Stanford,2006;K?l?çarslan,2006).It is because that the physical self-perception contains the individual’s perception and evaluation of his/her motor skills (coordination, sports skill etc.) and physical compatibility parameters (power, strength, flexibility, etc.) (Özerkan, 2004; K?l?çarslan, 2006). Therefore, it is expected that the favorable physical self-perception in the students of physical education may influence the self-esteem positively. There are studies in the literature showing the favorable effect of making sports on development of self-esteem (Karakoç, 2012; Korkmaz, 2002).

This study was conducted to reveal self-esteem level of futsal players who were students at the department of physical education and sports, and accepted as having necessary qualifications for self-esteem, physical self-perception. 

Research sample

The universe of the research was composed of the futsal players who attented Interuniversity Futsal Group matches in Antalya between 23rd and 27th of February in 2015. A total of 215 futsal players among whom 119 are females and 96 are males from 12 universities took part in the study.Sampling is composed of a total of 215 futsal players whose average age was21.57± 2.20. Their consent was recieved orally and they attended the study voluntarily. Prior to the application of the scale, research samplewas informed about the study and the scale.

Most of the participants belong to 18-21 years of age group with 55.8% followed by 22-25 years of age group with 37.7% and 26-30 years of age group with %6.5%. The percentages of participants with perceived income levels of very low, low, medium, high and very high were 4.2, 18.1, 64.7, 11.6 and %1.4 respectively.

Data collection ?nstrument

The study data were collected by using the Student Information Form, Coopersmith Self-Esteem Inventory (CSEI).

The Student Information Form: It includes sociodemographical characteristics and information related to school choices of the students.Thisformwasdeveloped by researcher through searching the literature about investigation field and receiving opinions of the experts charging in the field.

Coopersmith Self-Esteem Inventory (CSRI): The inventory developed by Coopersmith (1986) has a long form composed of 58 questions and a short form composed of 25 questions. The short form was used in the present study. Validity and reliability studies were performed on the short form after being adapted to Turkish by Turan and Tufan (1987). The consistency coefficient of alpha was found as 0.62 in the study performed. The score of the inventory ranges from 0 to 100. The level of the self-esteem according to the score was as the following: 10 to 30 points is low, 31 to 70 points is moderate and 71- 100 points is high (Tufan 1987).

Statistical analysis

The data obtained in the study was evaluated by the use of SPSS 19.0 statistical software .The Cronbach’s alpha internal consistency was found to be .70 which confirms the reliability of the test. This result shows that the scale is reliable. The investigation of thedata revealed that they were not showing a normal distribution. That was why non parametric test was employed. The self-esteem of the futsal players according to sex was investigated by the use of   Man-Whitney U test while the dependence of self-esteem according to age and perceived income levels were evaluated by Kruskall-Wallis analysis. 

Table 1 shows the percentage and frequency distribution of self-esteem levels of the players according to age and perceived income levels. When we look at the self-esteem levels of the research sample it is seen that the 93% of the futsal players have medium and high levels of self-esteem. In accordance with the aim of the study, firstly, normality test is conducted for the purpose of determining through which method the test will be conducted. To that end, Kolmogroy-Simirnoy normality test was conducted.

H₀: Data were gathered from normal distribution.

H₁: Data were not gathered from normal distribution.

H₀ hypotheses was rejected because p-value was smaller than α-value (0,05) when Kolmogrov-Simirnov test wasanalysed. That is to say, data were not collected from normal distribution. Thus, non-parametric techniques were used for analysis.

Table 2 shows the self-esteem levels of the participants according to sex, age and perceived income levels.

Table 3lists the self-esteem levels of the participants according to sex. According to Mann-Whitney U test results there was a statistically significant difference between the self-esteem values of the female and the male players [Z(215)= -3.86; p<0.05]. The female participants were observed to have higher self-esteem levels.

Table 4shows the comparison of the self-esteem levels of the participants according to the age groups. According to the data obtained by the use of Kruskal-Wallis Analysis there were statistically significant differences  between 18-21 , 22-25 and 26-30 years age groups [ (215)= 12.92; p<0.05]. These results indicate that the self-esteem levels decrease as the age of participants’ advances.

