiological measures of arousal (Thayer, 1967, 1970). A problem with commonly employed physiological measures of arousal (heart rate, skin conductance, etc.) is that these physiological systems exhibit relatively little intra-subject correlation (Lacey & Lacey, 1958). Therefore, problems arise when any single system or combination of system measurements is used to derive a meaningful indication of arousal level (Thayer, 1967). The construct of felt arousal assesses the personal perception of arousal in the athlete. Studies have been reported that show that felt arousal and physiological arousal are related (Coventry & Hudson, 2001 ; Logan et al., 2001). Consistent with the results of these reported studies, a number of researchers have used felt arousal as the sole measure of physiological arousal (Kerr & Tacon, 1999; Kerr, Yoshida, Hirata, Takai, & Yamazaki, 1997; Pietrowsky, Rudolf, Molle, Fehm, & Born, 1997).

Although the Sport Grid provides a viable measure of felt arousal and hence physiological arousal, its shortcoming in terms of testing catastrophe theory is its conceptualization of cognitive anxiety and self-confidence as bipolar opposites. The Sport Grid incorporates as one of its two dimensions "thoughts and feelings." Raedeke and Stein (1994) described this dimension as measuring "how positive and pleasant a person was thinking and feeling over-all" (pg. 364). The continuum for this dimension ranges from "extremely negative thoughts and feelings" (cognitive anxiety) to "extremely positive thoughts and feelings" (self-confidence). Research, however, does not support the conceptualization of self-confidence and cognitive anxiety as bipolar opposites (Hardy, 1990, 1996b; Jones et al., 1993, 1994; Martens et al., 1990; Parfitt & Pates, 1999; Vealey, 1990). Therefore, the combination of two distinct constructs on one dimension makes it difficult to determine exactly what the continuum is measuring.

The cusp catastrophe model of the relationships among arousal, anxiety and sport performance seems promising. This is true despite possible limitations in the model as suggested by Cohen, Pargman and Tenenbaum (in press). For example, Hardy and colleagues have continued to test the model and have proposed higher order variations of the original catastrophe model (Hardy, 1996b). Hardy (1996b) suggests that future research investigate the different catastrophe surfaces for different tasks. For this to be possible, several criteria must be met. First, instruments used to test the catastrophe model must measure the basic components of cognitive anxiety and physiological arousal, as opposed to cognitive anxiety and somatic anxiety. Second, the catastrophe model is grounded in the conceptualization that cognitive anxiety and physiological arousal are independent constructs. In order to test the catastrophe model, an instrument must be selected that has demonstrated independence of the two constructs that it measures (e.g., cognitive anxiety and arousal). A close inspection of the catastrophe model reveals that it cannot work if the constructs of cognitive anxiety and physiological arousal are not independent of each other. The perception and measurement of physiological arousal and cognitive anxiety dimensions must be independent of each other. If an increase in physiological arousal yields an automatic increase in cognitive anxiety, then the constructs are not independent. This was a major shortcoming of a study reported by Durr and Cox (1997) in that the constructs that they used to test the catastrophe model were highly correlated (cognitive and somatic anxiety). It must be possible, in the same instant, to perceive a high level of physiological arousal but a low level of cognitive anxiety, or visa versa. Finally, the instrument must be practical for use immediately before performance as well as during breaks in performance.

Consistent with the above review and discussion, the Sport Grid (Raedeke & Stein, 1994) promises to be a more appropriate instrument for field-testing the catastrophe model than the Mental Readiness Form (Krane, 1994) or the Anxiety Rating Scale (Cox, Russell, & Robb, 1998). However, because the Sport Grid confounds the measurement of cognitive anxiety with self-confidence, it is questionable if even the Sport Grid can adequately test the catastrophe model.

To address this perceived shortcoming in the Sport Grid, the revised Sport Grid (Sport Grid-R) was developed. The better instrument for field testing catastrophe models will be the inventory that can measure cognitive anxiety and arousal unambiguously, and has demonstrated independence of the two constructs that measure cognitive anxiety and arousal. Therefore, in order for the Sport Grid-R to be deemed superior to the original Sport Grid for testing catastrophe models, two specific hypotheses must be verified. These two hypotheses are as follows:

1. The correlation between cognitive anxiety as measured by the CSAI-2 and cognitive anxiety as measured by the Sport Grid-R will be larger than the correlation between cognitive anxiety as measured by the CSAI-2 and thoughts/feelings as measured by the Sport Grid

2. The correlation between cognitive anxiety and felt arousal for the Sport Grid-R will be smaller (greater independence) than the correlation between thoughts/feelings and felt arousal as measured by the Sport Grid.

Furthermore, to build additional evidence for the notion that somatic anxiety is an inappropriate substitute for physiological arousal, we sought to examine the relationship between somatic anxiety and felt arousal. As mentioned previously, physiological arousal has been found to minimally correlate with somatic anxiety (Hardy & Whitehead, 1984). Therefore, as felt arousal has been found to be an indication of physiological arousal (Coventry & Hudson, 2001; Logan et al., 2001), a third hypothesis was that there would be a small relationship, if any, between somatic anxiety and felt arousal.

