Concurrent Validity of the Revised Anxiety Rating Scale.

Cox, Richard H. ; Robb, Marshall ; Russell, William D. 等

The Anxiety Rating Scale (ARS) is a shortened version of the Competitive State Anxiety Inventory - 2 (CSAI-2). Subsequent use of the ARS led to the suggestion that minor modifications in the ARS might enhance the concurrent validity of the instrument. It was the purpose of this investigation to compare the concurrent validity of the original ARS with two revisions of the ARS. Participants in the investigation were 180 college age intramural volleyball participants. Results revealed that for females, revision two yielded the largest correlations with CSAI-2 subscales; for males, revision two again yielded the largest correlations relative to cognitive and somatic anxiety, but not for self-confidence. Collapsing across gender revision two yielded the highest correlations for cognitive anxiety, somatic anxiety, and self-confidence. These correlation coefficients were .67, .69 and .75 respectively. It was concluded that revision two exhibited the highest level of concurrent validity when research participants wer e intramural volleyball players.

The Anxiety Rating Scale (ARS) was developed by Cox, Russell and Robb (1998, 1999) as a short rating scale version of the Competitive State Anxiety Inventory - 2 (CSAI-2) (Martens, Burton, Vealey, Bump, & Smith, 1990). The ARS was not merely conceptualized as a short version of the CSAI-2 with fewer items devoted to each anxiety construct, but as a single statement rating scale that would allow an athlete to quickly rate how they felt in terms of precompetitive cognitive state anxiety, somatic state anxiety, and self-confidence. In this regard, it was patterned after the Borg Scale (1973) designed to estimate perceived exertion during exercise. Evidence of concurrent validity of the ARS with the CSAI-2 has been provided with team sport intramural athletes (Cox et al., 1998, 1999) and with individual sport intramural athletes (Cox, Reed, & Robb, 1997). Results of these investigations have shown the ARS to be moderately correlated with anxiety and self-confidence subcomponents of Martens' CSAI-2 (.60 to .70).

The ARS has been compared favorably with Krane's (1994) modification of the Mental Readiness Form (MRF-L). The MRF-L is a short version of the CSAI-2 originally developed by Murphy, Greenspan, Jowdy, and Tammen (1989). Comparative correlations between the anxiety components of the ARS and MRF-L with the cognitive and somatic state anxiety subscales of the CSAI-2 have consistently been higher for the ARS (Cox, Reed, & Robb, 1997; Cox et al., 1999). As originally developed (Cox et al., 1998), the ARS measured competitive somatic state anxiety and competitive cognitive state anxiety, but not state self-confidence. State self-confidence was later added to the ARS (Cox et al., 1999).

In crafting the statements associated with cognitive state anxiety, somatic state anxiety, and state self-confidence, the developers of the ARS used stepwise multiple regression procedures to identify the "best" three items from the CASI-2 to represent each psychological state. These CSAI-2 items were then crafted into aggregate statements that represent cognitive anxiety, somatic anxiety, and self-confidence. The three aggregate statements were then set to a 7-point scale ranging from I (not at all) to 7 (intensely so), which allowed an athlete to rate how she felt immediately prior to competition. Respectively, the three statements for cognitive state anxiety, somatic state anxiety, and self-confidence read as follows:

1. I feel concerned about performing poorly and that others will be disappointed with my performance.

2. I feel nervous, my body feels tight and/or my stomach tense.

3. I feel secure, mentally relaxed, and confident of coming through under pressure.

Anonymous critics have argued that each of these statements contain more than one psychological construct or idea, and therefore the athlete might be confused as to how to respond. For example, the third statment contains the psychological constructs of "secure," "relaxed" and "confident." As the arguement goes, the athlete might feel secure and confident, but not relaxed, and therefore would not know how to respond to this statement on a 7-point scale. Evidence suggests, however, that a respondent would simply provide a "global" response as to how they felt relative to feeling secure, relaxed, and confident (Cox, Russell & Robb, 1996). In the developmental phase of the ARS, Cox et al., (1996) reported data that showed that an aggregate statement that included three psychological constructs in a single statement had a higher concurrent correlation with the parent CSAI-2 subscale than a statement that included only the element that had the highest simple correlation with the CSAI-2 subscale. Specifically, the correlations between ARS measured cognitive and somatic state anxiety and the corresponding CSAI-2 measured subscales were reported to be .717 and .761. The correlations between the single cognitive anxiety CSAI-2 item "I'm concerned about performing poorly" and the single somatic anxiety CSAI-2 item "I feel nervous" with the cognitive and somatic subscales of the CSAI-2 were .497 and .630 respectively. This was taken as evidence that an aggregated statement was superior to a single unambiguous statement in predicting cognitive and somatic state anxiety as measured by the CSAI-2.

