Self-perception of dual career barriers and athletic identity in student-athletes with disabilities according to disability type and level of professionalization

The objective of this study was to analyze the perceived barriers to dual career success and athletic identity of student-athletes according to disability type and level of professionalization. The final sample consisted of 203 student-athletes with disabilities from five European countries. The questionnaires used were ESTPORT, EBBS and AIMS. Depending on disability type, it was found that student-athletes with hearing and physical impairment showed the highest difficulty in reconciling sports and studies (p = 0.001); that student-athletes with a hearing impairment showed the highest score in the barrier ‘the cost of education is high’ (p = 0.023); that student-athletes with a physical impairment had the highest scores in the barrier ‘Exercise tires me’ (p = 0.013); that student-athletes with cerebral palsy showed the highest scores in the barrier ‘I do not have enough university/educational institution support’ (p = 0.014) and ‘Exercise facilities do not have convenient timetables for me’ (p = 0.001). Depending on sports professionalization level, semi-professional student-athletes showed the highest values in the barrier ‘the university/educational institution is far from my training center’ (p = 0.040); while professional student-athletes had the highest score in the barrier ‘exercise takes too much time from family responsibilities’ (p = 0.034). In most of the variables related to identity as athletes, professional student-athletes showed the highest values, followed by semi-professional athletes (p = 0.043- < 0.001). In conclusion, the self-perception of barriers is quite relevant, with differences arising from disability type and level of professionalization, whereas the identity as an athlete is only different according to the level of professionalization.


Participants
The sample size was calculated using Rstudio 3.15.0software (Rstudio Inc., USA).The significance value was set at α = 0.05.The standard deviation (SD) was established considering to perceived barriers of previous studies (SD = 0.75) 15 .With an estimated error (d) of 0.10, the required sample size for a 99% confidence interval (CI) was 200 subjects.The final sample consisted of 203 student-athletes with disabilities from five European countries (Spain, Portugal, Italy, Ireland, and Romania).The inclusion criteria were defined on the basis of previous studies 39 : (a) have a physical, sensory (visual or hearing) disability, or cerebral palsy; (b) been enrolled in a sports federation for at least three years; (c) to be currently enrolled in the last academic years of compulsory education (pre-university education), a university degree, a university master's degree, or a doctorate, and (d) have an adequate level of reading comprehension in English in order to complete the questionnaire.
The sample characteristics are shown in Table 1.Depending on the type of disability, significant differences were found only for level of professionalization.The visually impaired had a higher percentage of professional athletes, the hearing impaired had a higher percentage of semi-professional athletes, and the cerebral palsy group had a higher percentage of amateur athletes, while the physically impaired had the greatest heterogeneity in terms of level of professionalization (p = 0.007).
Depending on the level of sports professionalization, significant differences were found in the type of disability (p = 0.007).The highest percentage of amateur athletes had a physical disability or cerebral palsy; the highest percentage of semi-professional athletes had a physical or sensory disability; and the highest percentage of professional athletes had a physical or visual disability.Differences were also found according to the level of professionalization in the stage of their sports career (p = 0.002).Thus, amateur and semi-professional athletes were more likely to be at the beginning of their sporting career, while professional athletes were more likely to be at the peak of their career.Lastly, significant differences were found according to the level of professionalization in the hours per week studying (p = 0.002) and doing sports (p = 0.001).The Bonferroni adjustment showed that, regarding hours spent studying, professional athletes spent significantly fewer hours than semi-professional athletes (Mean dif.: 9.83 ± 2.84; p = 0.002; 95%ICC 2.96; 16.70) and amateur athletes (Mean dif.: 7.51 ± 2.92; p = 0.033; 95%ICC 0.44; 14.59).Regarding the hours spent in their sports, amateur athletes spent significantly fewer hours than professional athletes (Mean dif.: − 10.88 ± 3.04; p = 0.001; 95%ICC − 18.24; − 3.53) and semiprofessional athletes (Mean dif.: − 7.39 ± 2.77; p = 0.025; 95%ICC − 14.08; − 0.70).

