EQOL Journal (2020) 12(2):
ORIGINAL ARTICLE
Psychometric properties of a Serbian version of the
Damjan Jakšić1 • Jovana Trbojević Jocić2 ✉ • Stefan Maričić1 • Bülent Okan Miçooğulları3
Received: 9th October, 2020 |
DOI: 10.31382/eqol.201202 |
Accepted: 30th November, 2020 |
|
© The Author(s) 2020. This article is published with open access. |
|
Abstract
Given that anxiety is a significant individual disposition that affects sports performance, this study aimed to verify the latent variables obtained by the State – Trait Anxiety Inventory
✉jovana.trbojevic88@gmail.com
1University of Novi Sad, Faculty of Sport and Physical Education, Novi Sad, Serbia
2Matica Srpska, Novi Sad, Serbia
3Nevşehir Hacı Bektaş Veli University, Physical Education & Sport Department, Nevşehir, Turkey.
urth factor respectively. Since internal reliability of the fourth factor was 0.394 it was excluded from further consideration and interpretation. The achieved scores on four separate factors show that Serbian male and female handball players achieve average results on the first (Presence of Trait anxiety) and the third factor (Absence of proactivity), while they achieve scores above the average on the second factor (Positive affect). In order to examine the sex differences between the examined variables, a
Keywords anxiety • handball players • STAI questionnaire • gender differences.
Introduction
Modern sport has raised the bar of physical achievement and top athletes differ minimally in motor and morphological characteristics. When athletes are equal in all physical dispositions, individual psychological dispositions decide the winner. The psychology of sport has aimed at identifying the individual dispositions of top athletes that affect sports performance and results for the last thirty years (Lavallee et al., 2004). One of the most frequently examined dispositions is emotional regulation, which proved to be important for achieving top results in sport (Lava-
13
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
llee et al., 2004). More specifically, research has largely focused on understanding how athletes' levels of anxiety affect their results (Woodman & Hardy, 2003). Anxiety is an unpleasant emotion accompanied by a vague experience of insecurity, uneasiness and fear without a real stimulus, as well as a high degree of activation of the autonomic nervous system (Petrović & Trbojević, 2020). Martens et al. (1990) assume that anxiety is not a
-it can be momentary (State anxiety) or a personality trait (Trait anxiety) (Spielberger, Gorsuch, & Lushene, 1970). State anxiety refers to a temporary emotional state characterized by subjective feelings of tension that can be of varying intensity. Trait anxiety refers to a relatively consistent tendency to respond to stress with anxiety and a tendency to perceive environmental situations as threatening (Behzadi, Hamzei, Nori, & Salehian, 2011). Trait anxiety can therefore have a far longer and stronger negative impact on the athlete's results, as it has been confirmed with the research in sports psychology (e.g., Rice et al., 2019; Patsiaouras, Papanikolaou, Haritonidis, Nikolaidis, & Keramidas, 2008; Wilson, Wood, & Vine, 2009). Anxiety rates in the general population are between 10 and 12 percent (Somers et al., 2006), while the incidence rate in athletes is estimated to be about 8% (Schaal et al., 2011). Gender differences in the degree of anxiety between male and female athletes were also recorded, in favor of female athletes (Rice et al., 2019), as well as differences in the degree of anxiety in relation to the rank of the competition and the type of sport. Athletes playing in
Before theoretical advances encouraged the development of new measuring instruments for assessing Trait anxiety, anxiety in sport was often assessed using the Sport Competition Anxiety Test (SCAT; Martens, 1977). SCAT is a
14
the above types of anxiety, but primarily assesses somatic anxiety (Smith, Smoll, & Passer, 2002). It has proven to be reliable, but limited in assessing the cognitive aspects of anxiety (Giacobbi & Weinberg, 2000; Johnson, Ekengren, & Andersen, 2005). As a result, a need arose to construct a questionnaire that would assess a particular type of anxiety.
The
Although researchers are increasingly investigating the effects of anxiety on athletic outcomes, research is more focused on linking State anxiety to athletic performance, than on athletes’ Trait anxiety. The researchers primarily focus on pre- competition anxiety, as opposed to Trait anxiety.
