Female Athletes’ Life Satisfaction: The Contribution of Individual and Collective Self-Esteem
XiaoXueyan, Halijah bt. Ibrahim
Faculty of Education Science & Technology, Universiti Teknologi , Skudai, Johor, Malaysia
This study investigates the factors influencing life satisfaction among female athletes, with a particular focus on the roles of individual and collective self-esteem. Utilizing a questionnaire survey and statistical analyses of 139 female athletes, the research identifies individual self-esteem as the primary predictor of general life satisfaction. Conversely, collective self-esteem emerges as a significant predictor of satisfaction in training and competition, accounting for 67% of the variance. Dominance analyses were employed to elucidate the relative importance of each variable. The findings underscore the critical role of collective self-esteem in specific life domains, particularly within collectivist cultural contexts. While the study provides valuable insights, it is based on cross-sectional data, necessitating further research to establish causality and explore the applicability of these findings to other demographic groups.
Keywords: female athletes; individual self-esteem; collective self-esteem; general life satisfaction; specific life satisfaction
Life satisfaction is a cognitive evaluation of the quality of an individual’s life experiences.[1][2]In today’s society, life satisfaction has become an important indicator of an individual’s psychological well-being and social adaptability. It not only reflects an individual’s subjective perception of his or her life situation, but is also closely related to psychological health, well-being, and social behaviour. For the special group of athletes, the connotation of life satisfaction is richer and more complex. Life satisfaction can be divided into two kinds, general life satisfaction is the overall evaluation of personal life quality; special life satisfaction is the specific evaluation of different life areas, such as marital satisfaction, income satisfaction, job satisfaction, etc. General life satisfaction is more abstract and stable than special life satisfaction.[3]Personal self-esteem, on the other hand, is an individual’s intrinsic evaluation of his or her values and abilities, which reflects the degree of self-acceptance and self-affirmation in self-perception. For athletes, personal self-esteem is often closely linked to their achievements, skill levels, and self-efficacy in sports. An increase in personal self-esteem not only enhances athletes’ satisfaction with their sporting career but may also extend to their positive evaluation of other aspects of their lives, which in turn enhances their general life satisfaction. Collective self-esteem, on the other hand, is the self-esteem experience that results from an individual’s sense of identification with and belonging to the group to which he or she belongs. Athletes are usually affiliated with a sports team, club, or national team, and these groups provide them with a supportive, cooperative, and competitive environment. The formation of collective self-esteem is closely related to team cohesion, team achievement, and group culture. When athletes are part of a team that is victorious or honoured, they feel proud to be part of the team, and this sense of collective honour translates into collective self-esteem, which in turn affects their overall satisfaction with life. Additionally, collective self-esteem can alleviate some of the frustrations that athletes may face on an individual level, as the support and sense of belonging within a team can provide them with a psychological buffer. However, athletes’ life satisfaction is not solely determined by individual and collective self-esteem; it is also influenced by a variety of other factors such as social support, family environment, career development, physical health status, and socio-cultural background. For example, an athlete may have achieved great success in sports and possess high levels of individual and collective self-esteem, but if they lack support in their family life or encounter difficulties during career transitions, their life satisfaction may still be negatively affected (Smith et al., 2020).Therefore, a comprehensive understanding of the components of athletes’ life satisfaction and their interaction mechanisms is of great significance in promoting athletes’ psychological health and career development.
The present study aims to fill this research gap, providing a theoretical basis and practical guidance for athletes’ mental health intervention and career development by deeply exploring the influence mechanism of individual self-esteem and collective self-esteem on athletes’ sense of life satisfaction, and revealing the characteristics of their roles in different contexts. Through in-depth analysis of athletes’ life satisfaction, we can not only better understand the psychological needs of athletes, but also provide useful insights for the promotion of mental health in the sports sector and society as a whole.