Perceptions of dual career student-athletes
To measure the perceptions of dual career student-athletes, the 'Perceptions of dual career student-athletes' (ESTPORT) questionnaire 54 was used, as in previous research in dual career athletes with and without disability 14,39,55 .The questionnaires were completed in English by all participants.The internal consistency of the questionnaire is high (Cronbach's alpha coefficient > 0.70) with a value of α = 0.857 obtained in the present study, understood as a high reliability 56 .The original version is composed of 84 items.To obtain information about sociodemographic and contextual variables, questions number 1, 2, 7, 8, 9, were included.Furthermore, to know the difficulty of reconciling sporting and academic life, question 20 was also included.Finally, to discover the perceived barriers, the scores obtained in items 26 to 37 of the questionnaire were included.These questions used a Likert scale from 1 (strongly disagree) to 5 points (strongly agree).

Exercise benefits/barriers
To analyze the exercise benefits and barriers, the 'Exercise Benefits/Barriers Scale' (EBBS) 57 was used.The questionnaires were completed in English by all participants.The resulting instrument was tested for internal consistency (α = 0.954), validity of its constructs (variance explained: 65.2%), and test-retest reliability (ICC = 0.89) 57 .The Cronbach's alpha coefficient in this study was α = 0.776, understood as a high reliability 56 .This questionnaire has been used in research for the analysis of exercise benefits/barriers in population with disabilities 58 , athletes 59 or university students 60 .For this research, items about the barrier scale were included.These questions used a Likert scale from 4 (strongly agree) to 1 (strongly disagree).

Athletic identity
The reduced version of the ' Athletic Identity Measurement Scale' (AIMS) was used 61 to measure athletic identity.The questionnaires were completed in English by all participants.The Cronbach's alpha coefficient in this study was α = 0.776, understood as a high reliability 56 .This questionnaire has been widely used in athletes 62,63 or dual career students-athletes 64 .The AIMS is composed of seven items designed to assess aspects of athletic identification, with the athlete's role measured on a scale ranging from 1 (strongly disagree) to 7 (strongly agree).www.nature.com/scientificreports/

Procedure
Universities from the participating countries (Spain, Portugal, Italy, Ireland, and Romania) contacted their sports service to send the questionnaire to athletes with disabilities enrolled at the universities, as well as local, regional and national associations, and foundations whose main focus was on athletes with disabilities, and the country's Paralympic Committee.The questionnaire was then circulated by email to all disabled athletes in their databases, specifying that it should only be completed by those who were currently enrolled in pre-university studies, university degree, or university post-degree studies.The participants completed an informed consent form and a questionnaire anonymously and individually, without academic or competitive pressure.All the questionnaires were completed in English by all participants.The questionnaire was disseminated through Google Forms®, and the participants completed it in 20-30 min.All the data was collected anonymously.www.nature.com/scientificreports/

Statistical analysis
The normality of the data was initially assessed with the Kolmogorov-Smirnov test, homogeneity with the Levene's test, and sphericity with the Mauchly test.All the variables included in the analysis showed a normal distribution, so parametric tests were performed.The descriptive analysis of quantitative variables shows mean values and standard deviations, while frequencies and percentages were calculated for qualitative variables.The chi-square analysis (χ 2 ) made it possible to establish the differences in the qualitative variables (gender, type of disability, level of education, level of sport professionalism, and stage of the sports career), depending on type of disability and level of sports professionalization.Cramér's V was used for the post hoc comparison, and the contingency coefficient was used to obtain the statistical value.The maximum expected value was 0.707; r < 0.3 indicated a low association; r < 0.5 indicated a moderate association; and r > 0.5 indicated a high association 65 .
For the analysis of the differences in the quantitative variables (age, hours studying/doing sports, difficulty in reconciling sport and studies, barriers towards achieving a good balance between sport and studies, exercise benefits/barriers and athletic identity) depending on type of disability and level of sports professionalization, a one-way analysis of variance (ANOVA) was used, with the Bonferroni pairwise comparison carried out with the variables with statistical significance, adjusting for the value of p. Partial eta squared (η2) was used to calculate the effect size (ES), and was defined as small: ES ≥ 0.10; moderate: ES ≥ 0.30; large: ES ≥ 1.2; very large: ES ≥ 2.0 66 .The p < 0.05 value was set to determine statistical significance.The statistical analysis was performed using the SPSS statistical package (v.25.0;SPSS Inc., IL, United States).The database of the present project can be found in Zenodo (Zenodo, Netherlands) in open access under https:// doi.org/ 10. 5281/ zenodo.79568 95.

