Physical Training July 2003
 
Our Sponsor, SDKsupplies

Motivation Theory for Elite Karate Athletes: 
A Psycho-physiological Approach

Copyright © Shaun Galloway 2003. All rights reserved.
 
 

Shaun Galloway

Semmelweis University
Dept. of Combat Sports



Abstract

The purpose of this study was to distinguish whether or not the use of Borg's rating of perceived exertion RPE (1970b) as a goal setting measure would lead to increased self confidence levels. Each competitor was measured using Vealey's Trait and State Self Confidence scales, TSCI and SSCI (1986). There were also questions used to further understand the dynamic within motivation theory. Elite competitive karate athletes (n=19) completed 3 trials of the TSCI and SSCI. The first trial was given at the beginning of the experiment with the other two trials being given after each of the two training blocks, which were 3 weeks each / 6 training sessions for each block. At the end of the experiment, divisions were made between those who completed their goals, those who underachieved, and those who overachieved. Repeated measures ANOVA 3 x 3 (treatment to goal achievement) showed that there was a significant main effect regarding treatment and group interaction for the TSCI and SSCI (p < .05). Regression analysis supported Bandura's self-efficacy theory for both TSCI and SSCI. Using a stepwise method of multiple regression, the predictors: previous treatment scores and that level's goal achievement placement, all added to the model significantly. The implications of the use of the Borg RPE scale in enhancing self-confidence are complex; however as a psycho-physiological method to control goal setting it is very useful.

Introduction

Motivating athletes has always been a topic of interest for all those involved in competitive sport. The coach, athlete and sport psychologist/scientist all work very hard at motivating the athlete to obtain good results. The problem with this triad is that the coach often works as a coach and sport psychologist. This is not to say that coaches cannot be masters of motivation; however, coaches have little time for new information regarding motivational strategies, let alone preparation for the normal duties inherent to coaching. That said, what I hope to accomplish in this article is to provide a more economical form of motivation testing which is very practical for the coach as well as the athlete. Using Gunnar Borg's Rating of Perceived Exertion scale, the athlete has an elegant way to gauge motivation output. This scale also offers the coach a quick and easy way to use motivational theory within the course of a normal training session. The Borg RPE scale can be taught in a matter of minutes and refreshed on subsequent trainings. The more that the athlete uses the scale the more he will be in tune with his body and his own psycho-physiological level. By using the Borg RPE as a goal-setting tool, the coach and athlete will be able to use many motivational theories, thus bridging the gap between theory and practice. I will briefly outline the various theories needed to understand this research project.

The first is Borg's theory of rating of perceived exertion, which was developed in the late 1950's. The RPE scale had been used mostly for cardiac patients, but it has gained popularity in sport psychology literature. "The Borg RPE scale consists of 15 grades from 6 (very, very light) to 20 (very, very hard). These gradations correspond roughly to 1/10 the heart variation from 60 to 200 in healthy, middle aged individuals. The participant is asked to estimate the degree of exertion he or she feels and select the appropriate number on the scale corresponding to the perception of exertion" (Ostrow, 2002). Exercise intensity (which is often thought of as fatigue and exertion) can be measured in three distinct ways: 1) physically based: power, work and energy, torque; 2) physiologically, such as: VO2 or heart rate (HR); and 3) subjective intensity: concerning the ratings of subjective intensity as perceived by the athlete (Borg 1998). It is this last rating which Borg believes is the most individualized direct measure of exercise intensity. In fact, Borg mentions the connection between ratings of perceived exertion and motivation: "The special situation in which work is performed may emphasize certain aspects of it, suppressing some sensation and making the subject attend to and concentrate on other sensations. Motivation and emotions during exercise may also influence perception and performance." (Borg, 1998, p. 3).

Borg points out a very interesting fact: that many people only look at physiological signs when setting training loads. The mind and the body, however, can not be separated, especially when regarding top athletes. Doing this leads to disastrous results both to motivational outputs as well as physiological outputs (Meyers & Whelan, 1998).

One of the motivational theories that this program addresses is that of achievement motivation as represented by McCelland and Atkinson's theory of achievement (McCelland et al., 1953). To summarize, there are two main points which this theory supports: 1) the motive to achieve success, which is manifest in the athlete as intrinsic motivation, and 2) fear of failure. Intrinsic motivation is the energy found to do an activity for your own sake and not for other reasons. The other point is not focused on in this study.