Current status of research related to life satisfaction and self-esteem
Life satisfaction is an important area of research in “positive psychology” [4], and one of the main interests of the field is to find the best predictors of life satisfaction. Through long-term efforts, psychologists have found that self-esteem is one of the best predictors of life satisfaction among demographic variables (e.g., gender, age, marital status, family income, education, socioeconomic status, etc.), self-esteem, social support, personality traits, coping skills, and adaptability. [1] For example, Campbell, Converse, and Rodgers (1976) [5]found that self-esteem had the highest correlation with life satisfaction among all predictor variables related to life satisfaction.Diener and Diener (1995) [6]conducted a cross-cultural study involving 13,118 university students from 31 countries and found a correlation of 0.47 between self-esteem and life satisfaction. This finding further supports the importance of self-esteem in predicting life satisfaction.
In this type of research, however, the term self-esteem refers to personal self-esteem. This kind of self-esteem emphasises the personal aspect of the self, that is, the sense of worth, respect and goodness of the individual, which is derived from the characteristics of the individual, the individual’s abilities and the difference between “I” and “he/she”. The importance and emphasis on personal self-esteem is one of the characteristics of the individualistic culture. [7] The self nurtured by the collectivist culture, on the other hand, has a stronger collective and social character. [The self-esteem associated with this kind of self emphasises the sense of collective value, respect and goodness. These self-perceptions and self-evaluations derive from the characteristics of the collective, its capabilities, and the distinction between “us” and “them”.
Human self-esteem means feeling and evaluating the value of the individual self, while collective self-esteem means feeling and evaluating the value of the social group.[9] The difference between these two types of self-esteem comes from the distinction between the social self and the individual self. The difference between these two types of self-esteem stems from the distinction between the social self and the individual self. As early as the 1980s, unique European social psychologists such as Taifel and Turner (1981) proposed the distinction between social identity and individual identity in social identity theory [10]and social category theory[11];later,Triandis[12]proposed a three-dimensional theory of the self, which classifies the self into the individual self, the public self, and the collective self; Markus and Kirkland(1997)[13] proposed a three-dimensional theory of the self, which classifies the self into the individual self, the public self, and the collective self; Markus and Kirkland [14] proposed a three-dimensional theory of the self, which classifies the self into the individual self, the public self, and the collective self; and Markus and Kirkland [15] proposed the three-dimensional theory of the self. The self-construal theory of Markus and Kitayama [7] divides the human self into social and individual selves; Tajfel and Turner et al. raise the problem of the structure of the self at the group level, while Triandis and Markus and Kitayama raise the problem of the structure of the self at the culture level. The structure of the self is presented at the level of culture by Triandis Markus and Kitayama. Despite these differences, the three theories are united by the distinction between social and personal selves, which are considered to be important influences on cognition, motivation, and emotion, and can be used to predict human behaviour.
The notion that collective self-esteem is derived from the social self has only recently attracted the attention of psychologists. [9,13-19] A study of Americans by Crocker et al. [16] suggests that collective self-esteem may be an important predictor of life satisfaction. They found that even after controlling for the effects of individual self-esteem, some dimensions of collective self-esteem were reliably biased, albeit with low coefficients of bias, about life satisfaction.
In a cross-cultural study of 53 countries and regions, Hofstede [20] found that the United States was at the top of the individualism dimension (score of 1. The smaller the number, the stronger the individualistic tendency) and was considered to be a representative of an individualistic culture, while Taiwan, China, was at the lower end of the individualism dimension (score of 43. The bigger the number, the stronger the collectivistic tendency. The maximum value is 53), which is considered to be a representative of collectivist culture. Since then, many researchers [21-25] have tended to believe that American culture favours individualism and Chinese culture favours collectivism based on the results of empirical studies. Thus, the findings of Crocker et al. suggest that if collective self-esteem has a value-added contribution (i.e., a new contribution beyond individual self-esteem) to the prediction of life satisfaction in individualistic cultures (e.g., the United States), it is reasonable to hypothesise that this value-added contribution should be even more pronounced in collective cultures (e.g., the Chinese).
Crocker [16] and others, in examining the role of collective self-esteem on life satisfaction, controlled only for the effect of individual self-esteem. This control, however, still seems to be insufficient. We argue that since both social support and collective self-esteem are related to social interactions, and both have a predictive function on life satisfaction. Therefore, when examining the contribution of collective self-esteem to life satisfaction, the effect of social support should also be controlled for. If collective self-esteem can still make a new contribution to the prediction of life satisfaction after controlling for the two important variables of individual self-esteem and social support, we will have a deeper understanding of the nature and function of collective self-esteem.
