An investigation into effects and implications of anxiety Essay

INTRODUCTION

According to Lane et al (1999) anxiety is one of the most frequently investigated aspects of sports psychology and this is not surprising since anxiety has been proven to have direct influences upon performance, and so controlling anxiety can enhance performance, definitely an advantage in the world of competitive sport. All athletes are susceptible to the influence of anxiety but this project focuses upon soccer players in particular.

Anxiety can be classified into two categories, trait anxiety, which is general anxiety and state anxiety, which is situational. People with high levels of trait anxiety are anxious in a lot of situations in life, not just under specific circumstances, while state-anxiety is related to an individual or specific situation. Lane et al (1999) also suggest that anxiety is multi-dimensional in its structure, containing both cognitive and somatic components. They explain that cognitive anxiety can occur through media such as negative thoughts, negative self images and self doubt while somatic anxiety can be represented through tense muscles, increased heart rate and sweaty/clammy skin.

Anxiety affects almost all performers, but can affect different performers differently. Work by many investigative sports psychologists including Langer and Imber (1979), Masters (1992) and Hardy et al (1996) suggests that anxiety leads to decrements in performance due to attention focus being diverted, in relation to cognitive effects, and due to tense, rigid muscles in relation to somatic effects.

A substantial understanding of anxiety is imperative for sports psychologists, whose role it is to provide suitable and well-timed procedures in order to mentally prepare athletes for the next performance (Hanton et al 2002), especially in the world of professional soccer. The findings from this research project will aid the understanding sports psychologists have of competitive anxiety and how it affects different athletes. Hopefully they will use the information to make their methods of mentally preparing athletes for competition more effective.

The aims of this research project relate to the comparison of elite and non-elite performers and to do this, two sample populations were used. People who play university football were investigated as an elite population, and those who play college football the non-elite. The first aim of the project was to determine if the two subject populations experience different levels of anxiety before and during a football match, and secondly whether the levels of anxiety related to the level of performance standard. Do those playing high-level sport suffer higher levels of anxiety or lower levels? Another aim was to highlight the effects of anxiety that influence both populations, and to determine if there are any differences. Other objectives include determining if there are any differences in the timing of anxious feelings, and the duration of anxious sensations.

LITERATURE REVIEW

Traditional research into competitive anxiety has conveyed the notion of performance decrements in relation to anxious athletes. Deikman (1966) used the phrase ‘de-automatization’ to describe how performers can ‘choke under pressure’, and later work by Langer and Imber (1979) further explained this idea by saying that athletes who consciously focus their attention on processing and performing a skill disrupt the natural performance of this skill and hence do not perform the skill to the level they normally would. Therefore ‘de-automatization’ led to the natural performance being affected with too much attention focused upon the process of a skill leading to other factors of performance being neglected, for example environmental cues such as the opposition’s position. Masters (1992) later attributed ‘de-automatization’ to competitive state anxiety. He suggested explicit knowledge of a skill in situations of high anxiety leads to ‘de-automatization’ and therefore decreased level of performance.

Many sports psychologists believe that athletes regress a level in terms of Fitt’s and Posner’s (1966) ‘Stages of Learning Model’. This model has three stages, the first stage being the cognitive stage where individuals are learning the skill. They have to concentrate very hard upon the processing and execution of skills and do so with limited success. The second stage is the associative stage which involves a higher success rate as individuals become more fluent in the decision making and execution, and the autonomous stage is the final stage where individuals can automatically execute the skills from stored motor programmes they have developed. The autonomous stage brings much higher success rates.

Athletes do not have to be considered elite to reach the autonomous stage of learning, they simply have to be able to process and execute skills without having to consciously focus. Elite athletes will however, be at the autonomous stage for more skills than non-elite. Anxiety is considered to cause a regress from one stage to the previous and so athletes of all abilities are susceptible to the effects of anxiety. Mullen and Hardy (2000) support this claim by saying that anxious athletes re-invest in knowledge base and control strategies associated with novice performers. Mullen and Hardy (2000) related this notion to Eysenck’s Processing Efficiency Theory (1992) that states anxiety can cause worry, which in turn has two major influences on performance.

Firstly, increased effort affects processing efficiency and secondly, working memory is neglected somewhat which limits the resources usually required to complete the task. Understanding the effects of anxiety can aid the understanding of causes and can aid the identification of anxiety in athletes. Therefore, the above literature from Langer and Imber (1979), Deikman (1966), Masters (1992), Fitts and Posner (1967), Mullen and Hardy (2000) and Eysenck (1992) all has direct implications upon this project. One of the aims of this project is to identify differences in anxiety levels and effects upon elite and non-elite performers and so it is imperative to understand what the effects and symptoms of anxiety actually are. Their research combined with the interviews will allow more relevant questions to be developed for the questionnaires.

