Exploring Factors That Influence Taekwondo Student Athletes’ Intentions to Pursue Careers Contributing to the Sustainability of the Korean Taekwondo Industry Using the Theory of Planned Behavior

Due to recently declining fertility rates and the social climate of job preferences in professional occupations in South Korea, the number of teenagers practicing elite Taekwondo has decreased, and Korean Taekwondo is in crisis. In this regard, it is essential for Taekwondo coaches to create a favorable environment preventing student athletes from abandoning careers in Taekwondo and for physical education researchers to explore factors directly or indirectly influencing the intentions to pursue such careers. Thus, the purpose of this study was to examine the structural relationships among mentoring, attitudes, subjective norms, perceived behavioral control, and career pursuit intentions by applying the theory of planned behavior, as well as investigating the moderating influence of Taekwondo identification on these relationships. We collected data from athletes with more than a year of elite Taekwondo experience attending one of 15 high schools in South Korea. We asked a total of 270 athletes to participate in the survey. Of these, 250 completed the survey. We eliminated data for 15 athletes due to repetitive response patterns; thus, we analyzed 235 usable responses. We assessed the measurement scale’s validity and reliability with confirmatory factor analysis, Cronbach’s alpha coefficients, and correlational analysis. Structural equation modeling evaluated the effects of four factors—mentoring, attitudes, subjective norms, and perceived behavioral control—on career pursuit intentions. Additionally, we performed a hierarchical regression analysis to confirm the moderating effect of Taekwondo identification. The results revealed positive impacts of (a) mentoring on attitudes (0.760, p < 0.001), (b) attitudes on career pursuit intentions (0.681, p < 0.001), (c) subjective norms on career pursuit intentions (0.141, p < 0.01), and (d) perceived behavioral control on career pursuit intentions (0.138, p < 0.05). However, Taekwondo identification did not moderate the relationships among the research variables.


Introduction
Taekwondo has been the representative sport and brand of Korea during the last several decades [1]. Currently, 210 countries are affiliated with the World Taekwondo Federation, and Taekwondo has become a truly world-class martial arts sport [2]. South Korea has won the most Olympic medals in Taekwondo and boasts its commanding position as the birthplace of Taekwondo. Taekwondo has made long strides in the world, starting with the adoption of the official sport in the 2000 Sydney Olympics. Although the 2020 Tokyo Olympics were not held as scheduled due to COVID-19, Taekwondo solidified its status as one of the most critical sports [3]. Research related to Taekwondo has been multiplied since it became an official Olympic sport in 2000 [4].
Although South Korean Taekwondo has exhibited a continuous growth, researchers have pointed out several problems concerning this growth [5]. Due to recently declining model associated with the determinants of behaviors and intentions [12]. The central ideas of the TRA are that an action results from intentions to perform it and that action intentions can predict individual behavior [15]. The theory emphasizes that individuals' intentions to act determine their actual behavior. These intentions reflect two main attributes: attitudes and subjective norms [16]. Attitudes are based on the belief that a particular action would produce certain consequences and one's feelings about these outcomes [17]. Therefore, attitudes in the TPB refer to individuals' positive or negative feelings toward behaviors [18].
Concerning subjective norms, when an individual performs a particular behavior, he or she is likely to be influenced by perspectives of and information from (significant) others (e.g., family, friends, colleagues, teachers, etc.). For example, when family members and friends do not support an individual's behavior, the individual might become less likely to perform that behavior in the future. Therefore, subjective norms represent the normative beliefs about an action. They refer to the individuals' belief about whether they will follow other people's directions and the motivation to conform to that belief [16]. The theory of reasoned action is useful when applied to situations that can be handled by the individual will; however, in cases where control of the individual will is more difficult, the theory has limitations. As external factors influence subjective norms, tangible differences in results pertaining to behavioral intentions and correlated variables emerge across research areas, subjects, and contexts. Specifically, there are cross-cultural differences in the importance of subjective norms. In the East there exist strong collectivist cultures, whereas in the West firm individualism dominates; in the latter case, individualism as a subjective norm has a greater influence on behavioral decision-making [19]. In the case of elite Taekwondo athletes in Korea, subjective norms constitute a significant variable.
Let us suppose that someone greatly enjoys working out at the Taekwondo gym and shows a positive attitude, and that other people assume that Taekwondo training has a positive influence on them. However, uncontrollable circumstances-such as the administrative order to ban gatherings due to COVID-19 or the suspension of business by gym owners for personal reasons-will always remain, which can affect the behavioral intentions. Accordingly, Ajzen [16] found that human behaviors exist on a continuum from completely controllable to uncontrollable by the human will, later publishing the TPB to encapsulate this insight. The theory aimed to overcome the limitations of the theory of reasoned behavior. Researchers added the construct of PBC to the TPB, which is the subjective evaluation of an individual's control over and potential for performing an action. Furthermore, PBC entails the confidence necessary to perform an action and the perception of one's ability to overcome obstacles. Factors that can affect PBC include opportunities, money, resources, and technology, which are not easy for individuals to control entirely. Thus, humans consider both internal capabilities and external factors, determining their behavioral intentions and ultimately leading to actual behaviors [20,21].