Method

Participants

Participants for this study were 209 male and female intramural athletes competing in volleyball or basketball at a large Midwestern University. One hundred nine participants were male while 100 were female. One hundred forty participants played volleyball and 69 played basketball The ages of the participants ranged from 18 to 22 years (M= 19.3, SD = 1.1). One hundred five participants completed the Sport Grid-R while 104 completed the Sport Grid. Research exists which demonstrates the importance of sex as an individual difference variable in the anxiety response (Jones, Swain, & Cale, 1991; Krane, 1994). Therefore, differences by sex in cognitive and somatic anxiety as measured by the CSAI-2 were investigated before analyses. T-tests revealed no significant differences (p > .05), and the samples were subsequently combined.

Instruments

Competitive State Anxiety Inventory-2. Martens et al. (1990) developed the CSAI-2 to be a sport-specific measure of the competitive state anxiety subcomponents of somatic and cognitive anxiety. The validation process for the CSAI-2 also produced a state self-confidence component. Thus, the CSAI-2 measures the separate components of state somatic anxiety, state cognitive anxiety, and state self-confidence. The CSAI-2 is a 27-item instrument with each of the three subscales represented by nine items. An item intended to measure somatic anxiety reads, "My body feels tense." An item meant to measure cognitive anxiety reads, "I'm concerned about performing poorly." An example of a self-confidence item is, "I'm confident about performing well." Each item is completed by responding to a four-point Likert scale ranging from "Not at all" (1) to "Very much so" (4). Scores for each of the three subscales can range from nine to 36.

A meta-analysis of 15 studies investigating the interrelationships between CSAI-2 components was conducted in search of support for the components' independence (Vealey, 1990). The mean true score correlation between state cognitive and somatic anxiety was found to be .63 with a standard deviation of.07. For state cognitive anxiety and state self-confidence, the mean true score correlation was -.64 with a standard deviation of. 10. The mean true score correlation between state somatic anxiety and state self-confidence was -.60 with a standard deviation of .03. These correlational results suggest moderate dependence among the constructs measured by the CSAI-2.

Sport Grid. The Sport Grid (Raedeke & Stein, 1994; Raedeke et al., 1993) is a 9 x 9 grid that is intended to measure felt arousal level and thoughts/feelings. The vertical axis of the Sport Grid assesses felt arousal level, and the horizontal axis assesses thoughts/feelings. The felt arousal component ranges from "Extremely Low Arousal" to "Extremely High Arousal." The thoughts/feelings continuum ranges from "Extremely Negative Thoughts/Feelings" to "Extremely Positive Thoughts/Feelings." Completing the Sport Grid entails placing a mark in the box that best represents how one presently feels both mentally and physically. Separate scores for felt arousal and thoughts/feelings are obtained by converting the location of the mark on the grid to a 1-9 score for each axis.

Construction of the Sport Grid was based on the general scale-development research by Russell, Weiss, and Mendelsohn (1989) which demonstrated the affect grid's interrater reliability, and convergent, discriminant, and predictive validity. Raedeke et al. (1993) reported initial support for the Sport Grid's psychometric properties in that the Sport Grid demonstrated acceptable construct validity. Thoughts/feelings on the Sport Grid related to both self-confidence (r = .58) and cognitive anxiety (r =.47) on the CSAI-2. Felt arousal on the Sport Grid moderately correlated (r = .40) with the CSAI-2's measure of somatic anxiety. The Sport Grid's felt arousal and thoughts/feelings continua were observed to be independent (r = .11). A measure of the Sport Grid's internal item consistency is incomputable due to its being a single-item instrument.

Sport Grid-Revised. The Sport Grid-R (see Appendix) differs from the Sport Grid in that (a) while the vertical dimension is still meant to measure felt arousal, the anchors are different; (b) the horizontal dimension now measures cognitive anxiety rather than positive or negative thoughts/feelings; and (c) the instructions for the two grids are necessarily different. The procedure for completing the Sport Grid-R is the same as for the Sport Grid, as is the scoring of the dimensions (see above).

The felt arousal anchors were changed in order to make them more easily understandable to the athlete. The felt arousal anchors on the Sport Grid are "Extremely High Arousal" and "Extremely Low Arousal." "Arousal" or "aroused" are not terms typically used by athletes to describe themselves in an athletic environment. Therefore, their meaning is likely to be misunderstood, or at least less well understood than alternative terms. The Sport Grid-R has "Extremely High Activation (Extremely Pumped-Up)" and "Extremely Low Activation (Extremely Flat or Sluggish)" as anchors for the felt arousal dimension (see Appendix). While for this study "arousal" and "activation" are meant to be synonyms for felt arousal, it was thought that the term "arousal" had a greater potential to be misunderstood by the present sample of college students. The vernacular interpretations of "pumped up" and "flat or sluggish" for high and low activation are examples of terms that athletes use, and therefore believed to be better understood. Unfortunately, it is outside the scope of this particular study to investigate which pair of anchors elicits the better measure of felt arousal.