Bearing in mind that anxiety represents a significant individual disposition that affects sports performance, this research aims at checking the latent dimensions obtained by the State - Trait Anxiety Inventory
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
sample of Serbian handball players on the State - Trait Anxiety Inventory
Method
The sample consisted of 170 handball players i.e., male (N = 99) and female handball players (N = 71), who have been playing handball for an average of
9.39years; aged 14 to 39 (Mean = 21.9). At the end of 2019, respondents filled out the Trait anxiety STAI questionnaire.
Table 1. Structure of sample of participants
|
Male handball players |
Female handball players |
||
|
|
|
|
|
|
Frequency |
Valid Percent |
Frequency |
Valid Percent |
|
|
|
|
|
Highest rank |
47 |
47.5 |
46 |
64.8 |
First league |
22 |
37.4 |
11 |
15.5 |
Second league |
29 |
14.1 |
14 |
19.7 |
|
|
|
|
|
Total |
99 |
100.0 |
71 |
100.0 |
|
|
|
|
|
The
Exploratory factor analysis with certain modifications of the algorithm was applied to determine accurately latent space. Firstly, the Spearman rank correlation matrix was calculated because of the ordinal nature of the variables. The mentioned matrix was loaded as the initial one for factor analysis instead of raw data. Then, the main components were calculated and presented, and the number of significant factors was determined on the basis of four criteria:
That was followed by determining the differences between male and female handball players in previously defined factors. The results of the items that define the factors formed a summary variable, and for each variable thus obtained, the median, interquartile range and significance of deviations from the normal distribution were calculated using the
female handball players. Furthermore, differences in the obtained aggregate variables formed by the factors, and between the competition ranks, were determined using the
The overall analysis was performed using the R statistical package.
Results
The verification of the latent dimensions obtained by the
15
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
Table 1. Spearman’s intercorrelation matrix
|
1. |
2. |
3. |
4. |
5. |
6. |
7. |
8. |
9. |
10. |
11. |
12. |
13. |
14. |
15. |
16. |
17. |
18. |
19. |
20. |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Item1 |
|
.002 |
.464 |
.423 |
.198 |
.001 |
.116 |
.005 |
.328 |
.000 |
.021 |
.001 |
.000 |
.001 |
.000 |
.000 |
.214 |
.013 |
.000 |
.866 |
Item2 |
|
.258 |
.002 |
.000 |
.000 |
.498 |
.000 |
.000 |
.017 |
.011 |
.001 |
.037 |
.041 |
.008 |
.005 |
.000 |
.000 |
.001 |
.005 |
|
Item3 |
.087 |
|
.008 |
.052 |
.685 |
.403 |
.001 |
.006 |
.063 |
.000 |
.000 |
.001 |
.648 |
.000 |
.009 |
.000 |
.019 |
.002 |
.000 |
|
Item4 |
.239 |
.203 |
|
.000 |
.156 |
.500 |
.001 |
.092 |
.020 |
.077 |
.002 |
.037 |
.389 |
.006 |
.116 |
.001 |
.050 |
.228 |
.060 |
|
Item5 |
.457 |
.149 |
.275 |
|
.000 |
.081 |
.000 |
.000 |
.002 |
.031 |
.000 |
.038 |
.002 |
.005 |
.000 |
.000 |
.000 |
.000 |
.001 |
|
Item6 |