Mentoring
There is a saying that "successful people have mentors". Mentoring refers to interactions in which more experienced (and oftentimes older) individuals provide guidance, personal and career development, and assistance to less experienced individuals [22]. As successful people in various fields tend to have plenty of support from mentors, we can infer that mentors are likely a decisive factor in their success [23]. Mentors must play a variety of roles to help mentees lead stable and successful lives, emotionally and socially. In other words, mentors must determine and fulfill various functions that are helpful to mentees, including technical instruction, counseling, friendship, and assistance. Therefore, mentoring encompasses the systematic and continuous relationship between mentors and mentees. Mentors should prepare mentees for challenges and growth based on their own experience and knowledge [24].
Mentoring is common in business and academic contexts. It also entails the process of providing steady guidance and the handing down of techniques by veterans with a wide range of experience and knowledge. The person who provides advice is called the Sustainability 2021, 13, 9893 4 of 16 "mentor", and the recipient of the advice is called the "mentee". A mentor's role is to make the best of the mentee's abilities to help them achieve their goals [25,26]. Therefore, mentoring must be based on trust and dedication, and the relationship should be relatively close for mentees to thrive [27,28].
Humans tend to imitate or follow other people, and by doing so they develop their attitudes and values. Therefore, mentors should serve not only as advisers but also as role models to facilitate mentees' growth. Research related to mentoring has concentrated primarily on mentees' changes, as mentoring offers optimism, education, and life lessons to both mentors and mentees. For instance, Nash [29] studied how mentoring systems function as coaching, and Pastore [30] investigated mentoring functions' impact in sports management. These studies identified performance gaps between formal and informal mentoring and the differential effects of mentoring depending on organizational characteristics.
In a study of sports leaders and managers in US universities, Weaver and Chelaudurai [31] identified several key characteristics of mentoring. First, they categorized mentors by age, social experience, social status, and personality. Second, they categorized mentees according to ability, gender, race, and personality. Third, they determined the key characteristics of mentoring functions. They also explained the relationship between mentors and mentees and the impact on organizational performance [31].
The research on the relationship between mentors and mentees has steadily progressed in various fields. Mentoring represents one key educational method utilized by sports leaders; thus, much research has examined the mentoring role of leaders in a variety of professional sports. Numerous prior studies emphasize that mentoring improves athletes' commitment to achieving their goals, as well as their confidence [32]. In addition to its important role in professional sports, mentoring also serves as an educational tool in schools' physical education programs, dance programs, and other physical exercise programs [33]. Sports leaders must train athletes and understand each athlete's psychological states and physical characteristics in order to bond with them and to help them achieve optimal results. Thus, sports leaders should recognize their mentoring roles' im-portance and develop professional and systematic coaching methods to enhance performance.