We could have chosen to use cognitive anxiety, self-confidence, or any other construct as the horizontal dimension of the Sport Grid-R. The decision to have cognitive anxiety as the horizontal axis was made with the testing of the cusp catastrophe model of arousal, anxiety, and performance in mind (Hardy, 1990).

The instructions for the Sport Grid-R are based on the instructions for Raedeke et al.'s (1994) Sport Grid. Due to the differences between the scales, the instructions for completing each one are necessarily different. Care was taken to make the instructions for each grid as similar as possible in order to better ensure that any observed differences in the scales' abilities are a reflection of their structure rather than their instructions.

Revised MRF-3 Self-Confidence Item. Because self-confidence has been shown to have an influence on performance (Hardy, 1996a; Jones & Hanton, 2001 ; Jones et al., 1993; Parfitt & Pates, 1999), the Sport Grid-R will be associated with a single-item assessment of self-confidence. We chose to look to the various MRF self-confidence items due to their simple, bipolar structure. It was determined that the MRF-3 self-confidence item was preferred to the MRF-L self-confidence item because the former showed the stronger relationship with CSAI-2 self-confidence (r = -.77) and the weaker relationship with CSAI-2 somatic anxiety (Krane, 1994). The original MRF-3 self-confidence statement reads, "I am feeling," and the Likert scale anchors for this item are confident and not confident.

In this study, we used a revised version of the MRF-3 self-confidence item. The rationale for the change lies in the idea that confident versus not confident are not appropriate anchors for a Likert scale because they do not lend themselves well to communicating differences in varying levels of self-confidence. For example, the original MRF-3 self-confidence item does not allow for the distinction between the individual who is confident and the individual who is very confident. The revision of the MRF-3 self-confidence item involved changing one of the anchor terms, changing the direction of the anchors, and a reduction in the number of points on the scale. The revised self-confidence item reads: "Relative to the upcoming competition, I am feeling...." Participants respond on a nine-point Likert scale where a score of one represents not confident, and a score of nine represents very confident. The nine-point scale was used in order to be consistent with the Sport Grid-R nine-point scales.

Procedure

Ten minutes before scheduled game time, an investigator approached volleyball or basketball teams and requested to speak with the team members. The athletes were briefed concerning the study and were informed of the voluntary nature of their participation. If any or all of the briefed individuals chose to participate, they were given a packet that contained, in order, an informed consent form, a brief demographics questionnaire, and three instruments. Half of the packets contained the Sport Grid and the other half contained the Sport Grid-R. It was thought that including both grids in all research packets would prompt participants to make the exact same mark on, or skip, the second grid thinking that it was a duplicate page. Packets were randomly distributed to participants.

After completing the demographics questionnaire, the first instrument the participants completed was either the Sport Grid or the Sport Grid-R, depending on the packet they happened to receive. Because the grids' directions are the longest and most detailed, it was thought that introducing the grids first would facilitate the likelihood that the directions were read. Second, all of the participants completed the revised MRF-3 self-confidence item. The CSAI-2 was always administered last due to the fact that it takes longer to complete the CSAI-2 than the other instruments combined. It was thought that administering the CSAI-2 first might cause participants to rush through the subsequent measures.

Results

As shown in Table 1 for the total sample, cognitive anxiety as measured by the Sport Grid-R has a larger correlation (.43) with cognitive anxiety as measured by the CSAI-2 than does the Sport Grid's measure of thoughts/feelings (-.25). A test of differences between these two independent correlations (Bruning & Kintz, 1987) approaches significance (p = .067), suggesting that the Sport Grid-R offers a better measure of cognitive anxiety than does the Sport Grid. Furthermore, the Sport Grid-R's measure of cognitive anxiety is in the same direction as CSAI-2 measures of anxiety, while the Sport Grid's measure of thoughts/feelings is not.

Interestingly, for the total sample, neither grids' felt arousal/activation component is correlated with somatic anxiety as measured by the CSAI-2 (Table 1). Also, neither grids' felt arousal/activation component is associated with cognitive anxiety as measured by the CSAI-2, while both show a slight to moderate correlation with the CSAI-2's measure of self-confidence. These results provide support for the notions that somatic anxiety is a suspect measure for arousal, and that cognitive anxiety and self-confidence are not bipolar constructs.

As shown in Table 2 for the total sample, the correlation between cognitive anxiety and felt activation for the Sport Grid-R appears to be smaller (-.07) than the correlation between thoughts/feelings and felt arousal as measured by the Sport Grid (.13). Also, cognitive anxiety as measured by the Sport Grid-R is not significantly correlated (-.19) with the single-item self-confidence scale, while thoughts/feelings as measured by the Sport Grid is (.40). This evidence in combination with other results (e.g., the direction of correlations among the thoughts/ feelings construct and the CSAI-2 sub-components) would seem to depict the Sport Grid's thoughts/feelings dimension as a measure of a construct similar to self-confidence rather than cognitive anxiety. The result that cognitive anxiety as measured by the Sport Grid-R is not significantly correlated with the single-item self-confidence scale is further evidence that cognitive anxiety and self-confidence are not bipolar constructs. The correlation of the single-item self-confidence scale with self-confidence as measured by the CSAI-2 was significant and identical for all participants (.51), regardless of whether they completed it as part of the Sport Grid or Sport Grid-R packet (Table 2).