Based on the above discussion, it is hypothesised that collective self-esteem can make a new contribution to the prediction of life satisfaction even after controlling for three types of variables: demographic variables, personal self-esteem and social support. In the present study, athletes were invited as subjects to test the above hypothesis. This is because athletes live in an environment with a strong collectivist atmosphere. Their food, clothing, housing, and transport are mostly in the sports team; their training and competition have clear collective goals; and their success greatly depends on the quality of their social interactions with teammates, coaches, and team leaders. Athletes are an ideal group to test the relationship between collective self-esteem and life satisfaction.
Literature research method
This study employs the literature review method to systematically examine relevant domestic and international literature on life satisfaction, individual self-esteem, and collective self-esteem. By reviewing existing research, we have clarified the theoretical connotations and measurement methods of each concept and summarized the main achievements and shortcomings of previous studies. The literature review not only provides a solid foundation for constructing the theoretical framework of this study but also helps identify gaps in the current research field, offering guidance for the innovative direction of this study.
This study adopts the questionnaire survey method to collect quantitative data to support subsequent statistical analyses. The participants include 139 female athletes from a sports team and a sports university, aged between 12 and 19 (mean age: 19.64 years), with training experience ranging from 2 to 17 years (mean training duration: 8 years). The questionnaire is designed based on athletes’ life satisfaction[26], individual self-esteem, and collective self-esteem[27], covering basic information and relevant evaluation indicators. To ensure sample diversity, the questionnaire was distributed among athletes from different sports, competitive levels, and age groups.
Rationale for Research Design
This study selects the questionnaire survey method primarily because it efficiently collects large amounts of quantitative data, making it suitable for exploratory research. Through the questionnaire, we can systematically measure athletes’ life satisfaction, individual self-esteem, and collective self-esteem, providing reliable data support for subsequent statistical analyses. Additionally, the flexibility and broad applicability of the questionnaire method make it the best choice for this study.
Sources of Questionnaire Design
The questions in the questionnaire were designed with reference to established and mature scales, particularly those measuring life satisfaction and self-esteem. Specifically, the measurement of life satisfaction is based on the Life Satisfaction Scale developed by Diener et al., while individual self-esteem and collective self-esteem are measured using the Rosenberg Self-Esteem Scale and the Collective Self-Esteem Scale by Luhtanen and Crocker, respectively. These scales have high reliability and validity in the fields of psychology and sociology, effectively measuring the core variables of this study.
Through these methods, this study ensures the scientific rigor and reliability of the data, providing a solid foundation for subsequent analyses and conclusions.
Mathematical and statistical methods
The questionnaire data were analyzed using advanced mathematical and statistical techniques to ensure robust and reliable findings. Descriptive statistics were employed to summarize the overall characteristics of the variables, providing a foundational understanding of the data distribution. Correlation analysis was conducted to identify potential associations between variables, offering insights into their interrelationships. Furthermore, regression analysis was utilized to examine the predictive roles of individual self-esteem and collective self-esteem on life satisfaction, allowing for a nuanced understanding of their impacts [19].The application of mathematical statistics was essential for several reasons. First, it enabled the distillation of complex data into interpretable information, facilitating hypothesis testing and the derivation of accurate conclusions. Second, the use of these methods ensured the validity and reliability of the findings, which is crucial for drawing meaningful inferences in social science research.
Logical analyses
The logical analysis method is used throughout the study to ensure that the research ideas are clear and the arguments are reasonable. The questions and hypotheses are clarified in the research design stage, the results are rationally interpreted in the data analysis stage, and the report is written to ensure that the content is coherent and the conclusions are scientific. Logical analysis presents a clear framework for the study, making it easy to understand and learn from.
The mean, standard deviation and zero-order correlation of the main variables are given in Table 1. As shown in Table 1, the mean of athletes’ general life satisfaction was 4.52 on a 7-point Likert scale, indicating that their general life satisfaction was moderate to high. The mean of training and competition satisfaction was 4.15, which was lower than general life satisfaction, t= 3.38, p< 0.01.