Sports Identification
Previous studies have defined identification as a sense of belonging or being one with an organization [34]. In social psychology, social identification theory addresses the emergence of the identification with the groups or specific organizations to which individuals belong and how this identification affects attitude formation. Individuals strongly identifying as leaders tend to be more immersed and interested in the needs of leaders [35]. Additionally, this increases subordinates' self-esteem and the emotional bonds with leaders. Thus, organizational identification plays a key role in improving organizational outcomes, such as job performance [36]. While business-related identification literature has focused on employees' organizational identification, the variable of interest in the sports literature is fans' team identification [37]. However, some researchers have highlighted that individuals might identify both with their favorite sports teams and with the sport itself [38]. The latter situation is called "sport identification" [39]. Individuals interested or immersed in practicing Taekwondo are likely to become more connected to the sport overall. Thus, such individuals might say "I am into Taekwondo" rather than "I am a fan of this Taekwondo team".

Research Hypotheses Development
Based on previous studies, mentoring likely influences attitudes. In the physical education context, Yoo and Nam [40] investigated the relationship between instructors' mentoring and the training attitudes of Judaka, demonstrating mentoring as a predictor of attitudes. Further, Shin, Jun, Zang, and Kwon [41] explored the impact of mentoring on employees' job attitudes and performance in sport centers, confirming the former's explanatory power with respect to the latter. In a similar vein, Kim, Im, and Hwang's [42] Sustainability 2021, 13, 9893 5 of 16 study developed and tested a research model examining the effects of mentoring on role stress, job attitudes, and turnover intentions, finding that mentoring indeed positively influenced job attitudes. Based on these findings, we proposed the following hypothesis: Hypothesis 1 (H1). Mentoring will positively influence Taekwondo student-athletes' attitudes.
Many studies support the attitude-behavioral intentions link. For example, a recent study conducted by Jeong, Kim, and Yu [43] used the TPB to test the process behind sports fans' decisions to attend sports matches amid the pandemic, showing that attitudes crucially affected event attendance intentions. Further, Kim, Chung, Chepyator-Thomson, Lu, and Zhang [44] explored factors influencing audiences' attitudes and intentions to watch LPGA tournaments and uncovered a positive association between the two variables. To better understand the intentions to attend a sporting event (i.e., hockey), Cunningham and Kwon [45] used the TPB as an empirical framework. Their findings also showed that attitudes could affect attendance intentions. Thus, we formulated the following hypothesis: Hypothesis 2 (H2). Attitudes will positively influence career pursuit intentions among Taekwondo student-athletes.
Many studies have explored the relationship between subjective norms and behavioral intentions. Based on the extended TPB, Zhang and Cha [46] examined the factors that influence intentions to watch e-sports competitions in China, showing that subjective norms positively affected viewing intentions. Bae, Won, Lee, and Pack [47] analyzed adolescents' participation in new sports using the extended TPB, including prior knowledge. As expected, subjective norms influenced participation intentions. Moreover, a study conducted by Chuan, Yusof, Soon, and Abdullah [48] found that high school students' intentions to participate in recreational sports activities significantly depended on subjective norms. Therefore, we proposed the following hypothesis: Hypothesis 3 (H3). Subjective norms will positively influence career pursuit intentions among Taekwondo student-athletes.
There has been an increasing awareness that PBC is related to behavioral intentions. Rigby, Vela, and Housman [49] investigated the effect of athletic trainers' beliefs about the current concussion-management guidelines using the TPB framework and found that PBC predicted behavioral intentions. In a different context, Verma and Chandra [50] demonstrated PBC as a determinant of young consumers' intentions to visit a "green" hotel. Yadav and Pathak [51] attempted to incorporate additional constructs (environmental concern and environmental knowledge) in assessing young consumers' intentions to purchase green products, reporting the significant influence of PBC on purchase intentions. These findings led to the following hypothesis:

Hypothesis 4 (H4).
Perceived behavioral control will positively influence career pursuit intentions among Taekwondo student-athletes.
Another question concerns whether the identification with the sport of Taekwondo exerts a moderating influence on the relationships between attitudes and career pursuit intentions, subjective norms and career pursuit intentions, and PBC and career pursuit intentions among Taekwondo student-athletes. As discussed previously, attitudes, subjective norms, and PBC likely influence career pursuit intentions [43][44][45][46][47][48][49][50][51]. With respect to the relationship between identification and behavioral intentions, Suh, Ahn, and Pedersen [52] examined the impact of team identification, e-service quality, and satisfaction on consumers' intentions to revisit sports websites. They showed that team identification positively affected revisit intentions. Recent research conducted by Kim, Rogol, and Lee [53] explored a conceptual model delineating the impact of ice hockey spectators' satisfaction on team identification and revisit intentions, revealing a relationship between team identification and revisit intentions. Moreover, many researchers have analyzed the moderating role of identification in sports contexts. For example, Theodorakis, Koustelios, Robinson, and Barlas [54] tested team identification's moderating role in the relationship between service quality and repurchase intentions among spectators of professional sports. As hypothesized, they found that spectators' intentions differed depending on the level of team identification. On the other hand, in the context of sports marketing, Jeong et al. [43] showed that team identification had no moderating effect in the relationships of attitudes, subjective norms, and PBC with intentions to attend a sporting event amid the COVID-19 pandemic. Therefore, the present study explored the moderating role of Taekwondo identification, and we proposed the following hypotheses: Hypothesis 5 (H5). Taekwondo identification will moderate the relationship between attitudes and career pursuit intentions among Taekwondo student-athletes.
Hypothesis 6 (H6). Taekwondo identification will moderate the relationship between subjective norms and career pursuit intentions among Taekwondo student-athletes.

Hypothesis 7 (H7).
Taekwondo identification will moderate the relationship between perceived behavioral control and career pursuit intentions among Taekwondo student-athletes.
Based on the preceding thorough review of existing studies, we proposed the conceptual model shown in Figure 1.
intentions among Taekwondo student-athletes. As discussed previously, attitudes, subjective norms, and PBC likely influence career pursuit intentions [43][44][45][46][47][48][49][50][51]. With respect to the relationship between identification and behavioral intentions, Suh, Ahn, and Pedersen [52] examined the impact of team identification, e-service quality, and satisfaction on consumers' intentions to revisit sports websites. They showed that team identification positively affected revisit intentions. Recent research conducted by Kim, Rogol, and Lee [53] explored a conceptual model delineating the impact of ice hockey spectators' satisfaction on team identification and revisit intentions, revealing a relationship between team identification and revisit intentions. Moreover, many researchers have analyzed the moderating role of identification in sports contexts. For example, Theodorakis, Koustelios, Robinson, and Barlas [54] tested team identification's moderating role in the relationship between service quality and repurchase intentions among spectators of professional sports. As hypothesized, they found that spectators' intentions differed depending on the level of team identification. On the other hand, in the context of sports marketing, Jeong et al. [43] showed that team identification had no moderating effect in the relationships of attitudes, subjective norms, and PBC with intentions to attend a sporting event amid the COVID-19 pandemic. Therefore, the present study explored the moderating role of Taekwondo identification, and we proposed the following hypotheses: Hypothesis 5 (H5). Taekwondo identification will moderate the relationship between attitudes and career pursuit intentions among Taekwondo student-athletes.
Hypothesis 6 (H6). Taekwondo identification will moderate the relationship between subjective norms and career pursuit intentions among Taekwondo student-athletes.

Hypothesis 7 (H7). Taekwondo identification will moderate the relationship between perceived behavioral control and career pursuit intentions among Taekwondo student-athletes.
Based on the preceding thorough review of existing studies, we proposed the conceptual model shown in Figure 1.