The McCelland - Atkinson model later developed the theory of extrinsic motivation due to the gap that this model had as a result of very little explanation for outward or socially motivated people. Extrinsic motivation is any reward which is generally connected to ego satisfaction, such as medals, praise from others, and money. This type of athlete needs satisfaction derived from social sources. Where the extrinsically motivated athlete differs from the intrinsically motivated athlete is that even when performance goals are set they will not be satisfied with these goals as there is no social comparison involved in the evaluation process. Where as the intrinsically motivated athlete will base their performance on how they complete the task (Roberts, 2001).

Nicholls' theory of achievement has components of Bandura's self-efficacy theory (Bandura, 1977) and the theory of competence motivation, (Harter 1982). This is developmental theory which supports the concept that children are either task-orientated or ego-orientated depending on the situation. With the use of the Borg scale as a goal setting tool, the positive environment that best supports the task-orientated position is created. This form of controlling the psychological and physiological training levels rids the athlete of the ego environment as the athlete is directed towards self controlled training (task-orientated) as opposed to training which is compared to others (ego-orientation).

Finally the last motivational theory that is used when applying the Borg RPE scale is Bandura's Self-Efficacy theory. It is here that this method of goal setting is best exemplified through theory. Bandura's theory proposes that self-efficacy (which is the belief that a person can achieve a specific task) is increased through: successful performance, verbal persuasion, vicarious experience (modeling) and emotional arousal. Using the Borg RPE a coach can now touch on two of the four principles (successful performance and emotional arousal)

By using the Borg RPE to as a goal setting tool the coach can be employing many of the precepts of motivational theory: intrinsic/extrinsic motivation theory, Nicholls' theory of perceived ability (task and ego orientation) and Bandura's theory of self-efficacy. The context of this study is within the overall study of a PhD thesis and represents one part of this doctorial dissertation. In this research article I will focus on the group which received the "Borg RPE" treatment.

It is my hypothesis that personal cognition of training levels will vary in self-confidence levels due to the ability of those athletes to either, reach, exceed or fail in gaining the goals set for them in the form of the Borg rating of perceived exertion scale and that the items which were given to support self-efficacy theory will be found to be significant.

Methods

Participants

The participants were competitive karate athletes competing using World Karate Federation (WKF) rules. There were n=20 athletes chosen at random from the pool of elite karate athletes within the dojo N=39. One athlete had to drop out of the study, leaving n=19 in the total study. An elite athlete in karate was defined as an athlete who has competed at the inter-provincial level and someone who trains 4 to 5 times a week. The elite karate athlete's main focus is on doing well during a tournament season. The analysis was further broken down into gender groupings.

Male athletes

The men who participated in the study (n=13) were from British Columbia, Canada. They were all members of B.C Karate and Karate Canada, which is part of the World Karate Federation (WKF). The athletes had a mean age = 22.69; SD. 3.7 and their training average in years were M years = 9.61; SD 4.66.

Female athletes

The women who participated in the study (n=6) were from British Columbia, Canada. They all had the same affiliations as the men. The athletes had a mean age = 23.33; SD 4.80 and their training average in years were M years = 9.80; SD 3.72.

Measures

The athletes completed two types of self-report inventories at three different intervals. The first was at the beginning of the experiment, prior to any treatment. The second was at three weeks, which was the mid point of the experiment. They completed the final interval at the conclusion of the experiment, which was three weeks after the mid point. The total length of time to collect the measures was 6 weeks.
A general information questionnaire was handed out at the beginning of the experiment, so as to better build a model of understanding into the complex nature of this experiment.

Finally, each athlete was asked to keep a training log in which they wrote down the Borg RPE that they attained for that training. They kept this log for the duration of the experiment.

State Sport-Confidence Inventory. The State Sport-Confidence Inventory (SSCI; Vealy, 1986) was used to assess the athlete's level of certainty that a person has at one particular situation related to their ability to be successful in sport. The SSCI measures individuals' state confidence using 13 measures with a 9 - point Likert scale (see appendix 1). It is important to note that the individuals are asked to compare their confidence to the "most self-confident athlete they know." The SSCI has an alpha reliability coefficient of .95. Though the original inventory did not evaluate test-retest reliability, this study did address test-retest reliability with elite karate athletes (n=20), which were .97 (3 week interval), and .94 (6 week interval). This inventory is a more direct measure of specific confidence at a certain point in time, for example, the confidence level just prior to a competitive match.

Trait Sport-Confidence Inventory. The Trait Sport-Confidence Inventory (TSCI; Vealy, 1986) was used to assess the athlete's level of certainty a person has about their ability to be successful at sport in general. The TSCI measures individuals' trait confidence using 13 measures with a 9 - point Likert scale (see appendix 2). Again this scale compares the athletes abilities to the, "most confident athlete they know." The TSCI has an alpha reliability of .93. The test-retest reliability coefficients were .86 (1-day interval), .89 (1-week interval) and .83 (1-month interval). The TSCI is a more global and static confidence level.