Results
Table 2 shows the differences in dual career barriers; exercise barriers and athletic identity depending on type of disability.Student-athletes showed statistically significant differences in difficulty in reconciling sports and studies depending on type of disability (p = 0.001), with the Bonferroni adjustment showing that visually impaired student-athletes perceived statistically less difficulty than those with a hearing (Mean dif.: − 0.54 ± 0.19; p = 0.029; 95%ICC − 1.06; − 0.03) or physical (Mean dif.: − 0.59 ± 0.15; p = 0.001; 95%ICC − 1.01; − 0.18) impairment.
Statistically significant differences were also found in the perception that 'the cost of education is high' (p = 0.023) and 'I do not have enough university/educational institution support' (p = 0.014) as a barrier, with student-athletes with a hearing impairment showing higher scores as compared to those with a visual impairment in the first one (Mean dif.: − 0.92 ± 0.31; p = 0.022; 95%ICC − 1.76; − 0.08), and with student-athletes with cerebral palsy showing higher scores as compared to those with a visual impairment in the second (Mean dif.: − 0.97 ± 0.29; p = 0.008; 95%ICC − 1.77; − 0.18).
With regard to athletic identity according to the disability type, no statistically significant differences were found in the data analyzed.
Table 3 shows the differences in dual career barriers; exercise barriers and athletic identity depending on sports professionalization level.On the student-athletes' reported barriers to dual career success, significant differences were found in the barrier 'the university/educational institution is far from my training center' (p = 0.040), with semi-professional athletes showing statistically higher scores than professional athletes (Mean dif.: 0.58 ± 0.23; p = 0.037; 95%ICC 0.02; 1.15).Significant differences were also found in the barrier 'exercise takes too much time from family responsibilities' (p = 0.034), with professional athletes showing significantly higher values than amateur athletes (Mean dif.: − 0.45 ± 0.17; p = 0.033; 95%ICC − 0.87; − 0.02).
On the athletic identity scale, significant differences were found in the dimensions