Participants
The current study utilized a convenience sampling approach. We distributed the survey to elite Taekwondo athletes involved in a mentoring program, each of whom was attending one of 15 high schools in Seoul, South Korea. The online questionnaires took the form of a Google survey from 1 to 30 March 2021. Since the academic year in South Korea begins during March and ends in the middle of July (1st semester) and starts again in August, ending in mid-February (2nd semester), and since the competitive season of Taekwondo begins in early April and ends in late November, we collected data from athletes for March not to interrupt a mid-term exams, a final exam, or a competition. We contacted 15 high school teachers and coaches using phone, fax, or e-mail and received their students' e-mail addresses or Kakao Talk (the most widely used messaging app for Sustainability 2021, 13, 9893 7 of 16 smartphones and personal computers in South Korea) IDs after obtaining mutual consent. We asked a total of 270 athletes to participate in the survey (without reward). Of these, 250 completed the survey, yielding a response rate of 92.6%. We eliminated data for 15 athletes due to repetitive response patterns; thus, we analyzed 235 usable responses. We gathered demographic information including sex (male: 66%, n = 155; female: 34%, n = 80), grade (first-year students: 28.9%, n = 68; second-year students: 36.2%, n = 85; third-year students: 34.9%, n = 82), and Taekwondo experience (less than 2 years: 3.4%, n = 8; more than 2 years and less than 4 years: 33.6%, n = 79; more than 5 years and less than 7 years: 46.8%, n = 110; 8 years or more: 16.2%, n = 38).

Measurement
The survey instrument consisted primarily of measures adapted from previous studies. We assessed mentoring using four items adopted from Noe [55]. We assessed attitudes using four items adapted from Kim et al. [44], Ajzen [56], and Han, Meng, and Kim [57]. We measured subjective norms with four items adopted from Jeong et al. [43], Kim et al. [44], and Ajzen [56]. We derived our measure of PBC from Ajzen [16], Han et al. [57], and Perugini and Bagozzi [58]. We used three items from Robinson, Trail, and Kwon [59] to assess Taekwondo identification. Finally, for career pursuit intentions, we used three items from Davis, Bagozzi, and Warshaw [60] and Lee [61]. The response format was a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). A panel of two physical education professors and one Taekwondo coach reviewed the instrument's content validity. We revised the preliminary questionnaire based on the panel's feedback before distributing it to participants. In addition, a professor affiliated with the department of interpretation and translation in a University in Seoul was asked to validate the translation of the questionnaire.

Validity and Reliability
We employed confirmatory factor analysis (CFA) with maximum likelihood estimation to assess the dimensionality of the measurement model using AMOS (version 24). The goodness-of-fit indices for the CFA (χ 2 /df = 2.310, NFI = 0.912, IFI = 0.948, TLI = 0.938, CFI = 0.948, and RMSEA = 0.075) were all within the recommended ranges [62]. Next, we calculated the factor loadings, construct reliability (CR), and average variance extracted (AVE) to assess the convergent validity based on the measurement model. As shown in Table 1, all factor loading values (0.669-0.952) were statistically significant (p < 0.001) and greater than the cutoff value of 0.50 [63]. All CR values (0.864-0.967) exceeded the recommended minimum value of 0.70, and all AVE values (0.679-0.882) exceeded the minimum of 0.50 [64]. Hence, the convergent validity was satisfactory. For the satisfactory discriminant validity, the diagonal elements in Table 2 should be greater than the off-diagonal elements. Comparing all correlation coefficients with the square roots of AVE demonstrated a satisfactory discriminant validity. In terms of the survey instrument's reliability, the Cronbach's alphas for the six factors ranged from 0.840 (PBC) to 0.948 (attitudes), all of which exceeded the recommended threshold of 0.70. This result suggests that the measures were sufficiently reliable [64].