The SSCI was designed to positively correlate to the TSCI and both are important in the understanding of behaviour. Vealy has reported that subjective outcomes such as causal attribution, perceptions of success and satisfaction have an interaction with the TSCI and the athlete's disposition to competitive orientation (Feltz D.L., and Chase M.A., 1998).

Borg Rating of Perceived Exertion. The Borg rating of perceived exertion was first developed in 1974 for use with cardiac patients (Borg RPE; 1974). The Borg RPE scale consists of 15 grades from 6 (very, very light) to 20 (very, very hard). These gradations correspond roughly to 1/10 the heart variation from 60 to 200 in healthy, middle aged individuals (see appendix 3). Gunnar Borg, himself, saw the connection between his exertion scale and motivation, "The special situation in which work is performed may emphasize certain aspects of it, suppressing some sensation and making the subject attend to and concentrate on other sensations. Motivation and emotions during exercise may also influence perception and performance." (Borg, 1998, p. 3). Ratio scaling reliability produced reliability coefficients above .90 (Borg, 1962a). Test-retest coefficients of .80 to .98 were reported over a 2 to 4 week period among 21 male and 15 female participants. Validity was assessed through correlation coefficients of .65 and .68 which were reported with heart rate. Test-retest reliability coefficients of .80 to .98 were reported over a 2 to 4 week time period (Ostrow, A., 2002).

Items designed to test motivation theory. These items were developed to understand the motivational dynamics of this experiment. It was hypothesized that there would be a wide range of results in the SSCI and TSCI scores due to each individual's make up. For example, there would be those who reached their goal, those that did not reach the goal and those who exceeded their goal. These items were used to better understand the goal achievement model

The first two questions were chosen to test whether or not karate athletes gained self-confidence through age (as a causation of maturity) and years training (as a causation of repeated training exposure). The next two questions were to examine the difference between perceived success and actual success. Tournament frequency and location were gauged so as to complete the model of an elite karate athlete. Finally the difference between the total sum of the Borg RPE levels that were set by the instructor and the total of the athlete's actual Borg RPE levels was examined as the sum total of the goal achievement abilities of each karate athlete. All questions with the exception of the 'perceived success' question were objective measures (see appendix 4).

Procedure

This study was a field study and as such is considered a quasi-experimental design. The first process in starting the study was finding three or more karate clubs which would have enough elite athletes from which a random sample could be chosen. I was able to find three clubs which were interested in being involved in the research and which had enough athletes to conduct the study. This research took place for 6 weeks in July and August, which is usually a "down time" for most karate clubs. I chose this time because those athletes who usually train during this time are dedicated.
The clubs all had instructors who trained in the same "mother club" and all three clubs were going through the specialized training phase. I then presented the idea of the experiment to each group and gave out consent forms to those athletes who were randomly chosen. There was no one who declined to be involved in the research. However, one athlete dropped out of the Borg group. The reason for leaving the study "was to go on an unexpected vacation."

The Borg karate group had a total class size 39 elite karate athletes. The men in study were n=13 and the women in study were n=6. The total in the study were N=19 and the total group mean age was = 22.89 / SD 3.72, while the total group mean years training were = 9.80 / SD 3.72.

I taught the Borg group instructor the Borg rating of perceived exertion scale. He then practiced giving me the instructions as set out by Borg, 1997. There has been much concern over the fact that the poor instructions lead to poor prediction and results. I then showed him the training logs in which the athletes were to place the date, Borg level to be met (which was set by myself and given by the instructor) and the Borg level that was attained for that training. I was only at the club for the three times that I delivered and collected the inventories. The inventories were delivered and collected by one researcher. I made it explicitly clear to the instructor to remind and give the athletes time to put a Borg entry into their training logs after training. To check on whether or not the athletes had been involved in the training sessions, I compared the attendance records with their Borg log book for discrepancies. There were none to note.

The training levels were set according to training theory (Bompa, 1999). In traditional training theory there are four phases which are: general preparation phase, sport specific training phase, competitive training phase and regeneration training phase. We set the 6 week experiment at the sport specific preparation phase which is signified by high volume / high intensity sport specific actions. This is considered the hardest part a periodized training program.

The training levels were specifically set high to see if intense training and the cognition of reaching, exceeding or not reaching the goal had an effect on the athlete's trait and state confidence. Through this manipulation we were able to put to the test the concept of, "no pain no gain."

Results

Data Screening

The Kolmogorov-Smirnov tests for SSCI treatments were not significant at p < .200. The TSCI tests of normality were: p < .200, p < .174 and p < .200. From these results I could continue on with parametric statistics.