Table 1 also shows correlations of 0.39 for general life satisfaction and training match satisfaction, 0.37 for individual self-esteem and collective self-esteem, and 0.29 for similarity to significant others and concordance with significant others, which suggest that (1) these three pairs of significant variables have different psychological significance, and (2) that the likelihood of multiple covariations in subsequent multiple regression analyses was The correlations indicate that (1) these three pairs of important variables have different psychological significance; (2) in the subsequent multiple regression analysis, the possibility of multicollinearity is small.
Table 1 Mean, standard deviation and zero-order correlation of main variables
Note: N= 125-139, p<0.05 for a correlation of 0.19 and above, p<0.01 for a correlation of 0.23 and above.
| variant | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Life satisfaction | — | |||||||||
| Training and competition satisfaction | 0.39 | — | ||||||||
| personal self-esteem | 0.41 | 0.25 | — | |||||||
| collective self-esteem | 0.21 | 0.50 | 0.37 | — | ||||||
| Coaching Support | 0.07 | 0.27 | 0.12 | 0.13 | — | |||||
| Team Support | 0.20 | 0.25 | 0.99 | 0.19 | 0.14 | — | ||||
| Friends Support | 0.19 | 0.08 | 0.06 | -0.07 | 0.04 | 0.41 | — | |||
| Family support | 0.19 | -0.03 | 0.31 | 0.03 | 0.05 | 0.13 | 0.19 | — | ||
| Similarity to significant others | 0.29 | 0.24 | -0.01 | 0.04 | 0.01 | 0.11 | -0.05 | 0.16 | — | |
| Harmony with significant others | 0.23 | 0.04 | 0.11 | -0.01 | 0.09 | -0.15 | -0.03 | 0.21 | 0.29 | — |
| mean value | 4.53 | 4.15 | 5.15 | 5.13 | 2.02 | 2.38 | 2.74 | 2.72 | 4.80 | 5.96 |
| (statistics) standard deviation | 1.14 | 1.23 | 0.91 | 0.94 | 0.53 | 0.54 | 0.62 | 0.73 | 1.07 | 0.81 |
Value-added contribution of group self-esteem
The present study used stratified regression to examine whether collective self-esteem contributes value-added to the prediction of athletes’ general life satisfaction and training and competition satisfaction after controlling for three commonly used predictors, namely demographic variables, personal self-esteem and social support. To do this, the variables were stratified into regression equations to calculate
The change in R2 produced between the two strata and the F-test value of this change were examined to see if there was a reliable improvement in R2 after inputting a new class of predictors into the regression equation.[29]The procedure was as follows: in the first step, demographic variables were first entered, including gender, age, sport, sport level and years of training; in the second step, individual self-esteem was entered; in the third step, social support was entered; and finally, in the fourth step, collective self-esteem was entered.
The results of the regression analyses for predicting general life satisfaction are given in Table 2. The results show that, in the first step, demographic variables failed to predict general life satisfaction reliably; in the second step, individual self-esteem reliably explained 15 per cent of the total variance in general life satisfaction; in the third step, the variables of social support reliably explained another 14 per cent of the total variance in general life satisfaction; and, in the fourth step, collective self-esteem failed to make a new contribution to the prediction of general life satisfaction based on demographic variables, individual self-esteem and social support. Make a new contribution to the prediction of general life satisfaction. This result did not support the hypothesis proposed in this study.
The results of the regression analyses for predicting training match satisfaction are presented in Table 3. The results show that, in the first step, demographic variables failed to predict training match satisfaction reliably; in the second step, individual self-esteem reliably explained only 5 per cent of the total variance in training match satisfaction; in the third step, the social support variables reliably explained another 18 per cent of the total variance in training match satisfaction; and, in the fourth step, collective self-esteem succeeded in making a new contribution to the prediction of general life satisfaction, i.e., reliably explaining another 13 per cent of the total variance in training match satisfaction, on top of the demographic variables, individual self-esteem and social support. Made a new contribution to the prediction of general life satisfaction by reliably explaining another 13% of the total variance in training match satisfaction. This result supports the hypothesis proposed in the present study.