Moderating Effect of Taekwondo Identification
We used a hierarchical regression analysis to evaluate the moderating effect of Taekwondo identification. As shown in Table 3, in Model 1, the regression was significant (F = 99.849, p < 0.001, R 2 = 0.559). Attitudes (β = 0.556, t = 7.472, p < 0.001), and PBC (β = 0.152, t = 2.192, p < 0.05) had significant impacts on career pursuit intentions, whereas subjective norms did not. In Model 2, the regression was also significant (F = 97.847, p < 0.001, R 2 = 0.623). When entering Taekwondo identification into the model, the variable was significant (β = 0.383, t = 6.368, p < 0.001). Finally, in Model 3, the regression equation had a significant effect (F = 58.559, p < 0.001, R 2 = 0.633), and the Durbin-Watson statistic value of 2.102 indicated no autocorrelation in the sample. When entering three interaction variables (i.e., attitudes × Taekwondo identification, subjective norms × Taekwondo identification, and PBC × Taekwondo identification) into the model, these variables did not emerge as significant predictors of career pursuit intentions. Therefore, Taekwondo identification did not moderate the relationships between attitudes and career pursuit intentions, subjective norms and career pursuit intentions, and PBC and career pursuit intentions; thus, Hypotheses 5, 6, and 7 were not supported.

Moderating Effect of Taekwondo Identification
We used a hierarchical regression analysis to evaluate the moderating effect of Taekwondo identification. As shown in Table 3, in Model 1, the regression was significant (F = 99.849, p < 0.001, R 2 = 0.559). Attitudes (β = 0.556, t = 7.472, p < 0.001), and PBC (β = 0.152, t = 2.192, p < 0.05) had significant impacts on career pursuit intentions, whereas subjective norms did not. In Model 2, the regression was also significant (F = 97.847, p < 0.001, R 2 = 0.623). When entering Taekwondo identification into the model, the variable was significant (β = 0.383, t = 6.368, p < 0.001). Finally, in Model 3, the regression equation had a significant effect (F = 58.559, p < 0.001, R 2 = 0.633), and the Durbin-Watson statistic value of 2.102 indicated no autocorrelation in the sample. When entering three interaction variables (i.e., attitudes × Taekwondo identification, subjective norms × Taekwondo identification, and PBC × Taekwondo identification) into the model, these variables did not emerge as significant predictors of career pursuit intentions. Therefore, Taekwondo identification did not moderate the relationships between attitudes and career pursuit intentions, subjective norms and career pursuit intentions, and PBC and career pursuit intentions; thus, Hypotheses 5, 6, and 7 were not supported.   Note: TI: Taekwondo identification. * p < 0.05, ** p < 0.01, *** p < 0.001.