Looking at the descriptive statistics in table 1 it is clear that those athletes who were classified as 'reaching their goals' had increases in their mean averages. This is in contrast to those that did not reach their goals or those who exceeded their goals.

Table 1. Means and Standard Deviations for SSCI and TSCI treatments and goal achievement
 
SSCI Treatments Groupings M SD   TSCI Treatments Groupings M SD
     
1st SSCI treatment Reached goals 6.39 .65 1st TSCI treatment Reached goals 6.42 .68
Over reached goals 6.67 .68 Over reached goals 6.71 .65
Did not reach goals 6.19 .35 Did not reach goals 6.47 .40
     
2nd SSCI treatment Reached goals 6.52 .69 2nd TSCI treatment Reached goals 6.51 .67
Over reached goals 6.53 .78 Over reached goals 6.56 .73
Did not reach goals 6.15 .37 Did not reach goals 6.34 .41
     
3rd SSCI treatment Reached goals 6.63 .78 3rd TSCI treatment Reached goals 6.65 .74
Over reached goals 6.39 .97 Over reached goals 6.34 .95
Did not reach goals 6.07 .47 Did not reach goals 6.34 .42

Repeated measures ANOVA analysis

Table 2 Repeated Measures ANOVA 3X3 (SSCI Treatments x Goal Achievement)

 
 Source df SS MS F
Between Subjects
Treatment 1.04 .026 .025 .32
Treatment x Goal Achievment 2.08 .47 .226 2.9 (p<.08)
Error 16.65 1.29 7.79
Within subjects
Treatment
level 1 vs. 2 1 .004 .004 .16
level 2 vs. 3 1 .02 .02 .43
level 1 vs. 3 1 .05 .05 .57
Treatment x Goal achievement
level 1 vs. 2 2 .24 .12 4.21*
level 2 vs. 3 2 .22 .11 1.98
level 1 vs. 3 2 .94 .15 2.98 (p<.07)
Error
level 1 vs. 2 16 .47 .02
level 2 vs. 3 16 .90 .05
level 1 vs. 3 16 2.51 .15
*p < .05.

Fig 1

Figure 1 - SSCI mean averages for repeated measures

All main effects for treatments were found not significant F(2, 54) = .329, p>.72, ?2 = .02.

The effects for goal achievement to treatments had mixed significance levels. Mauchly's test indicated that the assumption of sphericity had been violated, ?2 (2) = 38.24, p<.001, therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (e = .52). The results show that the main effect for treatment (training intensity) and the Borg results (the goal achievement) was not significant, F(2.08, 16.66) = 2.90, p< .08, ?2 = .26. indicating that there was a small effect on state sport confidence related to the ability to realize goal outcomes.

Although, at the on-set of this research, I left my alpha level at the traditional .05, Stevens (2002), suggested that the alpha level should be a subjective level set by each individual researcher. In this situation I felt it warranted to continuing the search further. Taking Stevens' advice it might have been more proper to set an alpha level of .10 due to the small sample size and the fact that the results will not have serious consequences by accepting them at a higher alpha level.

Due to the border line significant result for treatment to Borg effect I looked at the level comparisons. There was no significance between level 2 and level 3 F(2, 54) = 1.28, p>.17, ?2 = .19. However, the results showed that level 1 to level 2 interaction was significant, F(2, 54) = 4.21, p<.05, ?2 = .34, indicating that the training load between level 1 and 2, taken together with the goal achievement knowledge, resulted in a significant increase in state sport confidence. The results also showed that level 1 and level 3 interactions were border line significant, F(2, 54) = 2.90, p<.08, ?2 = .27.

It is very easy to see from figure 1 that those who reached their goals increased their confidence levels after each training block. This is in contrast to those who did not reach their goals and those who exceeded their goals.

Table 3 Repeated Measures ANOVA 3X3 (TSCI Treatments x Goal Achievement)
 
 Source df SS MS F
Between Subjects
Treatment 1.15 .08 .04 1.45
Treatment x Goal Achievment 2.3 .61 .26 5.55**
Error 18.4 .89 .04
Within subjects
Treatment
level 1 vs. 2 1 .07 .07 3.04
level 2 vs. 3 1 .01 .01 .30
level 1 vs. 3 1 .15 .15 1.48
Treatment x Goal achievement 
level 1 vs. 2 2 .23 .11 4.50*
level 2 vs. 3 2 .42 .21 5.51**
level 1 vs. 3 2 1.20 .60 5.83
Error
level 1 vs. 2 16 .41 .02
level 2 vs. 3 16 .61 .03
level 1 vs. 3 16 1.64 .10
*p < .05. **p < .01

Borg TSCI Repeated measures ANOVA

Using Repeated measures ANOVA (3 X 3) goal achievement to trait sport confidence treatment. I measured the interaction between treatments and goal achievement / treatments.