Table 2 Stratified regressions predicting general life satisfaction
Note:**p<0. 01
| Input Variables | R | R2 | △R2 | △F |
| Level 1: Demographic variables | 0.164 | 0.027 | 0.027 | 0.611 |
| distinguishing between the sexes | ||||
| (a person’s) age | ||||
| sports programme | ||||
| campaign level | ||||
| training period | ||||
| Level 2: Personal self-esteem | 0.425 | 0.181 | 0.154 | 20.484** |
| Tier 3: Social support | 0.565 | 0.320 | 0.139 | 3.498** |
| Coaching Support | ||||
| Team Support
|
||||
| Friends Support | ||||
| Family support | ||||
| Similarity to significant others | ||||
| Harmony with significant others | ||||
| Tier 4: Collective self-esteem | 0.569 | 0.324 | 0.004 | 0.650 |
Table 3 Hierarchical regressions predicting training match satisfaction
Note:**p<0. 01
| Input Variables | R | R2 | △R2 | △F |
| Level 1: Demographic variables | 0.271 | 0.074 | 0.074 | 1.746 |
| distinguishing between the sexes | ||||
| (a person’s) age | ||||
| sports programme | ||||
| campaign level | ||||
| training period | ||||
| Level 2: Personal self-esteem | 0.344 | 0.119 | 0.045 | 5.586** |
| Tier 3: Social support | 0.547 | 0.229 | 0.181 | 4.423** |
| Coaching Support | ||||
| Team Support
|
||||
| Friends Support | ||||
| Family support | ||||
| Similarity to significant others | ||||
| Harmony with significant others | ||||
| Tier 4: Collective self-esteem | 0.653 | 0.426 | 0.127 | 22.546** |
Relative weight of collective self-esteem
To determine more clearly the relative importance of collective self-esteem in predicting athletes’ general life satisfaction and training and competition satisfaction, a new statistical method, namely dominance analysis, was used in the present study for the calculation. This is because the traditional method of determining the relative importance of each predictor in multiple regression equations is characterised by model dependency and is unable to keep the relative importance indicators constant across the different sub-models derived from the multiple regression model, which is overcome by the dominance analysis recently proposed by Budescu [30]. Dominance analysis decomposes the contribution of each predictor to the total variance of the dependent variable into the percentage of predicted variance, thus allowing the relative importance of each predictor to be constant across different sub-models derived from the regression model.
At the same time, the percentage of predicted variance of each predictor produced by the dominance analysis is also model-independent and is not affected by different combinations of different predictors in the multiple regression model.
At present, there is no computer programme to perform the advantage analysis automatically, so some parameters need to be calculated manually, and the complexity of the calculation mainly depends on the number of predictors. To reduce the difficulty of manual calculation, the present study first used the combination of hierarchical regression and stepwise regression to select the best indicators for predicting athletes’ general life satisfaction and training and competition life satisfaction. The results showed that personal self-esteem, similarity to significant others, and support from friends were the best predictors of athletes’ general life satisfaction, which explained 27% of the total variance; and collective self-esteem, similarity to significant others, and support from coaches were the best predictors of athletes’ training and competition satisfaction, which explained 31% of the total variance. These two sets of predictors were used in subsequent strengths analyses.
To analyse the relative importance of the different predictors, it is first necessary to regress general life satisfaction on different predictors (expressed as X1, X2, etc.) of personal self-esteem X (X1), similarity to significant others (X2) and support from friends (X3), as well as on different combinations of these three predictors. In total, seven different regression equations can be generated with three predictors: three regression equations with one predictor (X1, X2, X3); three regression models with two predictors (X1X2, X1X3, X2X3); and one regression equation with all three predictors (X1X2X3). Please note that “X1X2” does not mean X1 times X2. “X1X2” means that the regression equation with two predictors includes predictor 1 and predictor 2, and the same nomenclature rule applies to all regression equations described in this paper.