While the above results support the three hypotheses of this study, we also used demographic data to examine differences between persons who had experience with competitive sports participation as compared with persons who had relatively little competitive sport experience. To examine differences between the two sub-samples, we divided the total pool of subjects into "Athlete" and "Non-athlete" groups. The athlete sub-sample is comprised of participants who reported having participated in sport at the high school varsity, sport club, or intercollegiate levels. The non-athlete sub-sample is those subjects who had played sport at the high school junior varsity level or lower, or who had never participated in competitive athletics.

Table I displays some interesting results for the athlete sub-sample. Related to the first hypothesis, the Sport Grid-R's measure of cognitive anxiety was significantly correlated (.41) with cognitive anxiety as measured by the CSAI-2, while thoughts/feelings as measured by the Sport Grid was not (-.21). Regarding the third hypothesis, it was again demonstrated that felt arousal and somatic anxiety are minimally correlated. Related to the second hypothesis, Table 2 displays for the athlete sample the independence of the Sport Grid-R constructs of felt activation and cognitive anxiety (r =-.11), while the Sport Grid constructs of felt arousal and thoughts/feelings were shown to be slightly to moderately dependent (r = .35). The difference between these two independent correlations (Bruning & Kintz, 1987) approaches significance (p = .07), suggesting that the Sport Grid-R constructs of felt activation and cognitive anxiety are more independent than the Sport Grid constructs of felt arousal and thoughts/feelings.

For the non-athlete population, cognitive anxiety as measured by the Sport Grid-R has a larger correlation (.48) with cognitive anxiety as measured by the CSAI-2 than does the Sport Grid's measure of thoughts/feelings (-.24) (Table I). Also, as shown in Table 2, the Sport Grid-R appears to demonstrate greater construct independence (r = .03) than does the Sport Grid (r = -. 18). Interestingly, non-athletes appeared to demonstrate a higher (though not significant) correlation between somatic anxiety and felt arousal than did the athlete subgroup.

In summary, all groups demonstrated the Sport Grid-R's measure of cognitive anxiety as having a larger correlation with cognitive anxiety as measured by the CSAI-2 than the Sport Grid's measure of thoughts/feelings. Also, all groups demonstrated that the Sport Grid-R constructs of felt activation and cognitive anxiety are more independent than are the Sport Grid constructs of felt arousal and thoughts feelings. These important findings were even more pronounced with an athlete sub-group. Finally, all groups demonstrated correlations between somatic anxiety and felt arousal that were small and not significant.

Discussion

The primary purpose of this investigation was to measure and compare the concurrent validity of the Sport Grid and the Sport Grid-R relative to the CSAI-2. The three hypotheses tested were: (a) The correlation between cognitive anxiety as measured by the CSAI-2 and cognitive anxiety as measured by the Sport Grid-R would be larger than the correlation between cognitive anxiety as measured by the CSAI-2 and thoughts/feelings as measured by the Sport Grid, (b) the correlation between cognitive anxiety and felt arousal for the Sport Grid-R would be smaller (greater independence) than the correlation between thoughts/feelings and felt arousal as measured by the Sport Grid, and (c) there would be a small relationship, if any, between somatic anxiety and felt arousal.

The first hypothesis of this investigation was supported. The Sport Grid-R's measure of cognitive anxiety had a larger correlation with cognitive anxiety as measured by the CSAI-2 than did the Sport Grid's measure of thoughts/feelings. Although the Sport Grid-R offers a better measure of cognitive anxiety, the magnitude of the grid's correlation with cognitive anxiety as measured by the CSAI-2 was lower than expected. Based upon previous research involving the CSAI-2 and instruments providing brief indicators of cognitive and somatic anxiety, we expected the correlation to be .50 to .70 (Cox, Russell, & Robb, 1999). The correlation between cognitive anxiety as measured by the CSAI-2 and the Sport Grid-R was .43. The original validation study for the Sport Grid (Raedeke & Stein, 1994) reported that the correlation between thoughts/feelings and the CSAI-2 cognitive anxiety subscale was -.47, while in the present study this correlation was -.25 for the original Sport Grid.

There are several possible explanations as to why this correlation was not higher in the present investigation. One possible explanation involves the formal education of participants relative to the meaning of the construct of cognitive anxiety. In the present investigation, participants were asked to follow instructions to complete their grid, including a description and example of cognitive anxiety (see Appendix). It is probable that when athletes are educated to more thoroughly understand the concept of cognitive anxiety prior to completion of the Sport Grid-R, the correlation between the Sport Grid-R's cognitive anxiety component and cognitive anxiety as measured by the CSAI-2 would increase. A second explanation relates to the difference in wording of the cognitive anxiety items in the CSAI-2 and the Sport Grid-R. In the Sport Grid-R we used the anchor term "worried" as opposed to "concerned" as used in the CSAI-2. We made this change consistent with research reported by Lane, Sewell, Terry, Bartram, and Nesti (1999), in which they argued that the word "worried" more adequately captures the essence of the meaning of cognitive anxiety than does the word "concerned."