Theoretical Implications
The present study extended the findings of previous studies with respect to the impact of mentoring on athletes' attitude formation. More specifically, our results supported the mentoring's strong influence, aligning with several existing studies' results [40][41][42]. Thus, it is evident that coaches' mentoring acts to foster athletes' positive attitudes toward their sports. If mentees receive assistance in achieving personal goals, learning new physical skills, choosing careers, and overcoming slumps through a successful mentoring program, they will likely surmount challenges and gain confidence in the process of playing sports, fostering positive attitudes toward sports. Therefore, mentors or instructors should invest heavily in improving sports mentoring programs to engender these favorable attitudes.
Additionally, we considered the argument advanced by previous studies that attitudes can shape behavioral intentions. For example, a recent study conducted by Wang, Won, and Jeon [65] examined the influences on students' sports gambling behavioral intentions and actual behavior using the TPB as a framework, with the results showing that attitudes were the most critical determinant of intentions. Further, Tariq, Sajjad, Usman, and Amjad [66] tested an extended model of the TPB to understand the social networking website (particularly, Facebook) use among 480 students between 16 and 24 years of age, showing that favorable attitudes toward Facebook significantly predicted use intentions. Moreover, our findings (specifically regarding Hypotheses 1 and 2) indicated that mentoring indirectly affected career pursuit intentions. Thus, a well-functioning mentoring program or system will likely improve career pursuit intentions [65].
This study also clarified the relationship between subjective norms and career pursuit intentions, demonstrating the former's positive impact on the latter. However, some researchers have put forth an alternative argument to explain the relationship. For example, Wang, Wong, Narayanan Alagas, and Chee [67] found a significant negative relationship between subjective norms and green product purchase behavioral intentions. In contrast, Paul, Modi, and Patel [68] validated the TPB and its extended form to predict consumers' green product purchase intentions but found the path from subjective norms to purchase intentions to be non-significant. Nevertheless, the current study provides sup-porting empirical evidence for the positive effect of subjective norms on behavioral intentions, which aligns with many other previous studies' findings [46][47][48]. Therefore, Taekwondo instructors should maintain good relations with students' families and pro-vide them with a thoughtful educational philosophy that could help motivate students to pursue their dreams.
This study contributes to the debate about the role of PBC in the TPB model. Although PBC has been viewed as a moderating variable, our study found that PBC was a direct determinant of career pursuit intentions. Ajzen [10], a pioneer of the TPB, has pointed out that, theoretically, PBC could be considered as a moderating variable, in accordance with several studies [69]. On the other hand, many previous studies have treated PBC as a direct determinant of intentions. For example, Rahaman, Stouten, and Guo [70] explored the antecedents of ethical leadership using the TPB, demonstrating PBC as a direct determinant of ethical intentions. Similarly, Schuster, Kubacki, and Rundle-Thiele [71] examined the factors influencing walk-to-school behavior through the application of the TPB and found that that PBC was a direct antecedent of behavioral intentions. Therefore, student-athletes' decisions about whether to continue practicing Taekwondo may result partially from their sense of competence. Suppose that Taekwondo student-athletes perceive some barriers to practicing elite Taekwondo, such as the lack of money, time, knowledge, and skills. In that case, their parents and Taekwondo instructors should provide the athletes with sensible solutions in order to foster their career pursuit intentions.
Although we expected a moderating effect of Taekwondo identification in the relationships between attitudes and career pursuit intentions, subjective norms and career pursuit intentions, and PBC and career pursuit intentions, this study's results did not support our hypothesis. This result can be explained in several ways. First, our sample's characteristics differed from those of previous studies' samples. This study's data came from high school students, while prior studies have examined various age groups. For in-stance, participants in a sample recruited by Theodorakis et al. [54] ranged from 18 to 72 years of age, with an average age of 34.81 years. Second, as elite athletes, Taekwondo student-athletes have stronger attachments to Taekwondo than do amateur athletes. Generally, elite athletes have much experience and desire to continue their careers in the sport, which cannot exist without an attachment to or identification with sports. However, among amateur athletes, sport identification could moderate the relationships among the research variables. Thus, in future research, it would be worth examining the moderating effect of sport identification among amateur athletes.