Based on Table 5the Kruskal-Wallis analysis results reveal that there was a statistically significant correlation between the perceived income levels and the self-esteem levels of the participants[ (215)= 13.47; p<0.05]. The participants with a medium income level were found to have the highest self-esteem level.

Table 1 shows the percentage and frequency distribution of self-esteem levels of the players according to age and perceived income levels. When we look at the self-esteem levels of the research sample it is seen that the 93% of the futsal players have medium and high levels of self-esteem. In accordance with the aim of the study, firstly, normality test is conducted for the purpose of determining through which method the test will be conducted. To that end, Kolmogroy-Simirnoy normality test was conducted.

H₀: Data were gathered from normal distribution.

H₁: Data were not gathered from normal distribution.

H₀ hypotheses was rejected because p-value was smaller than α-value (0,05) when Kolmogrov-Simirnov test wasanalysed. That is to say, data were not collected from normal distribution. Thus, non-parametric techniques were used for analysis.

Table 2 shows the self-esteem levels of the participants according to sex, age and perceived income levels.

Table 3lists the self-esteem levels of the participants according to sex. According to Mann-Whitney U test results there was a statistically significant difference between the self-esteem values of the female and the male players [Z(215)= -3.86; p<0.05]. The female participants were observed to have higher self-esteem levels.

Table 4shows the comparison of the self-esteem levels of the participants according to the age groups. According to the data obtained by the use of Kruskal-Wallis Analysis there were statistically significant differences  between 18-21 , 22-25 and 26-30 years age groups [ (215)= 12.92; p<0.05]. These results indicate that the self-esteem levels decrease as the age of participants’ advances.

Based on Table 5the Kruskal-Wallis analysis results reveal that there was a statistically significant correlation between the perceived income levels and the self-esteem levels of the participants[ (215)= 13.47; p<0.05]. The participants with a medium income level were found to have the highest self-esteem level.

Although our study aims at analyzing the self-esteem vis-à-vis the variables of sex, age, and perceived income, in the literature, the dependence of the  self-esteem  levels on other variables has also been investigated by some studies done in this area. Thus, there is a need to present thefindings of these studies before coming to the analysis of our results. For instance, according to some, the location of the people has a statistically significant effect upon the self-esteem values (Cengil,  2009;  Alt?ok et al., 2010). Also I??klar (2012) claims that there is a positive and statistically significant relation between the one of thesub dimension of the life quality and the self-esteem levels. It was reported that the increase in the life quality causes a corresponding increase in self-esteem levels in adults (I??klar, 2012).In another study, the physic-education methods based on cognitive behavioral approach were observed to improve the self-esteem levels of the people in the experimental group (?ahan and Birsen, 2012). ?ahin and Gizir (2014) investigated self-esteem levels of the 859 students (450 females and 409 males) studying in Mersin University and the self-esteem values were found to describe the shyness level of the total and female participants to a statically significant extent. There was a negative correlation observed between urgency and the self-esteem levels ofthe obesity patients (Sar?soy et al., 2013). The smokers fasting during the holly Ramadan were found to have low self-esteem levels. The students were found to have medium to high self-esteem levels (Çal?yurt et al.,2005). The people addicted to narcotic compounds were also found to have low self-esteem values (Toker et al., 2011).

The main goal of this study is to investigate the self-esteem levels of the futsal players according to certain variables that are sex, age, and perceived income. The study we conducted was participated by 119 females and 96 males;total of 215 futsal players with an average age of 21.57± 2.20 years.

Results of our study show that ninety three percent of the participating futsal players were observed to possess medium and high level self-esteem. When compared with other studies on this subject it is observedthat different studies have different standing points about the role of self-esteem. Er?an et al. (2009) reported that the people who are doing team sports possess higher self-esteem level than the others and an adverse mental state develops in the people as the self-esteem levels are decreased. In another study, the self-esteem value of the general people was found to be medium (SalaRaz? et al., 2009). On the other hand,Tiryaki and Moral? (1992) do not support the opinions that the self-esteem levels are positively or negatively affected by the participation of the sportive competitions.

Yet another studyclaimed that the approach of the problem in a negative manner, self-esteem, narcism and smart problem solving are significantly related to the hostility (Temel, 2008).