The second hypothesis of this investigation was also supported. The correlation between cognitive anxiety and felt arousal for the Sport Grid-R was smaller (less dependent) than the correlation between thoughts/feelings and felt arousal as measured by the Sport Grid. This is a critical finding relative to the viable prospect of examining catastrophe models in a field setting. The chosen instrument must (a) provide valid measures of the two constructs hypothesized in the cusp catastrophe model (i.e., cognitive anxiety and physiological arousal), and (b) the two constructs must be independent of each other as evidenced by previous testing of the instrument. Based on support for hypotheses one and two, we believe that this investigation provides research support for the use of the Sport Grid-R to test catastrophe models in a field setting.

The third hypothesis in this investigation was also supported. The small correlation between felt arousal and somatic anxiety suggested strong support for the independence of the two constructs. Because physiological arousal has been found to be minimally correlated with somatic anxiety, and because felt arousal is thought to indicate physiological arousal, it was expected that somatic anxiety and felt arousal would also demonstrate a weak relationship. This result would seem to build further support for a) the use of felt arousal as an indication of physiological arousal and b) the importance of refraining from using somatic anxiety as an indication of physiological arousal.

In previous laboratory based research involving the catastrophe model, physiological arousal has been systematically and successfully manipulated (Cohen et al., in press; Hardy & Parfitt, 1991; Hardy et al., 1994). Successful manipulation of cognitive anxiety, however, has been a much more difficult problem. Hardy and Parfitt (1991), for example, manipulated cognitive anxiety as a function of time and importance of contest, while Cohen et al. (in press) manipulated it as a function of threat of shock. A strong argument can be made that neither method of manipulation was particularly effective.

In previous field based research, somatic anxiety was used as a proxy for physiological arousal (Durr & Cox, 1997; Krane, Joyce & Rafield, 1994). The primary shortcoming of both of these investigations was that somatic anxiety and physiological arousal are not equivalent constructs and somatic anxiety and cognitive anxiety are moderately correlated.

As supported by the discussion presented above, the Sport Grid-R is the preferred instrument for testing catastrophe models of sports performance for the following reasons: (1) it provides a measure for the critical constructs of arousal and cognitive anxiety, (2) the measurement of these two constructs is independent, and (3) it can be administered immediately and repeatedly before and during athletic competition. Future research may utilize the Sport Grid-R to test the basic catastrophe model, as well as to investigate the specific performance surfaces of particular sports and skills (e.g., public speaking, test-taking, etc.).

Future research may also use the Sport Grid-R in combination with another grid or instrument to investigate higher order catastrophe models. For example, the self-confidence ("not confident--very confident") item used and validated in this investigation could be combined with a measure of self-control to form an entirely new grid, which in conjunction with the Sport Grid-R, could be used to test a higher order butterfly catastrophe model proposed by Hardy (1990). Finally, further knowledge concerning the relationship between felt arousal and physiological arousal, as well as between somatic anxiety and physiological arousal, would benefit the investigation of anxiety and arousal's relationship with sports performance.

Appendix

Sport Grid-Revised

Athletes experience a variety of thoughts and feelings before and during competitions. The Sport Grid-R allows you to describe some of these thoughts and feelings. The instructions for completing the grid are as follows:

1. Put an X in the box that indicates both the degree to which your body feels activated and the extent to which you are worried about your individual performance.

Top to bottom, the grid measures how activated or "pumped-up" your body feels--regardless of whether the feeling is positive or negative. The higher you go on the grid, the more activated your body feels.

Left to right, the grid measures how worried you are about your performance (not your team's performance) in the upcoming event. The further right you go, the more worried about your individual performance you are.

EXAMPLE:

[ILLUSTRATION OMITTED]

* From this completed example, we can speculate that this athlete is feeling a little sluggish and he/she is quite worried about performing poorly in the upcoming competition.