Table 4 Relative Contributions of Three Indicators of Personal Self-Esteem, Similarity to Significant Others, and Friend Support in Predicting General Life Satisfaction
| variant | R2 | X1 | X2 | X3 |
| – | 0 | 0.167 | 0.085 | 0.035 |
| X1 (personal self-esteem) | 0.167 | – | 0.086 | 0.026 |
| X2 (similarity to significant others) | 0.085 | 0.168 | – | 0.040 |
| X3 (friend support) | 0.035 | 0.158 | 0.090 | – |
| X1X2 | 0.253 | – | – | 0.031 |
| X1X3 | 0.193 | – | 0.091 | – |
| X2X3 | 0.125 | 0.159 | – | – |
| X1X2X3 | 0.284 | – | – | – |
| Decomposition of R2 | 0.163 | 0.088 | 0.033 | |
| Percentage of predicted variance | 57.39 | 30.99 | 11.62 |
Strengths analyses require that the squared multiple correlations of each indicator be compared across all regression equations containing different combinations of predictors. Table 4 presents the results of the dominance analysis for the regression equations predicting general life satisfaction. The first (leftmost) column of Table 4 represents the seven regression
variables included in the equation; the second column represents R2 for that regression equation. The third, fourth and fifth columns represent the increase in R2 when the predictor is added to the regression equation. For example, the second column of the third row of Table 4 shows that if personal self-esteem alone is used to predict general life satisfaction, R2 is 0.167; the fourth column of the third row shows that if similarity to significant others is added to the regression equation with only one predictor, personal self-esteem, R2 increases by 0.086; and the fifth column of the third row shows that if support from friends is added to the regression equation with only one predictor, personal self-esteem, R2 increases by 0.086. The symbol “-” in Table 4 indicates the first or the first block of predictors in the regression equation (the first block in SPSS). Thus, in each row of Table 4, we can see the increase or change in the value of R2 when another predictor (the second block in SPSS) is added. The fifth, sixth and seventh rows of Table 4 show the new contribution of the indicator if another indicator is added to the regression equation with two predictors (X1X2, X1X3 and X2X3), and the row where X1X2X3 is located shows that the R(2) of the full regression model with all three predictors is 0.284.
The key step in the strengths analysis is that by averaging the contribution (R2) of each predictor in a regression equation containing the same number of predictors and then averaging these averages for each predictor, the R2 of the full regression model can be disaggregated into components that reflect the relative importance of each predictor.[30-31] The penultimate row of Table 4 shows that the relative contributions of the three predictors of individual self-esteem, similarity to significant others, and support from friends can be recalibrated to 0.163, 0.088, and 0.033, respectively. The three predictors used for predicting athletes’ satisfaction with training competitions were collective self-esteem, similarity to significant others, and coach support. The same dominance analysis procedure was used to decompose the total variance of athletes’ training game satisfaction. The results are shown in Table 5. The penultimate row of Table 5 shows that the relative contributions of the three predictors of collective self-esteem, similarity to significant others, and coaching support can be recalibrated to 0.230, 0.055, and 0.059, respectively.
Adding the three values (0.163,0.088,0.033) in the penultimate row of Table 4 equals the R2 (0.284) of the full regression model for predicting satisfaction with life in general; adding the three values (0.230,0.055,0.059) in the penultimate row of Table 5 equals the R2 (0.343) of the full regression model for predicting satisfaction with the training competition. We thus see that the average contribution of each predictor from the dominance analysis is more refined and more stable than that from the traditional regression method.
The last rows of Tables 4 and 5 summarise the main results of the dominance analysis, i.e. the percentage of predicted variance. For the regression equation predicting athletes’ general life satisfaction (table 4), personal self-esteem contributed 57 per cent of the predicted variance, similarity to significant others 31 per cent, and friend support 12 per cent. This order of relative importance is the same as the ordering of the standardised regression coefficients in Table 1. For the regression equation predicting athletes’ satisfaction with training and competition (Table 5), collective self-esteem contributed 67% of the predicted portion of the variance, coaching support contributed 17%, and similarity to significant others contributed 16%. These results suggest that individual self-esteem contributes most to the prediction of general life satisfaction and collective self-esteem contributes most to the prediction of training and competition satisfaction.