All effects for treatments were found not significant F(2, 54) = 1.45, p>.24, ?2 = .08.

When examining the treatment to goal achievement interaction, Mauchly's test indicated that the assumption of sphericity had been violated, ?2 (2) = 19.76, p<.001, therefore degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity (? = .58). The results show that the main effect for treatment (training intensity) and the Borg results (the goal achievement) was significant F(2.31, 18.47) = 5.55, p = .01, ?2 = .41.

The follow up results found that the interaction between level 1 and level 2 was significant, F(2, 54) = 4.50, p<.05, ?2 = .30. Significance was also found between levels 2 and 3 was F( 2, 54) = 5.51, p<.05, ?2 = .40. Level 1 and 3 was significant at, F(2, 54) = .5.83, p>.01.

Regression Analysis

Diagnostics Regression Analysis for both SSCI and TSCI models

Next I conducted multiple regression analysis with the supplement items as predictors as can be seen in tables 4 and 6. The scores from the second and third sport state and trait confidence inventory were used as outcome variables. As a result I completed two models for the SSCI and two models for the TSCI.

The main specification assumptions for the regression models were met. The relationships for the variables were linear, all relevant predictors were included and no irrelevant predictors were included in the motivation model.

Table 4 SSCI Borg correlation matrix: model #1
 
Variable M SD 1 2 3 4 5 6 7 8
2nd treatment 6.41 .63 .94*** -.24 .34 .29 .16 .32 -.25 .007
Predictors
1) 1st treatment 6.42 .58 -- -.13 .14 .11 -.008 .13 -.33 -.03
2) borg vs. borg 1.94 .84 -- -74*** -.74*** -.71*** .13 -.002 -.26
3) percieived success 64.8 18.2 -- .85*** .87*** -.70*** -.05 .13
4) actual success 3.57 3.48 -- .92*** .73*** -.03 .25
5) number of tournaments 7.89 2.66 -- .68** .05 .25
6) competition in or out of dojo 1.47 .51 -- -.08 -.01
7) age 22.8 3.72 -- .74***
8) years training 9.84 4.83 --
**p < .01. ***p < .001.

Table 5 Stepwise Regression Analysis for Motivation variables explaining Elite Karate Athletes State Self-Confidence Inventory (SSCI): Model one (N=19)
 
Variable B SEB ? R2 ?R2
1) 1st treatment 1.05*** .04 .98*** .90 .89
2) 1st goal achievement -.33** .05 -.27*** .97 .96
**p < .01. ***p < .001.

Table 6 SSCI Borg correlation matrix: model #2
 
Variable M SD 1 2 3 4 5 6 7 8
3rd treatment 6.38 .76 -.18 .03 .51* .44* .34 .48 .95*** -.49*
Predictors
1) age 6.42 .58 -- .74*** -.05 -.03 .05 -.08 -.25 -.08
2) years training 1.94 .84 -- .13 .25 .25 -.01 .007 -.01
3) perceived success 64.8 18.2 -- .85 .87*** .82*** .34 -.47*
4) actual success 3.57 3.48 -- .92*** .73*** .29 -.34*
5) number of tournaments 7.89 2.66 -- .68** .16 -.39*
6) competition in or out of dojo 1.47 .51 -- .32 -.42*
7) 2nd treatment 22.8 3.72 -- -.30
8) 2nd goal achievement 9.84 4.83 --
*p < .05. **p < .01. ***p < .001.

Table 7 Stepwise Regression Analysis for Motivation variables explaining Elite Karate Athletes State Self-Confidence Inventory (SSCI): Model two (N=19)
 
Variable B SEB ? R2 ?R2
1) 2nd treatment 1.03 .05 .85 .90 .91
2) 1st treatment -2.19 .06 -.17 .95 .04
3) 2nd goal achievement .006 .002 .14 .95 .01
**p < .01. ***p < .001.

Regression Analysis for SSCI models

Multicolinearity is a usual problem in regression analysis, so I will briefly report the basic results with this problem in mind. The Durban-Watson test was 2.87, which was higher than the suggested 2.0; however it wasn't higher than the 3.0 which suggests problems in multicolinearity. The tolerance levels were all above the .02. (Menard, 1995) and the VIF levels did not approach the 10 point danger level (Myers, 1990).

As a final validation test of the SSCI regression models, I used Stein's adjusted R2 which was .96 for model one and .89 for model two. The first model had very little variation between R2 and Stein's adjusted R2. However, there is more variance between Stein's adjusted R2 and the second model, indicating that the future predictive power of the model might not be very strong.