Practical Implications
The role of coaches is to strengthen athletes' psychological stability so that athletes can display their best abilities in the stadium. A superior mental health can decide games between athletes with similar skill levels. To fulfill this role, coaches must lead athletes through a systematic preparation, resulting in increased trust and, ultimately, a strong athletic performance. Therefore, even if coaches' technical guidance is essential, mentoring (or emotional guidance) is still important. While psychology is not necessarily more important than physical strength or skills, to perform at a higher level, athletes must possess mental toughness.
During the ongoing COVID-19 pandemic, continual viral mutation will likely pose challenges for athletes. Athletes may experience a variety of psychological difficulties due to COVID-19, including confusion, sense of loss, nervousness, sleep disorders, and lethargy due to reductions in competition and rule changes. Of course, more research will be needed to evaluate these potential effects. Nevertheless, if coaches attempt to encourage athletes who are anxious over a confirmed COVID-19 diagnosis not to worry about mis-takes during the training process and show empathy for their various hardships, this will strongly impact athletes' skills and performance. Therefore, Taekwondo coaches should mentor athletes like mother kangaroos taking care of their babies in their pouches, not giving off the appearance of strong lions. More than ever, now is the time for mentors not only to provide relevant knowledge but also to listen to athletes' concerns. When Taekwondo athletes struggle to reach their personal goals, coaches should help them to avoid repeating the same mistakes by addressing their deficiencies constructively.
It is imperative to provide practical assistance for elite Taekwondo athletes to continue their training. Listening to athletes' concerns and resolving conflicts with colleagues will also prevent athletes from abandoning the sport. Additionally, it is necessary to develop internal and external policies to recruit and train excellent Taekwondo athletes. However, there is little structure to accomplish this. Even if one is an outstanding athlete, it is still possible for them to want to quit. Through competitions for elementary, middle, and high school students, the World Taekwondo Federation should invest heavily in cultivating athletes' talent worldwide. On another note, under the extenuating circumstances imposed by COVID-19, it would be helpful for athletes to be able to assess their own psychological stability and receive feedback online. Moreover, for athletes suffering from mental health issues, providing HyFlex counseling and psychological technical training in self-discipline could be highly beneficial.
Furthermore, a warm support from those who exert strong influences on elite youth Taekwondo athletes is crucial. Praising athletes' efforts and success will give them the strength not to be frustrated even amid challenges. The pivotal element of subjective norm-based action plans is that coaches maintain good relationships with athletes' parents. Therefore, coaches should report to the athlete's parents the athletes' current skill level, elaborate on their educational philosophy, and build trust. Additionally, coaches should discuss athletes' developmental potential, share their opinions about the preparation for the next competition, and communicate with athletes to help them advance toward achievable goals. While also presenting confidence, vision, and goals to the parents of youth Taekwondo athletes, coaches must develop trust with parents, which may be just as important as developing trust with the athletes themselves.
Economic support is also a non-negligible factor. This is because even if athletes want to concentrate on Taekwondo, they will not be able to do so under heavy financial burdens. Regardless of the athletes' skill level, if they are in a situation where they may need to give up the sport for financial reasons, institutional support will be of great value to them. Even if the policy is pursued, it will be unfortunate if the system is not used well. Favorable outcomes will result from joint efforts to use the system appropriately. This is something upon which everyone who loves Taekwondo should reflect deeply.

Summary
Due to recently declining fertility rates and the social climate giving preference to professional occupations in South Korea, the number of teenagers practicing elite Taekwondo has decreased, and Korean Taekwondo faces a crisis. In this regard, it is important for Taekwondo coaches to foster a positive environment, preventing student athletes from abandoning careers in Taekwondo, and for researchers to explore factors directly or indirectly influencing the intentions to pursue such careers. Thus, the objective of this study was to examine the structural relationships among mentoring, attitudes, subjective norms, perceived behavioral control, and career pursuit intentions among elite high school Taekwondo athletes, applying the theory of planned behavior and assessing the moderating effect of Taekwondo identification. The results revealed positive impacts of (a) mentoring on attitudes, (b) attitudes on career pursuit intentions, (c) subjective norms on career pursuit intentions, and (d) perceived behavioral control on career pursuit intentions. On the other hand, Taekwondo identification did moderate the relationships between attitudes and career pursuit intentions, subjective norms and career pursuit intentions, or perceived behavioral control and career pursuit intentions. The findings provided above have important implications. (1) The current study extended the findings of existing studies concerning the impact of mentoring on athletes' attitude formation. (2) The present study considered the argument advanced by previous studies that attitudes can build behavioral intentions. (3) This study clarified the relationship between subjective norms and career pursuit intentions, demonstrating the former's positive impact on the latter. (4) Although PBC has been viewed as a moderating variable, our study found that PBC was a direct determinant of career pursuit intentions.

Future Study Directions
Despite these meaningful results, our study has some limitations. First, the sample included only elite high school student-athletes. To ensure the external validity of the measures of mentoring, attitudes, subjective norms, and PBC and to confirm the moderating effect of sport identification, it might be necessary to recruit participants from various age groups and to survey amateur athletes. Second, the TPB may not have had sufficient explanatory power. Hence, future studies could apply the model of goal-directed behavior, which has been introduced to overcome the limitations of the TPB. Third, we did not examine the beliefs influencing subjective norms and PBC, respectively. Thus, future research should explore the antecedents of subjective norms and PBC to understand them better. Finally, in future research, it would be fruitful to investigate the impact of educational service quality on attitudes, as the former likely plays a vital role in this context.