A second criticism of the ARS is that some of the psychological constructs that are included in each statement seem to cause some confusion in the mind of the responder relative to whether it is a cognitive anxiety, somatic anxiety, or self-confidence statement. For example, the word "nervous" which appears in the somatic anxiety statment might be interpreted by the respondent as a cognitive state anxiety construct. And again, the phrase "mentally relaxed" which appears in the self-confidence statement might be interpreted by the respondent as a somatic anxiety statement. This confusion is possible, despite the fact that the phrase "I feel nervous" is scored as a somatic anxiety element in the CSAI-2 parent inventory, and the phrase "I feel mentally relaxed" is scored as a self-confidence element in the CSAI-2 parent test.

It was the purpose of this investigation to develop revised versions of the Anxiety Rating Scale, and to compare them with the parent CSAI-2 for concurrent validity, in much the same way that the ARS was compared with the MRF-L for concurrent validity (Cox, et al., 1999). It was hypothesized that revised versions of the ARS would exhibit larger concurrent validity coefficients with the CSAI-2 than the original Anxiety Rating Scale (ARS).

Method

Participants

Participants for this research were 90 male and 90 female (N 180) intramural athletes (largely undergraduates) competing in volleyball at a large Midwest University. The average age of the male participants was 19.8 years (SD = 1.6) and the average of female participants was 19.3 years (SD = 1.4). Use of human subjects for the purpose of research was approved by the appropriate University human subject's committee and anonymity assured.

Instruments

The instruments used for collecting anxiety data were the Competitive State Anxiety lnventory-2 (CSAI-2), the Anxiety Rating Scale (ARS), and two revised versions of the ARS (ARS-l, ARS-2). The CSAI-2 is composed of 27-items that measure cognitive state anxiety, somatic state anxiety, and self-confidence. Details associated with the development and testing of the CSAI-2 are reported by Martens, Vealey, and Burton (1990). The Anxiety Rating Scale (ARS) was previously described in some detail in the introduction of this article.

In addition to the CSAI-2 and the ARS, two revised versions of the ARS were developed for testing. Details associated with the development of the two revised versions of the ARS are described in the following sections.

ARS revision one. The primary difference between the original ARS and revision one (ARS-l) was in the replacement of the word "nervous" with the word "jittery" in the somatic state anxiety statement, and the replacement of the phrase "mentally relaxed" with the phrase "I feel comfortable" in the self-confidence statement. The word "nervous" and the phrase "mentally relaxed" were replaced because of their potential ambiguous meaning in the minds of some respondents. Some respondents have reported that the word "nervous" brings to mind thoughts that are more mental than somatic. Similarly, some respondents have reported that the phrase "mentally relaxed" brings to mind thoughts that are more related to bodily tension/anxiety than to self-confidence.

Focusing upon these two words/phrases, we went back into the original data set from which the ARS was developed (Cox et al., 1998) to identify logical replacements. In the original data set the correlation between the CSAI-2 subscale for somatic anxiety and the phrase "I feel nervous" item was .778, while the same relationship for the phrase "I feel jittery" was .777. Additionally, the intercorrelation between these two phrases was relatively high, .69, providing a possible explanation why both were not included in the best three variable prediction model. Based upon this information, we replaced the word "nervous" in the somatic state anxiety statement of the ARS with the word "jittery." Similarly, we went back into the original data set to find a replacement for the phrase "mentally relaxed." In the original data set, the correlation between the CSAI-2 subscale for self-confidence and the "I feel mentally relaxed" item was .68, while the same correlation with the "I feel comfortable" item was .71. The two items were likely not included together in the best three variable model because of the relatively high correlation between the two items (r = .50) and their intercorrelations with other self-confidence items. With these modifications, the statements that appear in ARS revision one (ARS-1), with respect to cognitive anxiety, somatic anxiety, and self-confidence, are as follows:

1. I feel concerned about performing poorly and that others will be disappointed with my performance.

2. I feel jittery, my body feels tight and/or my stomach tense.

3. I feel comfortable, secure, and confident of coming through under pressure.

All three statements were anchored with a 7-point rating scale ranging from I (not at all) to 7 (intensely so).

ARS revision two. Revision two differed significantly from the original and the first revision. The ARS and ARS-l were both developed and based on the data from which the ARS was originally developed. ARS revision two (ARS-2), however, was developed with an entirely different frame of reference. We looked at the nine items for each subcomponent of the GSA I2 and selected three items that seemed to most logically be related to cognitive anxiety, somatic anxiety, or self-confidence. Thus, the ARS-2 was based on the collective judgement of the researchers. We selected CSAI-2 items 13, 16 and 22 to form the basis of the single aggregate statement for cognitive state anxiety; items 5,8 and 17 to form the basis of the somatic state anxiety statement; and items 6, 12, and 18 to form the basis of the self-confidence statement. The aggregate statements that appear in ARS revision two (ARS-2), with respect to cognitive anxiety, somatic anxiety, and self-confidence, were as follows:

1. I feel concerned about performing poorly, choking under pressure, and that others will be disappointed with my performance.

2. I feel jittery, my body feels tense, and my heart is racing.

3. I feel comfortable, secure, and confident about performing well.

As with the ARS and ARS-1, the ARS-2 was achored with a 7-point rating scale.