Table 8 TSCI Borg correlation matrix: model #1
 
Variable M SD 1 2 3 4 5 6 7 8
2nd treatment 6.47 .59 -.18 .04 .27 .28 .15 .30 .94*** -.04
Predictors
1) age 22.8 3.7 -- .74*** -.05 -.03 .05 -.08 -.32 -.11
2) years training 9.84 4.83 -- .13 .25 .25 -.01 -.03 -.08
3) perceived success 64.8 18.2 -- .85*** .87*** .82*** .03 -.64**
4) actual success 3.57 3.48 -- .92*** .73*** .10 -.59**
5) number of tournaments 7.89 2.66 -- .68** -.05 -.57**
6) competition in or out of dojo 1.47 .51 -- .09 .68**
7) 1st treatment 6.53 .58 -- .19
8) 1st goal achievement 2.47 .51 --
**p < .01. ***p < .001.

Table 9 Stepwise Regression Analysis for Motivation variables explaining Elite Karate Athletes Trait Self-Confidence Inventory (TSCI): Model one (N=19)
 
Variable B SEB ? R2 ?R2
1) 1st treatment 1.04 .04 1.02 .90*** .90
2) perceived success .008 .001 .26 .95*** .05
3) age .004 .01 .27 .97*** .018
4. years training -.02 .007 -.16 .98** .01
**p < .01. ***p< .001.

Table 10 TSCI Borg correlation matrix: model #2
 
Variable M SD 1 2 3 4 5 6 7 8
3rd treatment 6.47 .59 -.08 .12 .41 .41 .29 .43 .94*** -.44
Predictors
1) age 22.8 3.7 -- .74*** -.05 -.03 .05 -.08 -.18 -.08
2) years training 9.84 4.83 -- .13 .25 .25 -.01 .04 -.01
3) perceived success 64.8 18.2 -- .85*** .87*** .82*** .27 -.47
4) actual success 3.57 3.48 -- .92*** .73*** .28 -.34*
5) number of tournaments 7.89 2.66 -- .68** .15 -.39
6) competition in or out of dojo 1.47 .51 -- .30 -.42*
7) 2nd treatment 6.47 .597 -- .27
8) 2nd goal achievement 2.36 .579 --
**p < .01. ***p < .001.

Table 11 Stepwise Regression Analysis for Motivation variables explaining Elite Karate Athletes Trait Self-Confidence Inventory (TSCI): Model two (N=19)
 
Variable B SEB ? R2 ?R2
1) 2nd treatment 1.07 .075 .89 .90*** .90
2) 2nd goal achievement -.25 .075 -.21 .93** .04
**p < .01. ***p < .001.

Regression Analysis for TSCI models

The diagnostics for multicolinearity was similar to the SSCI analysis. The Durban-Watson 2.31, tolerance levels +.02 and VIF levels >10 were all at acceptable levels.

Steins adjusted R2 which was .96 for model one and .85 for model two. The first model had very little variation between R2 and Stein's adjusted R2. The second models variance differences between Stein's adjusted R2 and the R2 statistics indicating a poor match for the predictive power of the model.

Discussion

Hypothesis

Both of my hypotheses were supported: 1) That personal cognition of training levels will vary in self-confidence levels due to the ability of those athletes to either, reach, exceed or fail in gaining the goals set for them in the form of the Borg rating of perceived exertion scale and 2) That the items which were given to support self-efficacy theory will be found significant.

The athlete's self-confidence levels varied depending on their predisposition towards goal setting. Nicholls' theory of achievement motivation best explains this hypothesis. Those athletes who are task motivated (the goal achievers) were able to increase their confidence levels in the midst of two very hard training cycles. While, those with either weak (underachievers) or strong (overachievers) ego motivated behaviour had a decrease in confidence. In this experiment I can classify underachievers as ego orientated due to the fact that the goals which were set were self-reference goals. The Borg RPE is a self-perceived rating of exertion. As a consequence those who continually rate their Borg RPE levels below the levels that they are to achieve, are doing so because they are using others in the training group as a reference and are not able to compete, thus their ego' are involved. In contrast, those who could outperform (overachievers) the other athletes will want to consistently show others to they can dominant (Duda & Whitehead, 1998).

As figure 1 has shown, not only did the ego motivated overachievers decrease their confidence, they did so drastically. This could be due in part to the already large training load which was prescribed for the experiment and if those overachievers tried to out train others when the training level was already set at a ceiling, they experienced fatigue due to overtraining. This would further explain the decline in the state sport confidence levels as this inventory was administered prior to most of a training session and to kumite matches, while the trait sport confidence inventory was administered in a fairly rested state. I would call for future researchers to fully research the dynamic between the Borg RPE and overtraining so as to make a better informed decision regarding causality.