.264 |
|
.002 |
.032 |
.042 |
.000 |
.003 |
.039 |
.004 |
.248 |
.028 |
.002 |
.000 |
.010 |
.000 |
.019 |
||||
Item7 |
.121 |
.241 |
|
.037 |
.243 |
.006 |
.248 |
.000 |
.006 |
.005 |
.006 |
.000 |
.191 |
.020 |
.005 |
.255 |
||||
Item8 |
.323 |
.256 |
.250 |
.449 |
|
.001 |
.003 |
.000 |
.000 |
.000 |
.102 |
.000 |
.001 |
.000 |
.000 |
.000 |
.000 |
|||
Item9 |
.295 |
.210 |
.130 |
.293 |
.255 |
|
.061 |
.000 |
.000 |
.004 |
.029 |
.000 |
.000 |
.000 |
.000 |
.000 |
.000 |
|||
Item10 |
.528 |
.335 |
.212 |
|
.018 |
.000 |
.000 |
.010 |
.000 |
.000 |
.002 |
.000 |
.000 |
.050 |
||||||
Item11 |
.194 |
.348 |
.137 |
.166 |
.324 |
.403 |
|
.000 |
.000 |
.246 |
.000 |
.003 |
.000 |
.000 |
.000 |
.000 |
||||
Item12 |
.263 |
.270 |
.232 |
.401 |
.502 |
.366 |
.432 |
|
.000 |
.022 |
.000 |
.000 |
.000 |
.000 |
.000 |
.000 |
||||
Item13 |
.269 |
.222 |
.211 |
.267 |
|
.017 |
.000 |
.000 |
.005 |
.001 |
.000 |
.009 |
||||||||
Item14 |
.251 |
.089 |
.213 |
.197 |
.090 |
.183 |
|
.116 |
.010 |
.695 |
.000 |
.327 |
.595 |
|||||||
Item15 |
.201 |
.330 |
.210 |
.212 |
.357 |
.319 |
.266 |
.415 |
|
.000 |
.000 |
.000 |
.000 |
.000 |
||||||
Item16 |
.399 |
.241 |
.347 |
.557 |
.363 |
.198 |
|
.001 |
.000 |
.000 |
.000 |
|||||||||
Item17 |
.275 |
.279 |
.257 |
.435 |
.434 |
.469 |
.415 |
.376 |
.373 |
|
.000 |
.000 |
.000 |
|||||||
Item18 |
.269 |
.180 |
.152 |
.363 |
.401 |
.492 |
.351 |
.455 |
.379 |
.483 |
|
.000 |
.000 |
|||||||
Item19 |
.400 |
.331 |
.213 |
.422 |
.426 |
.076 |
.537 |
|
.000 |
|||||||||||
Item20 |
.216 |
.271 |
.145 |
.264 |
.337 |
.357 |
.274 |
.332 |
.302 |
.444 |
.389 |
|
Lower triangle – ρ coefficient; upper triangle – significance
16
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
Figure 1. Determining number of factors
Based on the presented intercorrelation matrix (Table 2), as well as the Figure 1 analysis it was possible to conclude that different criteria for determining a significant number of factors singled out a different number of significant factor solutions. The CG criterion, which takes into consideration all those factors that have eigenvalues values ≥1, generally has a tendency of hyperfactorization, and therefore, as a rule, it should be applied with caution. Especially since the applied factor analysis was performed using ordinal variables, the author highlights that special caution be taken when defining the number of significant main components. The Scree (Parellel) criterion i.e., its mathematical approximation, and the optimal coordinates reduced the number of significant principal components by 1, from 8 to 7. However, the logical sequence that followed did not indicate meaningful solutions. It seems that only by applying the fourth criterion, an optimal solution was achieved, with a minimum
number of single factors, and adequate values of communality and uniqueness.
The values of the first four main components are shown in Table 3, while the values of the assemblies by promax rotation of the main components into a more favorable factor solution, are shown in Table 4.