More recent research by Jones and Hanton (2001) does however state that not all anxiety is bad. This is supported by Martens et al (1990), who claim that ‘directional perception’ influences whether anxiety is facilitative or debilitative. If athletes interpret anxiety as being useful then it will be and if they see it as a problem, it will become one. Jones and Hanton (2001) believe elite athletes express more facilitative directional perceptions than non-elite and so they can use anxiety to benefit their performance. Jones (1995) contradicts these thoughts though and claims anxiety can never be facilitative and any facilitative symptoms reported are not really anxiety. He believes they are other factors incorrectly labelled as anxiety. This recent research is relevant to this project because it is important to understand there is a case for anxiety being facilitative and debilitative when considering structures for interviews and questions for questionnaires, to ascertain whether anxiety affects elite and non-elite performers differently, it is important that anxiety can be interpreted positively and negatively.

Research by Jones and Hanton (2001) coincides with the first objective of this project, which is to determine any differences in the levels of anxiety elite and non-elite athletes suffer. They state that there is no significant difference in the intensity of symptoms between groups. Therefore our research can be expected to show no difference in the levels of anxiety experienced by both the elite and non-elite populations.

This research will be extremely useful in discussing the results and patterns identified in this project and hopefully this project will help conclude the arguments contained in this literature. For example our research could aid the decision of whether anxiety can be facilitative or whether it always has a negative impact upon performance.

METHOD

The criteria selected for subjects to qualify for the two populations was quite strict in order to minimise the number of variables affecting the results and ensure more reliable results. All subjects were of similar age, between the ages of nineteen and twenty-one. This was to ensure variances related to age differences did not occur, for example it is quite likely that young children will suffer different levels of anxiety compared to older athletes. Only male subjects were used to prevent potential gender variances and finally all subjects played soccer to ensure standardisation of the effects of anxiety. Initially, the aim was to use random subjects for each population but limited interest, limited willingness, and time constraints forced a selective process. Subjects had to be chosen to complete the interviews and questionnaires, to speed up the data collection process. Therefore most of the subjects used were actually people known to the data collectors making the results not quite as reliable because random subject populations are the recommended option.

Two forms of data collection were employed. The first was an interview that was conducted upon eight subjects, four from each target population. The interview served as a qualitative form of data collection and also served as our pilot study. A guideline (see appendix for samples) was devised to direct the interview but subjects were encouraged to talk openly rather than answer questions. Interviews conducted in such an open manner allow themes to arise but, these themes are dictated by the subjects therefore, any repeated issues are likely to be dominant and influential upon the athletes because the athletes feel they are important enough to mention, they have not been prompted in a certain direction of conversation. Transcriptions of all interviews were recorded and common issues were identified and then incorporated into the questionnaires, in order to make the questionnaires more relevant.

Due to anxiety being so commonly investigated, there are existing methods of assessing anxiety, especially in sporting situations. This includes many questionnaires such as the State Competitive Anxiety Test (SCAT) or the Competitive State Anxiety Inventory (CSAI-2). The CSAI-2 is the most commonly used measure (Lane et al 1999) and it was designed by Martens et al (1990) in order to assess competitive state anxiety in athletes, before competition. Although some researchers such as Tabachnick and Fiddell (1996) and Thompson and Daniel (1996) claim the CSAI-2 is invalid because exploratory analyses at each stage of the questionnaire used a ratio of participants to items lower than that recommended (5:1), work by Martens (1990) showed that inter-correlations between the components of anxiety provided sufficient factorial validity.

Due to the fact the CSAI-2 is still the most commonly used measure of anxiety, it was employed as a foundation for the questionnaire used in this project, giving a proven and quantitative form of data collection for our data collection. Some of the questions in the original CSAI-2 were removed and others altered in order to give direction and relevance to this study and also to link some of the issues arising from the interviews. The validity of the questionnaire was maintained because the same four point Likert scale scoring system was used. By modifying the original CSAI-2, the reliability and relevance to this study was increased, making the test more reliable than the original CSAI-2 would have been.

By incorporating two methods of data collection, any correlations in one can be cross-referenced and supported supported by similar patterns in the other. Using both quantitative and qualitative methods of data collection gives more reliable findings and any conclusions can be supported by either specific examples or by statistics.