The second hypothesis was partly supported due to the fact that there were only 2 or 3 predictors which explained the self-confidence levels and not all 8 predictors. It might be suggested that the regression model should only include predictors from one theory as opposed to many theories. For example, including just questions related to positive enactment (positive results), physiological and affective States (autonomous recognition), the power of verbal persuasion and vicarious experience (modeling) which relate to Bandura's self-efficacy model. However these regression models did yield very interesting information. These results suggest that an athlete's initial disposition towards goal setting is very important. If the athlete is a goal achiever she/he will be a goal achiever throughout the process and they will have a positive result as a consequence. In this case, this program was very helpful for those athletes who achieve their goals and are task orientated. From the results I can suggest that athletes who are ego orientated will not benefit from such a program. This research design did not address the question of motivational environment but this theory could be an important concept to include in future research. I would argue that the nature of this design could lead to a task orientated environment but not at the intense training levels that were prescribed. A lower level training program would allow those athletes who are ego orientated not to achieve an overtraining state in such a quick period of time which then acts as a snowball effect producing negative results. Future experiments should use mid range Borg RPE levels and use brief group counseling sessions to help everybody understand that training should be wholly individual. The ability to reach your own goal which is not measured against others would be the main message to get across to the athlete.

Conclusion

In conclusion, I believe that the Borg RPE holds a great deal of promise as a goal achievement process. It is important for the practicing instructor to realize that athletes will come to training with a predisposition towards goals and to be able to consider these traits before starting a goal setting program which includes training levels. Another consideration is that the proper motivational environment is very important to stress when using goal setting which includes training levels. Through continued research in this area, researchers and instructors will be able to find a practical way to use motivation theory within the flow of a normal training session.

References

Bandura, A. (1997) Self-Efficacy: The exercise of control, New York: W.H. Freeman and Co.
Bompa. T.O.,(1999) Periodization: Theory and Methodology of Training, Champaign, IL: Human Kinetics.
Bompa. T.O.,(1999) Periodization Training for Sports, Champaign, IL: Human Kinetics.
Borg, G. (1998). Borg's perceived exertion and pain scales. Champaign, IL: Human Kinetics.

Cox. R. (1994) Sport Psychology: Concepts and Applications. Champaign, IL: Brown and Benchmark.

Duda, J.L., and Whitehead., (1998) Measurement of goal perspectives in the physical domain. In Duda J.L. (Ed.) Advances in Sport and Exercise Psychology Measurement. (pp 21-48), Morgan town, WV: Fitness Information Technology, Inc.

Feltz D.L., and Chase M.A., (1998), Measurement of self-efficacy and confidence in sport. In Duda J.L. (Ed.) Advances in Sport and Exercise Psychology Measurement. (pp 65-80), Morgan town, WV: Fitness Information Technology, Inc.

Meyers, A.W. & Whelan, J.P. (1998). A Systematic Model for Understanding Psychosocial Influences in Overtraining. Kreider, R.B. et al. (Eds.), Overtraining in Sport (pp 335-369). Champaign, IL: Human Kinetics.

Ostrow, A., (2002). Directory of Psychological Tests in the Sport and Exercise Sciences: Second Edition. Morgantown, WV: Fitness Information Technology Inc.

Roberts, G.C., (2001) Advances in Motivation in Sport and Exercise, Champaign, IL: Human Kinetics.

Stevens, J.P., (2002) Applied Multivariate Statistics for the Social Sciences, Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc.

Vealy, R.S (1986) Conceptualization of sport confidence and competitive orientation: Preliminary investigation and instrument development. Journal of Sport Psychology, 8, 221-246.

Verkhohansky, Y. (1998), Main Features of a Modern scientific sports training theory, New Studies in Athletics, 13:3, 9-20.

Verkhohansky, Y. (1998), Organization of the training process, New Studies in Athletics, 13:3, 21-31.

Authors Notes:

Various parts of this article will be presented at the FEPSAC conference in Copenhagen, Denmark on July 22, 2003.

If any martial arts instructors would be interested in getting involved in research please contact me at this email address: shaun_galloway@yahoo.com

I have various research projects focusing on psychological processes as well as training theory.


Appendix 1: State Sport Confidence Inventory - SSCI (Sample questions)
State Sport-Confidence Inventory


Name of Dojo:_____________________________ Athlete Number:_________

*Think about how confident you feel right now about performing successfully in the upcoming competition (karate: tournament or dojo competition).