Table 5 Relative contribution of three indicators of collective self-esteem, similarity to significant others and coaching support when predicting general life satisfaction
| variant | R2 | X1 | X2 | X3 |
| – | 0 | 0.249 | 0.059 | 0.074 |
| X1 (personal self-esteem) | 0.249 | – | 0.051 | 0.044 |
| X2 (similarity to significant others) | 0.059 | 0.240 | – | 0.037 |
| X3 (friend support) | 0.074 | 0.219 | 0.058 | – |
| X1X2 | 0.299 | – | – | 0.044 |
| X1X3 | 0.293 | – | 0.050 | – |
| X2X3 | 0.132 | 0.211 | – | – |
| X1X2X3 | 0.343 | – | – | – |
| Decomposition of R2 | 0.230 | 0.055 | 0.059 | |
| Percentage of predicted variance | 67.06 | 16.04 | 17.20 |
Discussion of the study
To the best of our knowledge, there are only a few studies on the relationship between collective self-esteem and subjective well-being, and the present study is the first to examine the relationship between collective self-esteem and life satisfaction among athletes. Frocker et al. (1994) conducted a pioneering study on the relationship between collective self-esteem and subjective well-being. However, their study used only general life satisfaction as the dependent variable rather than domain-specific life satisfaction. Additionally, they compared the predictive functions of individual and collective self-esteem dimensions separately rather than comparing individual and collective self-esteem as a whole. Furthermore, they controlled only for the effect of individual self-esteem on general life satisfaction, without considering the influence of other important variables.
Frocker et al. found that, when using the Collective Self-Esteem Scale with the general population as the reference group, only intrinsic collective self-esteem among the four dimensions was reliably correlated with subjective well-being (r=0.13, p<0.05) after controlling for the effect of individual self-esteem. However, when using the subject’s race as the reference group, only extrinsic collective self-esteem showed a significant correlation with subjective well-being (r=0.12, p<0.05). This finding suggests that different dimensions of collective self-esteem may have varying impacts on subjective well-being depending on the context.
The results of our study share some similarities with Frocker et al.’s research, particularly regarding the relationship between collective self-esteem and subjective well-being. However, our study extends the existing literature by exploring the relationship between collective self-esteem and life satisfaction among athletes and by controlling for additional potential variables. These findings provide new directions for future research, particularly in validating the relationship between collective self-esteem and subjective well-being across different groups and cultural contexts.
Referring to the Collective Self-Esteem Scale for groups, then, of the four dimensions of collective self-esteem, only extrinsic collective self-esteem was reliably correlated with subjective well-being, after controlling for the effect of individual self-esteem (r=0.12,p<0.05). A biased correlation of full-scale scores on the Collective Self-Esteem Scale with subjective well-being was not reported by Crocker et al. The present study differed from Crocker et al. in that, firstly, it included special life satisfaction (training match satisfaction) as a variable to provide a more comprehensive examination of the predictive function of collective self-esteem; secondly, it utilised full-scale scores on the Collective Self-Esteem Scale in correspondence with individual self-esteem to provide a clearer and more direct examination of both.
Third, the predictive function of collective self-esteem was examined under more stringent conditions by using demographic variables and social support as control variables; and fourth, a dominance analysis technique was utilised to more reliably assess the relative importance of each predictor variable in the regression equation.
The present study found that although the full-scale scores of the Collective Self-Esteem Scale had a reliable zero-order correlation with general life satisfaction (r=0.21,p<0.05), the Collective Self-Esteem failed to make a new contribution to the prediction of general life satisfaction after controlling for demographic variables, personal self-esteem and social support (△R2=0.004,p>0.05). The present study found that individual self-esteem was the best predictor of general life satisfaction, a result that merely replicated again the findings of previous studies. [1,5,32] But this is only half the story. The most surprising findings of this study are, first, that even after controlling for demographic variables, personal self-esteem, and social support, the results of the present study are not as surprising as they might have been.