In anotherstudy,theself-esteemlevels of the university students were found to be low and their seclusion levels were observed to be high. These students often need psychological assistance. Generally, studies show that people with high level of self-esteem have much better adaptation to new situations and easily establish communication with others. The lower the self-esteem levels ofthe students the higher their seclusion levels (Karahan et al., 2004).As such, the results of our study on futsal players and those who state that sports participation positively effects self-esteem levels are compatible.

The Mann-Whitney U test, which we used to determine the degree of the change in the self-esteem levels of the participants according to sex,revealed that there was a statistically significant difference between the female and male futsal players in favor of the females. Similarly, a previousstudycarried out by Alt?oket al.(2010) upon 1335 students in Adnan Menderes University revealed that the self-esteem levels of the female students were found to be higher than the male students. Correspondingly,K?mter (2012) suggested that the self-esteem levels of the female students were higher thanthe male students. On the other hand, there are also numerous studies in the literature reporting that there were no statistically significant differences observed between the self-esteem levels of the females and the males  (Cengil, 2009; I??klar, 2012;  Uçan and Ça?layan, 2012; Özdemir, 2014). Our study results are compatible with the studies claiming that self-esteem levels of females is found to be higher than those of males. The reason behind our finding about female futsal players’ higher self-esteem levels than that of males is thought to be that the intensiveness of the pace and the game speed in male futsal competitions is much greater than that in female futsal competitions.

According to Kruskal-Wallis Variance analysis results which we used to test the change of the self-esteem levels according to the age groups, there are statistically significant differences between 18-21, 22-25 and 26-30 years of age groups. The results show that as the ages of the participants increase there is a corresponding decrease in their self-esteem levels. However, contrary to our finding, in a study byAlt?oket. al. (2010) which was carried out on 1335 students,it is claimed that as the age increases, the self-esteem levels show an increase. Similarly with the study of Alt?oket. al. (2010),Er?an et al.(2009) observed that the self-esteem level increases as the age of the students advances. On the other hand, as to the study of Özdemir (2014), there were no statistically significant differences observed between the self-esteem levels of the university students in Turkey and the US according to the age variable. As such, the part played by the age variable remains as a still controversial issue among the studies done on the very area, and thus needs a further investigation.

The Kruskal-Wallis analysis also revealed statistically significant changes between the self-esteem levels and perceived income of the participants. According to the results of our analysis, the futsal players with a medium perceived income were observed to have higher self-esteem levels.

The studies in the literature reported similar results. Er?an et al. (2009) reported that there was a linear relation between income and the self-esteem levels. In another study, the self-esteem levels of the students with no financial problems were found to be higher (Alt?ok et al., 2010). Again K?mter (2012) claimed that the increase in the income levels of the people had a synergistic effect upon their self-esteem levels in a study he carried out upon the adults.   However, there are also studies stating that the income levels do not cause any statistically significant differences upon the self-esteem values (Cengil, 2012; Gacar and Yanl?ç, 2012;Özba? and Gürhan, 2013). The reason behind our findings on thepart played by the perceived income variable is thought to be the fact that the university teams participating in futsal competitions in Turkeyare the representatives of different parts of the country (eastern and western provinces) which possess great socio-economic andcultural differences that effect perceived incomes of the participants.

The study revealed that 93% of the participants have medium or high levels of self-esteem. The results of  Mann-Whitney U test employed to test the self-esteem levels according to sex  revealed a statistically significant differences between the male and female participants [Z(213)= -3.86; p<0.05]. The Kruskal-Wallis variance analysis results showed that self-esteem levels change with a statistically significant extent with the age groups [ (215)= 12.92; p<0.05]  and the perceived income level [ (215)= 13.47; p<0.05].

The females and the people with a medium perceived level of income were found to exhibit higher self-esteem levels. It was also found that the increase of the age is associated with a corresponding decrease in the self-esteem values.

• It is important that the factors affecting the self-esteem values of the futsal players should be determined and appropriate steps should be taken to increase it.

• The education and training programs should be developed according to the results obtained from  similar studies.

• The self-esteem levels of individuals doing different sports should be investigated and compared with each other.

The author have not declared any conflict of interests.