The procedure for both methods was very simple. The interviews were conducted on individual subjects in isolation from others. Each interview was of approximately ten minutes duration. In total eight interviews were conducted, four from each target population. Transcriptions were made of what the subjects said, and each subject was made aware of the fact that records were being taken, providing opportunity to refuse the interview if they were uncomfortable with the setting. Subjects were asked initial questions to guide them along a relevant course of conversation, with supporting questions used as prompts when necessary. Subjects were ensured confidentiality would be ensured and with no names recorded, so was anonymity.

Questionnaires were simply distributed to the athletes who completed them at a time convenient for them, but with a researcher in proximity should they have any queries. The approximate time taken to complete questionnaires was two minutes. It was important to ensure the questionnaires are not time consuming because of the difficulties encountered in finding random subjects for interviews. Time was one of the major reasons for unwillingness to participate.

RESULTS

The questionnaires were scored on a four-point Likert Scale, with the results being analysed using the statistical analysis software programme SPSS. (Full series of data results shown in appendix) Some questions are asked in a negative context and so to ensure reliable scoring, these questions used a reverse scoring system. So high mean scores for negative questions are actually equivalent to low scores for positive questions.

Fig 1. General statistics and analysis of questionnaires

Group

Mean Score

Std Deviation

T-Test Significance

College

67.18

11.15

0.894

University

66.63

7.51

As fig 1 shows, the overall mean scores from the questionnaire for each population were very similar, which explains why the T-test of t =0.894 showed the results were much higher than the recommended level of significance which is t<0.05. Such insignificant results would suggest that there is no difference in anxiety levels and effects between college and university performers. With no overall significant difference between the results, a more detailed analysis was necessary to ascertain any correlations on a more specific level. A T-test was done for each individual question to establish any patterns occurring for specific factors, and some significant results were noted.

Question 1’s t =0.003 shows significant difference between the two groups and with the college population having the highest mean score for this question it shows that in general college players are less concerned about imminent football matches than the university players.

Question 4 has a significance of t =0.088 which exceeds the recommended t ;0.05 guideline but does show some amount of significance in relation to the majority of results recorded. This question relates to the confidence of the performer and as the mean scores show, university players had the highest levels of self-confidence.

Question 8 shows significance of t =0.034 which is significant according to accepted values. Question 8 relates to the general state of relaxation before competition and again referring to mean scores for the question, university players are more relaxed than college players.

These three questions were the only questions that showed any acceptable significant difference between the two groups, however Question 23 did show extremely low levels of significance, which would suggest that all performers, regardless of performance standard, reported this factor. Question 23 relates to the use of mental imagery to control nerves during competition and it is possible that the highly insignificant difference between the two populations can be attributed to both populations using mental imagery, equally. Suggesting all football players use mental imagery to some extent. Mean scores for this question, for each population are very similar, supporting this claim.

DISCUSSION

As highlighted in the results section, the overall significance values and mean scores for both the college and university populations for the questionnaires showed no difference between the two groups. This suggests that overall both groups experience similar levels of anxiety intensity and present the same symptoms. This coincides with research done by Jones and Hanton (2001) that claimed this very same notion.

Reverting to more in-depth analysis of individual questions that showed a significant difference, results for Question 1 showed that college performers experienced consistently less concern about forth-coming football matches than university players. Further evidence to support this is found in Interview 1 (see appendix) where the subject volunteers the fact that he is concerned about the outcome of his football matches. To directly quote the statement “…..I care about how we play and what the outcome of the game will be.” provides the support that university players were more concerned about the games. The reason for this trend could be found in the fact that university players perceive the importance of their games as very high and they possibly feel under pressure to perform well because they are in the privileged position of playing such high standard football. The additional competitive aspects of higher standard sport could also make university players more anxious of performing and increasing their concern over the outcome of the match. Conversely, college players could see the matches they play as less significant and simply designed for participation and enjoyment rather than success. With college football matches being played in a less competitive manner, the games could be perceived as less important and so concern over the outcome is less. Further proof of this is found in the interviews, as one of the subjects from the college population stated “….it’s just a kick around with a few mates.”.

Question 4 relates to the issue of confidence, and university players reported significantly more confidence than those playing for the college teams. Although the significance value was above the acceptable level of being undoubtedly significant, the value still showed a fairly significant difference and there is a possibility that a study of more subjects would provide a more significant result.

University players will have greater belief in their football ability due to their greater level of skill, hence their self-confidence was reported as much higher. Hanton et al (2002) noted that an athlete’s perception of their own ability predicts their self-confidence and anxiety levels. And so university players have more confidence than college performers. Extending this theory is Martens (1990) idea that an athletes directional perception of anxiety determines whether anxiety is facilitative or debilitative. With a higher level of self-confidence, university athletes will find anxiety more useful than college performers, evidence that supports an idea proposed by Jones and Hanton (2001) that elite athletes report more facilitative interpretations of anxiety. Referring to the Interview 1 (see appendix) the university player stated that he is more confident in his ability playing football than most things in life, and so interprets anxiety before and during a football match as facilitative.