After the effects of the three types of variables, collective self-esteem was still able to make a new contribution to the prediction of athletes’ satisfaction with training and competition, i.e., explaining another 13 per cent of the total variance of the dependent variable; secondly, collective self-esteem was the best predictor in the regression equation for predicting athletes’ satisfaction with training and competition, while individual self-esteem failed to enter the regression equation; and thirdly, collective self-esteem contributed a surprisingly high 67 per cent of the variance in the portion of the satisfaction with training and competition that was explained. Third, collective self-esteem accounted for 67% of the variance explained by satisfaction with training competitions. This finding suggests that we need different self-esteem to predict different life satisfaction, and that the contribution of collective self-esteem to life satisfaction cannot be ignored, which may be particularly important for groups in collectivist cultures. The subjects in this study were professional and collegiate athletes who spent most of their lives training and competing with teammates and coaches. Athletes typically spend far more time and engage in far more social interactions within the sports team community than they do within other social groups, including even families. Athletes’ achievements also depend greatly on their team community. Thus, the team community has the most important influence on their satisfaction with training and competition.
This study also highlights the limitations of using only individual self-esteem to predict life satisfaction and emphasizes the importance of incorporating collective self-esteem in such predictions. This finding further underscores the significance of social identity theory and group categorization theory in distinguishing between individual and collective identity. These theories posit that the social dimension of identity stems from the groups to which we belong. We involuntarily compare the characteristics of “our group” with those of “other groups,” and such comparisons tend to favor the group to which we belong. These comparisons generate positive responses and feelings to the question “Who are we?” and foster an awareness of how to differentiate between in-group and out-group members, which is then put into practice.
The contribution of this study lies in its demonstration that both individual and collective identity play significant roles in people’s life satisfaction. When individuals engage in study, work, and other social activities with other group members, collective self-esteem can be a crucial source of occupational or job satisfaction, potentially even more so than individual self-esteem. This finding provides a new perspective on the impact of social identity on individual well-being and highlights the need for further exploration of the role of collective self-esteem across different social and cultural contexts in future research.
The present study used cross-sectional data to explore the relative contribution of predictors of life satisfaction; therefore, even if some statistically reliable predictors were found, it is difficult to derive conclusions of a causal nature from the regression model. There are at least four possible causal relationships that could explain our findings: first, self-esteem and social support are causal and life satisfaction is causal; second, life satisfaction is causal and self-esteem and social support are causal; third, the two are causal; and fourth, the two are only correlated, but not causally related, and another causal variable (Type III) affects the two at the same time, causing them to be correlated. We need a better research design such as an experimental design or cross-lagged survey design to determine the possible causal relationship between predictor and predicted variables. [33-36]
In this study, stepwise regression was used to screen the predictors before the dominance analysis so that the number of predictors was not too large, to reduce the complexity of manual calculations in the dominance analysis, which was a last resort. One of the reasons why stepwise regression has been criticised in social science research [37] and has been used so infrequently is that it is not based on theoretical hypotheses in a field, but only on statistical decisions such as the 0.05 entry and 0.10 screen-out. Such statistical decisions do not indicate that what goes in is gold and what comes out is rubbish. If conditions permit, the best approach is to conduct simultaneous dominance analyses of all variables of interest to the researcher to judge the relative importance of all predictors, guided by theoretical hypotheses. People can identify with several different social groups at the same time. A person’s country, race, gender, and family can all be reasons for social identification. For those who study at school or work in an organisation, the work group may be the most important source of social identity. For example, athletes on the Chinese diving team, both at home and abroad, may use their status as members of the Chinese diving team as a source of pride. The present study shows the important contribution of identification with the workgroup to life satisfaction, while the contribution of identification with other social groups such as country, race, and gender to life satisfaction is an interesting topic for future research.
The subjects of the present study were athletes, a highly specific group characterised by extremely frequent interactions of collective members, which have a significant impact on athletic performance.
This is extremely important. Therefore, future research needs to verify whether the findings from this special group, particularly the role of collective self-esteem, can be generalized to the general Chinese population. Such verification will not only help confirm the universality of collective self-esteem across different groups but also provide important insights into the applicability of social identity theory in broader cultural contexts. Additionally, given the diversity and complexity of Chinese society, future studies should also explore the differential manifestations of collective self-esteem across various social strata, age groups, and occupational groups.