17
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
Table 3. Principal components (H), eigenvalues (λ) and percentage of common variance explained (R2)
Item |
H1 |
H2 |
H3 |
H4 |
Item1 |
4.272 |
0.336 |
0.757 |
|
Item2 |
1.031 |
|||
Item3 |
1.485 |
|||
Item4 |
||||
Item5 |
0.503 |
|||
Item6 |
4.265 |
0.286 |
0.668 |
|
Item7 |
3.654 |
0.092 |
0.799 |
|
Item8 |
||||
Item9 |
0.092 |
1.268 |
0.543 |
|
Item10 |
5.165 |
0.034 |
0.802 |
0.131 |
Item11 |
1.574 |
0.577 |
||
Item12 |
0.112 |
0.306 |
||
Item13 |
4.965 |
0.192 |
0.461 |
|
Item14 |
3.278 |
2.190 |
1.263 |
|
Item15 |
0.483 |
|||
Item16 |
5.653 |
0.274 |
||
Item17 |
0.229 |
0.139 |
0.772 |
|
Item18 |
1.174 |
0.044 |
||
Item19 |
5.749 |
0.219 |
||
Item20 |
0.605 |
0.733 |
0.453 |
|
|
|
|
|
|
λ |
4.119 |
2.957 |
1.960 |
1.421 |
R2 |
0.206 |
0.148 |
0.098 |
0.071 |
Table 4. Pattern (A) matrix and communalities (h2)
Item |
A1 |
A2 |
A3 |
A4 |
h2 |
Item11 |
0.737 |
0.447 |
0.596 |
||
Item3 |
0.716 |
0.045 |
0.162 |
0.401 |
|
Item20 |
0.688 |
0.167 |
0.109 |
0.032 |
0.440 |
Item17 |
0.663 |
0.108 |
0.347 |
0.123 |
0.600 |
Item9 |
0.644 |
0.199 |
0.162 |
0.470 |
|
Item12 |
0.589 |
0.035 |
0.546 |
||
Item18 |
0.535 |
0.042 |
0.170 |
0.547 |
|
Item8 |
0.516 |
0.272 |
0.461 |
||
Item15 |
0.477 |
||||
Item13 |
0.375 |
0.141 |
0.030 |
0.398 |
|
Item1 |
0.243 |
0.825 |
0.028 |
0.599 |
|
Item10 |
0.054 |
0.798 |
0.047 |
0.648 |
|
Item16 |
0.609 |
0.067 |
0.196 |
0.632 |
|
Item19 |
0.578 |
0.603 |
|||
Item6 |
0.141 |
0.526 |
|||
Item2 |
0.005 |
0.722 |
0.573 |
||
Item5 |
0.156 |
0.011 |
0.673 |
0.640 |
|
Item4 |
0.193 |
0.041 |
0.385 |
0.220 |
|
Item14 |
0.207 |
0.038 |
0.860 |
0.706 |
|
Item7 |
0.284 |
0.130 |
0.431 |
0.342 |
|
|
|
|
|
|
|
Based on the conducted factor analysis of the latent space in the questionnaire, four factors were singled out. Ten items are projected on the first factor, one of which (item 13 I feel secure) is reversed. The items projected on this factor refer to characteristic cognitive and affective aspects of Trait anxiety. A high score of this factor indicates an increased degree of concern about one's own performance and ability, as well as a more pronounced negative affective experience such as sadness and anxiety. Therefore, this factor is named the Presence of Anxiety. The internal reliability of this factor is .760, however when item 13 is eliminated the internal reliability obtained by the Cronbach's Alpha coefficient is .832, therefore item 13 is excluded from further discussion and consideration of this factor.
Five
18
The third factor is formed by three positively directed items which are focused on the experience of fatigue and discouragement due to the assessment of the impossibility of coping with difficulties. Items indicate the absence of activity, so this factor is called the Absence of proactivity. The internal reliability of this factor is .565.
Only two items are projected on the last, fourth factor, with the item 7 directed towards the experience of athletes in which they feel detached when it comes to crisis situations, whereas item 14 is aimed at avoiding crises and difficulties. The internal reliability of this factor is .394. Due to the above, the fourth factor was not taken into further consideration and interpretation.
Table 5. Correlation between factors
|
Factor |
1. |
2. |
3. |
1. |
Presence of anxiety |
|
.000 |
.000 |
2. |
Positive affect |
|
.000 |
|
3. Absence of proactivity |
.466 |
|
||
|
|
|
|
|
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
Correlation analysis show that there is moderate negative correlation between factors Presence of anxiety and Positive affect, as between Positive affect and Absence of proactivity. As expected, there is a positive moderate correlation between factors Presence of anxiety and Absence of proactivity.