Table 1

Sport Grid and Sport Grid--Revised Correlations with CSAI-2 Subscales

 CSAI-2
 Subscales

 Somatic
Sample Grid N Subscale Anxiety

Total Sport Grid 104 Felt Arousal .06
 104 Thoughts/Feelings .35 *
 Sport Grid-R 105 Felt Activation .04
 105 Cognitive Anxiety .42 *
Athletes Sport Grid 68 Felt Arousal -.01
 68 Thoughts/Feelings -.18
 Sport Grid-R 65 Felt Activation -.10
 65 Cognitive Anxiety .30 *
Non-athletes Sport Grid 36 Felt Arousal .26
 36 Thoughts/Feelings -.53 *
 Sport Grid-R 40 Felt Activation .26
 40 Cognitive Anxiety .60 *

 CSAI-2
 Subscales

 Cognitive
Sample Grid N Subscale Anxiety

Total Sport Grid 104 Felt Arousal -.06
 104 Thoughts/Feelings -.25
 Sport Grid-R 105 Felt Activation .07
 105 Cognitive Anxiety .43 *
Athletes Sport Grid 68 Felt Arousal -.09
 68 Thoughts/Feelings -.21
 Sport Grid-R 65 Felt Activation .03
 65 Cognitive Anxiety .41 *
Non-athletes Sport Grid 36 Felt Arousal .20
 36 Thoughts/Feelings -.24
 Sport Grid-R 40 Felt Activation .13
 40 Cognitive Anxiety .48 *

 CSAI-2
 Subscales

 Self-
Sample Grid N Subscale Confidence

Total Sport Grid 104 Felt Arousal .37 *
 104 Thoughts/Feelings .33 *
 Sport Grid-R 105 Felt Activation .23 *
 105 Cognitive Anxiety -.32 *
Athletes Sport Grid 68 Felt Arousal .30 *
 68 Thoughts/Feelings .25 *
 Sport Grid-R 65 Felt Activation .32 *
 65 Cognitive Anxiety -.24
Non-athletes Sport Grid 36 Felt Arousal .35 *
 36 Thoughts/Feelings .43 *
 Sport Grid-R 40 Felt Activation .03
 40 Cognitive Anxiety .44 *

* p <.05

Table 2

Intercorrelations Between Sport Grid (SG) and Sport Grid-Revised (SG-R)
Constructs

 Grid Constructs

 Grid Construct 1 2 3

TOTAL
1. SG (Arousal) 1.00 .13 .41 *
2. SG (thoughts/feelings) 1.00 .40 *
3. SG (Single-item Self-confidence) 1.00
4. SG-R (Activation)
5. SG-R (Cognitive Anxiety)
6. SG-R (Single-item Self-Confidence)
7. CSAI-2 Self-confidence

ATHLETES
1. SG (Arousal) 1.00 .35 * .44 *
2. SG (thoughts/feelings) 1.00 .35
3. SG (Single-item Self-confidence) 1.00
4. SG-R (Activation)
5. SG-R (Cognitive Anxiety)
6. SG-R (Single-item Self-confidence)
7. CSAI-2 Self-confidence

NONATHLETES
l. SG (Arousal) 1.00 -.18 .23
2. SG (thoughts/feelings) 1.00 .41 *
3. SG (Single-item Self-confidence) 1.00
4. SG-R (Activation)
5. SG-R (Cognitive Anxiety)
6. SG-R (Single-item Self-confidence)
7. CSAI-2 Self-confidence

 Grid Constructs

 Grid Construct 4 5 6

TOTAL
1. SG (Arousal) -- --
2. SG (thoughts/feelings) -- --
3. SG (Single-item Self-confidence) -- -- --
4. SG-R (Activation) 1.00 -.07 .30 *
5. SG-R (Cognitive Anxiety) 1.00 -.19
6. SG-R (Single-item Self-Confidence) 1.00
7. CSAI-2 Self-confidence

ATHLETES
1. SG (Arousal) --
2. SG (thoughts/feelings) -- -- --
3. SG (Single-item Self-confidence) -- --
4. SG-R (Activation) 1.00 .11 .36 *
5. SG-R (Cognitive Anxiety) 1.00 -.11
6. SG-R (Single-item Self-confidence) 1.00
7. CSAI-2 Self-confidence

NONATHLETES
l. SG (Arousal) -- -- --
2. SG (thoughts/feelings) -- -- --
3. SG (Single-item Self-confidence) -- -- --
4. SG-R (Activation) 1.00 .03 .15
5. SG-R (Cognitive Anxiety) 1.00 -.32 *
6. SG-R (Single-item Self-confidence) 1.00
7. CSAI-2 Self-confidence

 Grid Constructs

 Grid Construct 7

TOTAL
1. SG (Arousal)
2. SG (thoughts/feelings)
3. SG (Single-item Self-confidence) .51 *
4. SG-R (Activation) --
5. SG-R (Cognitive Anxiety) --
6. SG-R (Single-item Self-Confidence) .51 *
7. CSAI-2 Self-confidence 1.00

ATHLETES
1. SG (Arousal)
2. SG (thoughts/feelings) --
3. SG (Single-item Self-confidence) .43 *
4. SG-R (Activation) --
5. SG-R (Cognitive Anxiety) --
6. SG-R (Single-item Self-confidence) .40 *
7. CSAI-2 Self-confidence 1.00

NONATHLETES
l. SG (Arousal) --
2. SG (thoughts/feelings)
3. SG (Single-item Self-confidence) .52 *
4. SG-R (Activation) --
5. SG-R (Cognitive Anxiety) --
6. SG-R (Single-item Self-confidence) .67 *
7. CSAI-2 Self-confidence 1.00

* p <.01

References

Baddeley, A., & Idzikowski, C. (1983). Waiting in the wings: Apprehension, public speaking and performance. Ergonomics, 26, 575-583.