*Answer the questions below based on how confident you feel right now about competing in the upcoming competition.

*Compare your self-confidence to the most self-confident athlete you know.

HOW CONFIDENT ARE YOU RIGHT NOW ABOUT COMPETING IN THE UPCOMING CONTEST? (circle a number)
 
1) Compare the confidence you feel right now in YOUR ABILITY TO EXECUTE THE SKILLS NECESSARY TO BE A SUCCESSFUL to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9
2) Compare the confidence you feel right now in YOUR ABILITY TO MAKE CRITICAL DECISIONS DURING COMPETITION to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9
3) Compare the confidence you feel right now in YOUR ABILITY TO PERFORM UNDER PRESSURE to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9
4) Compare the confidence you feel right now in YOUR ABILITY TO EXECUTE SUCCESSFUL STRATEGY to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9


Appendix 2: Trait Sport Confidence Inventory -TSCI (Sample questions)


Name of Dojo:_____________________________ Athlete Number:_________

*Think about how self-confident you are when you compete in sport (karate: tournament or dojo)

*Answer the questions below based on how confident you generally feel when you compete in your sport. Compare your self-confidence to the most self-confident athlete you know.

*Please answer as you really feel, not how you would like to feel. Your answers will be kept completely confidential.

WHEN YOU COMPETE, HOW CONFIDENT DO YOU GENERALLY FEEL? (circle a number)
 
1) Compare your confidence in YOUR ABILITY TO EXECUTE THE SKILLS NECESSARY TO BE A SUCCESSFUL to the most confident athlete you know. Low Medium High 
1 2 3 4 5 6 7 8 9
2) Compare your confidence in YOUR ABILITY TO MAKE CRITICAL DECISIONS DURING COMPETITION to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9
3) Compare your confidence in YOUR ABILITY TO PERFORM UNDER
PRESSURE to the most confident athlete you know. 
Low Medium High 
1 2 3 4 5 6 7 8 9
4) Compare your confidence in YOUR ABILITY TO EXECUTE SUCCESSFUL STRATEGY to the most confident athlete you know.  Low Medium High 
1 2 3 4 5 6 7 8 9


Appendix 3: Borg RPE scale and instructions as was collected from the athletes.


Borg Rating of Perceived Exertion (RPE) Scale Log Book

Borg instructions

While exercising I want you to rate your perception of exertion, i.e., how heavy and strenuous the exercise feels to you. The perception of exertion depends mainly on the strain and fatigue in your muscles and on your feeling of breathlessness or aches in the chest.

Look at this rating scale; I want you to use this scale from 6 to 20, where 6 means "no exertion at all" and 20 means "maximal exertion."
 
9 Corresponds to "very light" exercise. For a normal, healthy person it is like walking at his or her own pace for some minutes. 
13 On the scale is "somewhat hard" exercise, but it still feels OK to continue.
17 "very hard" is very strenuous. A healthy person can still go on, but he or she really has to push him or herself. It feels very heavy, and the person is very tired.
19 On the scale is an extremely strenuous exercise level. For most people this is the most strenuous exercise they have ever experienced.

Try to appraise your feeling of exertion as honestly as possible, without thinking about what the actual physical load is. Don't underestimate it, but don't overestimate it either. It's your own feeling of effort and exertion that's important, not how it compares to other people. What other people think is not important either. Look at the scale and the expression and then give a number.

Write down the training date.

Write down the training length.

Then circle the RPE number that best represents your exertion level for that training.

Example…
Training date:
Length of training:
 
6 No exertion at all
7 Extremely Light
8
9 Very light
10
11 Light
12
13 Somewhat hard
14
15 Hard (heavy)
16
17 Very hard
18
19 Extremely hard
20 Maximal exertion


Appendix 4: General information questionnaire


This questionnaire is to further help understand motivation factors in your training. Please answer as truthfully as possible. The only person who will be reading your responses will be the researcher: your confidentiality will be kept.

Name: _______________________ Birth date / Age: __________________________

1/ Rate this past tournament season on a percentile. 0% the worst season in your competitive career and 100% the best season you have had to date. ____________

2/ List the tournaments you went to this year, whether or not they were in your province or outside the province and the results to the 4th place finish. *Only include kimute results.

- Example:

1) BC provincials, 3rd place in weight division and 2nd place in team kimute.

2) Las Vegas Open, 1st place in open kimute


*Leave this space alone. It is for the input of your confidence scores and your Borg training log.
 
1st SSCI: 2nd SSCI: 3rd SSCI:
1st TSCI: 2nd TSCI: 3rd TSCI:
/ 1st Borg: 2nd Borg:

 
 
 


Our Sponsor, SDKsupplies

Physical Training July 2003