Below is a presentation of the scores obtained by Serbian male and female handball players at the questionnaire State - Trait Anxiety Inventory
Table 6. Differences between male and female handball players
|
Male Handball Players |
Female Handball Players |
|
|
||||
|
|
(N=97) |
|
|
(N=71) |
|
|
|
|
|
|
|
|
|
|
|
|
Factor |
Median |
IQR |
KS |
Median |
IQR |
KS |
U |
p |
Presence of anxiety |
16.50 |
.001 |
17.00 |
0.810 |
3307.50 |
.584 |
||
Positive affect |
18.00 |
.000 |
18.00 |
0.000 |
3407.50 |
.818 |
||
Absence of proactivity |
5.00 |
.000 |
5.00 |
0.000 |
3059.00 |
.174 |
||
|
|
|
|
|
|
|
|
|
IQR – Interquartile range. |
|
|
|
|
|
|
|
|
The obtained results show that there is no statistically significant difference between male and female handball players when it comes to the scores obtained at all three analyzed factors.
Figure 2 shows the difference in the scores obtained at the three factor solutions depending on the rank of the competition.
* – significant difference compared with Highest rank.
Figure 2. Differences between competition ranks
A statistically significant difference between male and female handball players of different competitive ranks exists only when it comes to the Positive Affect factor, and between athletes who compete in the highest rank and in the second league (p = .04). Athletes who compete in the second league scored higher on the Positive Affect factor than athletes who compete in the highest rank.
Discussion
An increasing number of researches in the fields of sport and sports psychology is aimed at understanding the individual characteristics that distinguish top athletes from athletes who do not reach that level.
Research so far has largely focused on examining the effects of anxiety on behavioral outcomes of athletes (Lavallee et al., 2004). Anxiety, as an unpleasant emotion, is a complex construct that can be analyzed from the State or Trait perspective. Research has focused more on examining the effects of anxiety as a condition on sports performance in the form of pre- competition anxiety, which is reflected in the cognitive and somatic elements. However, some research results have shown that athletes can experience a high degree of State anxiety as a facilitator i.e., that performance issues and somatic symptoms are motivating and athletes invest more effort to prevent these concerns from materializing (Lavallee et al., 2004). When it comes to Trait anxiety, the results consistently show that Trait
19
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
anxiety has a negative effect on sports performance (Rice et al., 2019). Therefore, this research is aimed at examining Trait anxiety in Serbian male and female handball players of different competitive rank, more precisely at examining one of the most common questionnaires for Trait anxiety assessment STAI (Lonner & Ibrahim, 1989) on a sample of Serbian athletes.
The initial aim of this research was to determine the latent space of the STAI version questionnaire for Trait anxiety in the population of Serbian handball players. Although the STAI questionnaire is valid in most cultures, the validation was done mainly on the student population or the clinical population, and additional data is missing on how this questionnaire describes the sport population, primarily their form of Trait anxiety. When applying and scoring the STAI questionnaire, most researchers take into account the global score achieved on the Trait or State form, very few studies have paid attention to checking the factor structure of separate questionnaire forms (Andrade et al., 2001). However, some studies single out a two- factor solution, a
In our sample, four factors were singled out, three of which were retained and named, according to the items projected on them: Presence of anxiety, Positive affect, and Absence of proactivity. Items aimed at assessing cognitive anxiety - concerns about one's own performance and abilities, as well as items focused on the emotional component - negative affective experience are projected on the first factor - the Presence of anxiety. This factor largely describes anxiety as a personality trait as defined by Spielberg et al. (1980). An item related to positive affectivity (item 13 I feel secure), which was eliminated from the first factor after the analysis of internal reliability, was also projected on this factor. High score on this factor indicates the presence of cognitive anxiety and negative affect. In contrast to the first factor, the second factor, Positive affect - consists of items that focus on positive emotional states, such as security, stability and
20
towards emotional stability and the other towards avoiding difficulties, the fourth factor was not taken into further consideration due to low internal reliability and validity of the factor formation based on two items. The correlations between the factors (Table 5) confirm the stated assumptions regarding the naming and interpretation of the selected factors. Athletes who have a high score on the Presence of anxiety factor will achieve lower scores on Positive affect and a high score on the Absence of proactivity. This means that athletes who have more pronounced anxiety, use passive forms of problem solving to a greater extent, and feel positive emotions to a lesser extent.
Since the
Although previous research has shown that there are significant gender differences in favor of female athletes (Rice et al., 2019) our sample showed no statistically significant differences between male and female handball players in three separate questionnaire factors. On the other hand, a statistically significant difference in the degree of Positive affect was noted between handball players of the highest competition rank and lower rank of the competition.