Results from the questionnaires showed that university players were more relaxed before a football match and the reason is probably related to Question 4, and their comparatively higher self-confidence than college players. Although Question 8’s result could be interpreted as a contradiction of results reported in Question 1 where university players said they were more concerned about the game, their self-confidence could enable them to remain relaxed before a game, even though they are under pressure to perform. Increased anxiety intensity can lead to reduced relaxation before a game, but again the fact confidence leads to facilitative interpretations of anxiety, could mean the elite players remain relaxed. Research by Burton (1998) also helps to explain this idea.

He stated that anxiety early in the temporal period (ie well before the game) is interpreted as facilitative and beneficial to performance, and anxiety directly before the game is debilitative to performance. With university athletes remaining more relaxed immediately before a game, anxiety is likely to be more facilitative to them, and beneficial to performance. College athletes however, who are less relaxed before a game, probably experience debilitative anxiety effects upon performance. This could be one of the reasons why they do not play at an elite level.

An issue arising from Question 23 (due to the fact that the T-test result showed extremely low levels of significance), was that no difference from the two groups can be identified and the mean scores for this question were very similar. This would suggest that both populations use mental imagery as much as the other, and with the averages being towards the higher end of the scale (average ;2.5) then both populations use mental imagery to calm nerves during competition. Due to the fact players of all abilities use mental imagery, I would propose that mental imagery is a skill acquired by all footballers in the early stages of learning the game, and possibly it could have other benefits to performance than simply calming nervous feelings. Mental imagery is proven to help the execution of skills and aid positive thoughts and therefore self-confidence.

CONCLUSION

Anxiety intensity levels and symptoms have been proven to be the same no matter what standard of football the subjects played, however the interpretations of anxiety varied between the subject populations. Elite athletes interpreted anxiety in as more facilitative to performance than non-elite athletes and higher self-confidence of elite the athletes is probably the reason for this difference.

The implications of these findings could be employed by sports psychologists, football coaches and captains of football teams alike, in order to enhance the performance of players and therefore the team. These individuals must concentrate on enhancing the self-confidence their players in order to encourage them to interpret anxiety as facilitative to performance rather than debilitative.

Using a combination of methods for data collection, interviews and questionnaires, was a successful technique that provided reliable and valid results. Results from the questionnaires were reflected in the interviews and visa-versa. Although some questions for the questionnaire were introduced due to issues arising in the interviews leading to expectations of some correlation between these two methods, some of the most significant correlations came from questions independent of the interview process. Both these methods were simple to implement and not very time consuming, making them effective and practical measures to use.

One possible limitation of this project could be the fact that out of twenty-three questions, only four provided significant results. This could be due to their being no significant difference between the groups, but it could also be because the sample populations were of a performance standard too similar. University footballers are selected from college teams and so the standard of university football is higher, but the difference between the two levels may not be sufficient to represent differences in anxiety related to performance standard. Repetition of this project should take this into account and use college football in contrast to a standard much higher than university football, such as semi-professional or professional football.

Possible extensions to this project could involve the research into effects of anxiety upon performance, by analysing the performance of footballers in competitive situations under different levels of induced anxiety.

REFERENCES

* Langer and Imber (1979), Deikman (1966), Masters (1992), Fitts and Posner (1967), Eysenck (1992) all cited in Mullen and Hardy (April 2000) State anxiety and motor performance: Testing the conscious processing hypothesis. Journal of Sports Sciences. 18 (pg 785-799). Taylor ; Francis

* Martens et al (1990), Jones (1995) both cited in Jones and Hanton (2001). Pre-competitive feelin states and directional anxiety interpretations. Journal of Sports Sciences. 19 (pg 385-395). Taylor ; Francis

* Hanton et al (2002). Re-examining the competitive anxiety trait-state relationship. Personality and Individual Differences. 33 (7). Elsevier Science

* Martens et al (1990), Tabachnick and Fiddell (1996) both cited in Lane et al (1999). Confirmatory factor analysis of the Competitive State Anxiety Inventory-2. Journal of Sport Sciences. 17 (pg 505-512). Taylor ; Francis

* Burton (1998) cited in Hanton et al (2002). A qualitative investigation of the temporal patterning of the pre-competitive anxiety response. Journal of Sport Sciences. 20 (pg 911-928). Taylor ; Francis

* www.dlibrary.acu.edu.au