Bruning, J.L., & Kintz, B.L. (1987). Computaional handbook of statistics (3rd ed., pp. 226-228). Scott, Foresman and Company.

Burton, D. (1988). Do anxious swimmers swim slower? Re-examining the elusive anxiety performance relationship. Journal of Sport and Exercise Psychology, 10, 45-61.

Cohen, A., Pargman, D., & Tenenbaum, G. (In press). Critical elaboration and empirical investigation of the cusp catastrophe model. The Sport Psychologist.

Coventry, K.R., & Hudson, J. (2001) Gender differences, physiological arousal and the role of winning in fruit machine gamblers. Addiction, 96, 871-879.

Cox, D., Hallam, R., O'Conner, K., & Rachman, S. (1983). An expirimental analysis of fearlessness and courage. British Journal of Psychology, 74, 107-117.

Cox, R.H., Robb, M., & Russell, W.D. (2000). Concurrent validity of the Revised Anxiety Rating Scale. Journal of Sport Behavior, 23, 327-334.

Cox, R.H., Russell, W.D., & Robb, M. (1998). Development of a CSAI-2 short form for assessing competitive state anxiety during and immediately prior to competition. Journal of Sport Behavior, 21, 30-40.

Cox, R.H., Russell, W.D., & Robb, M. (1999). Comparative concurrent validity of the MRF-L and ARS competitive anxiety rating scales for volleyball and basketball. Journal of Sport Behavior, 22, 310-320.

Durr, K.R., & Cox, R.H. (1997). Relationship between state anxiety and performance in high school divers: A test of catastrophe theory. Research Quarterly for Exercise and Sport, 68 (1), Supplement, Abstract 102.

Edwards, T., & Hardy, L. (1996). The interactive effects of intensity and direction of cognitive and somatic anxiety and self-confidence upon performance. Journal of Sport & Exercise Psychology, 18, 296-312.

Fazey, J.A., and Hardy, L. (1988). The inverted-U hypothesis: A catastrophe for sport psychology (British Association of Sports Sciences Monograph No. 1). Leeds, United Kingdom: The National Coaching Foundation.

Gould, D., Petlichkoff, L., & Weinburg, R.S. (1984). Antecedents of temporal changes in, and relationships between CSAI-2 sub-components. Journal of Sport Psychology, 6, 289-304.

Hardy, L. (1990). A catastrophe model of performance in sport. In J.G. Jones & Hardy (Eds.), Stress and performance in sport (pp. 81-106). England: Wiley.

Hardy, L. (1996a). Testing the predictions of the cusp catastrophe model of anxiety and performance. The Sport Psychologist. 10, 140-156.

Hardy, L. (1996b). A test of catastrophe models of anxiety and sports performance against multidimensional anxiety theory models using the method of dynamic differences. Anxiety, Stress, and Coping, 9, 69-86.

Hardy, L., & Parfitt, G.(1991). A catastrophe model of anxiety and performance. British Journal of Psychology, 82, 163-178.

Hardy, L., & Whitehead, R. (1984). Specific modes of anxiety and arousal. Current Psychological Research and Reviews, 3, 14-24.

Hardy, L., Parfitt, G., & Pates, J. (1994). Performance catastroophes in sport: A test of the hysteresis hypothesis. Journal of Sport Sciences. 12, 327-334.

Hockey, R., & Hamilton, P. (1983). The cognitive patterning of stress states. In R. Hockey (Ed.), Stress and Fatigue in Human Performance (pp. 331-362). Chichester: John Wiley.

Humphreys, M. S., & Revelle, W. (1984). Personality, motivation and performance: A theory of the relationship between individual differences and information processing. Psychological Review, 91, 153-184.

Janelle, C. M., Singer, R. N., & Williams, A. M. (1999). External distraction and attentional narrowing: Visual search evidence. Journal of Sport & Exercise Psychology, 21, 70-91.

Jones, J.G., & Cale, A. (1989). Precompetition temporal patterning of anxiety and self-confidence in males and females. Journal of Sport Behavior, 12, 183-195.

Jones, J. G., & Hanton, S. (1996). Interpretation of competitive anxiety symptoms and goal attainment expectancies. Journal of Sport & Exercise Psychology. 18(2), 144-157.

Jones, G., & Hanton, S. (2001). Pre-competitive feeling states and directional anxiety interpretations. Journal of Sports Sciences, 19, 385-395.

Jones, G., & Swain, A. B. J. (1992). Intensity and direction dimensions of competitive state anxiety and relationships with competitiveness. Perceptual and Motor Skills, 74, 467-472.

Jones, G., Hanton, S., & Swain, A. (1994). Intensity and interpretation of anxiety symptoms in elite and non-elite sports performers. Personality and Individual Differences, 17, 657-663.

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

Jones, G., Swain, A., & Hardy, L. (1993). Intensity and direction dimensions of competitive state anxiety and relationships with performance. Journal of Sports Sciences, 11,525-532.