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
have already achieved everything they wanted. While, athletes who are younger and play in lower leagues have an urge to achieve more and training are still a challenge and a source of satisfaction.
The factor structure of the STAI – Trait form in our sample is similar to the structure of individual studies that singled out a
References
Andrade, L., Gorenstein, C., Tung, T., & Vieira Filho, A. H. (2001). Psychometric Properties of the Portuguese Version of the
Behzadi, F., Hamzei, M., Nori, S., & Salehian, M. H. (2011). The Relationship Between Goal Orientation and Competitive Anxiety in Individual and Team Athletes Fields. Scholars Research Library,
Correia, M., & Rosado, A. (2019). Anxiety in Athletes: Gender and Type of Sport Differences. International Journal of Psychological Research,
Giacobbi, P. J., & Weinberg, R. S. (2000). An Examination of Coping in Sport: Individual Trait Anxiety and Situational Consistency. The Sport Psychologist, 42- 62.
Johnson, U., Ekengren, J., & Andersen, M. B. (2005). Injury Prevention in Sweden: Helping Soccer Players at Risk. Journal of Sport & Exercise Psychology, 32- 38.
Keedwell, P., & Snaith, R. P. (1996). What do Anxiety Scales Measure? Acta Psychiatrica Scandinavica, 177- 180.
Lavallee, D., Kremer, J., Moran, A. P., & Williams, M. (2004). Sport Psychology; Contemporary Themes. Hampshire and New York: Palgrave MacMillan.
Lonner, W. J., & Ibrahim, F. A. (1989). Assessment in
Martens, R. (1977). Sport Competition Anxiety Test. Champaign: IL: Human Kinetics.
Martens, R., Vealey, R. S., & Burton, D. (1990). Competitive Anxiety in Sport. Champaign IL: Human Kinetics.
Pacesova, P., Šmela, P., & Kraček, S. (2019). Personal
Patsiaouras, A., Papanikolaou, Z., Haritonidis, K., Nikolaidis, D., & Keramidas, P. (2008). The Use of the
Petrovic, J., & Trbojevic, J. (2020). Psihologija sporta. Novi Sad: Filozofski fakultet.
Rice, S., Gwyther, K.,
Schaal, K., Tafflet, M., Nassif, H., Thibault, V., Pichard, C., Alcotte, M., . . . Toussaint,
Seok, C. B., Hamid, H. S., Mutang, J. A., & Ismail, R. (2018). Psychometric Properties of the
Students. Sustainability, 10, 3311; doi:10.3390/su10093311.
Smith, R. E., Smoll, F. L., & Passer, M. W. (2002). Sport Performance Anxiety in Children and Youth. U F. L. Smoll, & R. E. Smith, Children and Youth in Sport (str.
Somers, J. M., Goldner, E. M., Waraich, P., & Hsu, L. (2006). Prevalence and Incidence Studies of Anxiety Disorders: A Systematic Review of the Literature. Canadian Journal of Psychiatry,
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the
Spielberger, C. D., Vagg, P. R., Barker, L. R., Donham, G. W., & Westberry, L. G. (1980). The Factor Structure of the
21
This HTML is created from PDF at
EQOL Journal (2020) 12(2):
C. D. Spielberger, Stress and Anxiety. Washington, DC, USA: Hemisphere.
Wilson, M. R., Wood, G., & Vine, S. J. (2009). Anxiety, Attentional Control, and Performance Impairment in Penalty Kicks. Journal of Sport & Exercise Psychology,
Woodman, T., & Hardy, L. (2003). The Relative Impact of Cognitive Anxiety and
How to cite this article:
APA:
MLA:
Chicago:
Jakšić, D., Trbojević Jocić, J., Maričić, S., & Miçooğulları, B. (2020). Psychometric properties of a Serbian version of the
Exercise and Quality of Life, 12(2),
Jakšić, Damjan, et al. „Psychometric properties of a Serbian version of the
Jakšić, Damjan, Jovana Trbojević Jocić, Stefan Maričić, and Bülent Okan Miçooğulları. "Psychometric properties of a Serbian version of the
22
This HTML is created from PDF at