Procedures

Starting members of intramural volleyball teams were approached approximately 15 mm prior to a round-robin competition and asked if they would participate in the study. Volunteers were given a pencil and an inventory packet that included instructions, consent form, an ARS version, and the CSAI-2, in that order. In completing the ARS and the CSAI-2, they were asked to respond relative to how they felt at that moment. The inventories were distributed to the athletes in random order, so that each athlete was as likely to receive a packet containing the ARS, ARS-l or ARS-2. Inventories were distributed in such a fashion, that the ARS was administered to 30 men and 30 women, and the same for the ARS- 1 and ARS-2. Teams were not approached for their participation more than one time during round-robin competition.

Results

Correlation procedures were used to compare the concurrent validity of participants who used the three different versions of the ARS. Each reported correlation coefficient was based upon 30 participants, unless gender was pooled, in which case each correlation was based upon 60 participants.

Correlation coefficients between CSAI-2 subscales and the three versions of the Anxiety Rating Scale are illustrated in Table 1. In this Table, data are displayed separately by gender, but also by collapsing across gender. As can be observed in this Table, the highest concurrent validity coefficients (correlations) for the females were obtained using the ARS-2, for cognitive anxiety, somatic anxiety, as well as self-confidence. The original version of the ARS produced the lowest correlation coefficients among the three measures of anxiety and confidence. Based upon this result, we might conclude that the ARS-2 is the preferred version of the scale.

Focusing upon the male participants, however, we get a slightly different result. Relative to cognitive and somatic state anxiety, the ARS-2 still exhibits the highest correlations with CSAI-2 subscales, but in the case of self-confidence, the original version of the ARS yielded the highest correlation coefficient. Collapsing across gender, however, we get a result that suggests that the ARS-2 is the preferred version for predicting cognitive anxiety, somatic anxiety, and self-confidence as measured by the CSAI-2.

Discussion

Three different versions of the Anxiety Rating Scale were compared for concurrent validity with the Competitive State Anxiety Inventory -2 (CSAI-2). For the female participants, the results seem to demonstrate the superiority of the ARS-2 over the ARS and the ARS-l for cognitive state anxiety, somatic state anxiety and self-confidence. Although there were some differences, this pattern of superiority remained generally consistent for the male participants. For cognitive state anxiety, the ARS-2 continued to exhibit a clear superiority over the ARS and ARS-1. For somatic state anxiety, the ARS-2 exhibited a correlation coefficient that was slightly larger than the correlation for the ARS, but clearly larger than for the ARS-1. In the case of self-confidence, however, the pattern changes suggesting that the ARS remains the best predictor of this subscale of the CSAI-2. Collapsing the two samples together (males and females), however, we again have a pattern of results that suggests that the ARS-2 is the best p redictor of the CSAI-2 subscales of cognitive state anxiety, somatic state anxiety, and self-confidence.

In this data set, using volleyball participants, the evidence seems to suggest that the second revised version of the ARS (ARS-2) is the preferred version. In terms of cognitive and somatic state anxiety, its superiority over the original ARS and the ARS-1 is fairly consistent, regardless of gender. In terms of self-confidence, the evidence favoring the ARS-2 is less convincing, although when the male and female samples are pooled, it does exhibit the largest correlation coefficient. Based upon these observations, it is concluded that the concurrent validity of the Anxiety Rating Scale can be improved by modifying it consistent with revision two (ARS-2).

In future investigations, it is recommended that the construct validity of the ARS-2 be tested. Construct validity of the ARS-2 could be established by demonstrating that groups of participants who theoretically should differ on state anxiety and confidence, do in fact differ on these psychological constructs as measured by the Anxiety Rating Scale (Thomas & Nelson, 1996).

References

Borg, G.A.V. (1973). Perceived exertion: A note on "history" and methods. Medicine and Science in Sports, 5, 90-93.

Cox, R.H., Reed, C., & Robb, M. (1997). Comparative validity of the MRF-L and ARS competitive state anxiety rating scales for intramural athletes competing in six individual sports. Research Quarterly for Exercise and Sport, 68(1), Supplement, A-101.

Cox, R.H., Russell, W.D., & Robb, M. (1996). Five phases in the development of an instrument for assessing competitive state anxiety during and prior to competition. Unpublished manuscript, University of Missouri - Columbia.

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, 31-40.

Cox, R.H., Russell, W.D., & Robb, M. (1999). Comparative concurrent validity of the MRF-L. and ARS competitive state anxiety rating scales for volleyball and basketball. Journal of Sport Behavior, 22, 1 - 11.

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

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

Murphy, S.M., Greenspan, M., Jowdy, D., & Tammen, V. (1989). Development of a brief rating instrument of competitive anxiety: Comparisons with the Competitive State Anxiety Inventory-2. Proceedings of the Association for the Advancement of Applied Sport Psychology (p.82). Seattle, WA.

Thomas, J.R., & Nelson, J.K. (1996). Research methods in physical activity (3rd edition). Champaign, IL: Human Kinetics.