Kerr, J.H., & Tacon, P. (1999). Psychological responses to different types of locations and activities. Journal of Environmental Psychology, 19, 287-294.

Kerr, J.H., Yoshida, H., Hirata, C., Takai, K., & Yamazaki, F. (1997). Effects on archery performance of manipulating metamotivational state and felt arousal. Perceptual and Motor Skills, 84, 819-828.

Krane, V. (1994). The mental readiness form as a measure of competitive state anxiety. The Sport Psvchologist, 8, 189-202.

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

Lacey, J.L.& Lacey, B.C. (1958). Verification and extension of the principle of autonomic response-stereotypy. American Journal of Psychology, 71, 50-73.

Lane, A.M., Sewell, D.F., Terry, P.C., Bartram, D., & Nesti, M.S. (1999). Confirmatory factor analysis of the Competitive State Anxiety Inventory-2. Journal of Sports Sciences, 17, 505-512.

Logan, H., Lutgendorf, S., Rainville, P., Sheffield, D., Iverson, K., & Lubaroff, D. (2001). Effects of stress and relaxation on capsaicin-induced pain. Journal of Pain, 2, 160-170.

Martens, R., Burton, D., Vealey, R.S., Bump, L.A., Smith, D.E. (1990). Development and validation of the competitive state anxiety inventory-2. In R. Martens, R., R.S. Vealey, & D. Burton, Competitive anxiety in sport (pp. 117-190). Champaign, IL: Human Kinetics.

Morris, L.W., Davis, D., & Hutchings, C. (1981). Cognitive and emotional components of anxiety: Literature review and revised worry-emotionality scale. Journal of Educational Psychology, 73, 541-555.

Murphy, S., Greenspan, M., Jowdy, D., & Tammen, V.(1989, October). Development of a brief rating instrument of competitive anxiety: Comparison with the CSAI-2. Paper presented at the meeting of the Association for the Advancement of Applied Sport Psychology, Seattle, WA.

Parfitt, C.G. (1988). Interactions between models of stress and models of motor control.Unpublished PhD thesis, University College of North Wales.

Parfitt, G., & Hardy, L. (1987). Further evidence for the differential effects of competitive anxiety upon a number of cognitive and motor sub-systems. Journal of Sports Sciences, 5, 62-63.

Parfitt, G. & Pates, J. (1999). The effects of cognitive and somatic anxiety and self-confidence on components of performance during competition. Journal of Sports Sciences, 17, 351-356.

Parfitt, C. G., Jones, J. G., & Hardy, L. (1990). Multidimensional anxiety and performance. In J. G. Jones & L. Hardy (Eds.), Stress and Performance in Sport (pp. 43-80). Chichester: John Wiley.

Pietrowsky, R., Rudolf, S., Molle, M., Fehm, H.L., & Born, J. (1997). Cholecystokinin-induced effects on selective attention depend on level of activation.Neuropsychobiology, 36, 87-95.

Raedeke, T.D., Stein, G.L., & Schmidt, G. (1993, June). A new look at arousal: A two-dimensional conceptualization. Paper presented at the North American Society for the Psychology of Sport and Physical Activity Conference. Brainerd, MN.

Raedeke, T. D., & Stein, G.L. (1994). Felt arousal, thoughts/feelings, and ski performance. The Sport Psychologist, 8, 360-375.

Russel, J.A., Weiss, A., & Mendelsohn, G.A. (1989). Affect grid: A single-item scale of pleasure and arousal. Journal of Personality and Social Psychology, 57, 493-502.

Schachter, S. (1964). The interaction of cognitive and physiological determinants of emotional state. In L. Berkowitz (Ed.) Advances in experimental social psychology (Vol. 1, pp. 49-80). New York: Academic Press.

Swain, A., & Jones, G. (1993). Intensity and frequency dimensions of competitive state anxiety. Journal of Sports Sciences, 11, 533-542.

Thayer, R. E. (1967). Measurement of activation through self-report. Psychological Reports, 20, 663-678.

Thayer, R. E. (1970). Activation states as assessed by verbal report and four psycho-physiological variables. Psychophysiology, 7, 86-94.

Thelwell, R. C., & Maynard, I. W. (1998). Anxiety-performance relationships in criketers: Testing the zone of optimal functioning hypothesis. Perceptual and Motor Skills, 87, 675-689.

Vealey, R.S. (1990) Advancements in competitive anxiety research: Use of the sport competition anxiety test and the competitive state anxiety inventory-2. Anxiety Research, 2, 243-261.

Woodman, T., Albinson, J.G., & Hardy, L. (1997). An investigation of the zones of optimal functioning hypothesis within a multidimensional framework. Journal of Sport and Exercise Psychology, 19, 131-141.

Send Correspondence To: Gant Ward, University of Missouri-Columbia, Department of Educational and Counseling Psychology, 16 Hill Hall, Columbia, MO 65211. Email Address: [email protected]. Telephone:(405)321-4733. Fax:(405)321-4733.