Discussion
The first aim of this study was to analyze the differences in dual career interferences and barriers, exercise barriers, and athletic identity depending on type of disability.This is an issue that has not been addressed in previous research, although it could be very relevant, as it appears that student-athletes with disabilities may perceive Vol:.( 1234567890 www.nature.com/scientificreports/more barriers against achieving success in their dual career than their non-disabled peers 39 .However, previous research has not compared the difficulties encountered by student-athletes according to the type of disability, which so far prevents us from knowing which groups of disabled people would need a stronger support network to compensate for the barriers they face in the pursuit of dual career success.Moreover, given the multifactorial nature of the barriers they suffer 67 , it would be appropriate to know what type of barriers are affecting each population the most, in order to be able to carry out specific measures aimed at reducing their incidence in a concrete and efficient way.An important finding of this research is that there were differences in the perception of barriers depending on the type of disability.More specifically, this study found that visually impaired student-athletes perceived the least difficulties in reconciling studies and sports.This could be due to a growing awareness of the need to adapt the education system to maximize learning opportunities for students with visual impairment 68 , as well as the inclusion of administrators, curriculum planners, classmates and families, for a holistic policy for the visuallyimpaired learners 69 .The scientific literature has found that pedagogical strategies, learning tools, and external support, are the most effective strategies for the inclusion of students with visual impairment 70 .These facilitators coincide with those required for the implementation of an effective dual career model for student-athletes 71 .
At the other end of the spectrum, student-athletes with cerebral palsy showed the highest scores on the barrier 'I do not have enough university/educational institution support' .Previous studies have found that an inclusive school culture was crucial for students with cerebral palsy, because this group has been classically excluded from the educational system 72 , being sent to separate classes from their peers 73 , which has led to people with cerebral palsy perceiving a lack of institutional and home-based support to achieve success in education 74 .Indeed, in the case of students with cerebral palsy, this perception of dependency led students to perceive that their success in education was substantially impacted by the capacity of adults in the student's life to collaborate with others 72 .Therefore, knowing that student-athletes with disabilities are especially vulnerable in the area of education, it would be necessary to promote measures for the pursuit of success in this population, to create a sufficient support network, thus putting into practice the European regulations that promote the pursuit of equal opportunities for all subjects in the area of education 38,39,52 .
But the perception of high barriers of students with disabilities is not limited to the area of education.Studentathletes with cerebral palsy also showed the highest scores in the barrier 'Exercise facilities do not have convenient timetables for me' .This barrier has already been pointed out, together with the lack of sports offerings and the scarcity of opportunities, as the main obstacles faced by athletes with cerebral palsy 75 .In addition, for this group, there are other personal and environmental conditioning factors that affect sports practice 75 .In this sense, it has been found that belonging to a sports club is defined as a successful strategy to increase the participation in sports of these subjects, which also directly increases their athletic identity and their quality-of-life values 76 .This may be due to the fact that the people around them and socialization through sports practice are the main facilitators of sports practice for this group 77 .Furthermore, in the research by Cleary et al. 78 , it was evidenced that physical activity was promoted in this population when academic work and physical activity were understood as equally important priorities at school, a fact that can be considered as the starting point of the benefits expected in this group through the implementation of the dual career model in higher education.
Another relevant finding of this research was that student-athletes with hearing impairment encountered the most barriers related to the cost of education.Previous studies have pointed to the large economic impact of hearing loss on earning power, which may be largely due to the fact that the disability makes it difficult for disabled students to attend university, hindering their ability to find work and to have an average quality of life 79 .In this regard, research by Hogan et al. 80 highlighted how people with hearing loss were less likely to be in highskilled jobs, and were overrepresented among the low-income population.In this regard, while generic job skills are readily available in the higher education environment, other skills important for professional development and job search are more difficult for deaf students to acquire 81 .In addition to job opportunities, deaf students encounter a number of barriers in the educational environment that can increase the investment needed to  www.nature.com/scientificreports/ensure their inclusion, such as the lack of sign language interpreters, which is a key factor in the accessibility of education for the deaf 82,83 .Finally, physical impairment student-athletes obtained the highest scores in the barrier 'Exercise tires me' , suggesting that they may have a greater physiological response to exercise than able-bodied athletes 84 .In addition, fatigue is frequent in adults with physical disabilities 85 , which may be accentuated by the use of mobility aids, such as crutches or wheelchairs 86 , as well as by the limitations of transportation for this group, sometimes requiring more time for travel 87 .In this regard, greater fatigue could be related to a lower socio-economic level, due to the high cost of these implements 88 , as well as the increased logistics and costs of adapted transportation, especially affecting team sports 89 .
Regarding the type of disability and athletic identity, no statistically significant differences were detected in the present research, coinciding with the previous results from Pans et al. 90 .Regardless of whether the athlete has a disability or not, as well as the type of disability, he/she has had to give up other roles to prioritize the sporting one, a fact that is especially detected when the athlete reaches an elite competitive level (e.g., Olympic/ Paralympic), which leads to a reinforcement of the athletic identity 91 .Therefore, at these levels, there are previous studies that highlight that in adapted sports it is observed that athletes perceive themselves only as athletes and not as people with disabilities who are practicing a sport 92,93 .However, a strong athletic identity in disabled athletes can be an obstacle when facing sporting retirement, negatively affecting employability 92 .
In view of the above results, the hypothesis that there are differences in the perception of barriers to success in dual careers depending on the type of disability can be accepted.On the other hand, although it was not possible to write a clear hypothesis regarding differences in the perception as an athlete, depending on the type of disability, in the present research it was found that there are no differences in the perception as an athlete depending on the type of disability.
A second aim of the present research was to analyze differences in dual career interference and barriers, exercise barriers, and athletic identity, as a function of the level of professionalization of the sport.Differences were found in the perception of barriers by student-athletes according to their level of professionalization.More semi-professional student-athletes had higher scores in 'the university/educational institution is far from my training center' .Previous studies have already shown, in non-disabled student-athletes, that semiprofessional student-athletes may have higher scores on perceived barriers than their professional counterparts 94 .This could be due to the fact that semi-professional athletes are forced to make a more equal distribution of time between sports and studies, as they themselves in a situation where they do not know whether to prioritize their sporting or educational career 15,95 .Not in vain, in this study, it was also found that professional athletes spent the most hours on their sport to the detriment of study hours, while semi-professional athletes sought a better balance between these facets.Prior research has also highlighted that athletes with disabilities self-perceive, to a greater extent, the barriers related to the distance from their training center to the educational institution, as compared to their non-disabled peers 39 .In light of the results of the present research, this is an issue that should be especially taken into account in the case of semi-professional student-athletes.Thus, programs where accessible sports facilities are provided within the educational environment 96,97 , or where online attendance at school is allowed 38 , could be a solution for this population.
The present study also found that professional athletes pointed out the highest values in 'exercise takes too much time from family responsibilities' , showing differences in this item as compared to amateur athletes.This may be due to the fact that professional athletes have employment contracts related to their sporting performance, which oblige them to devote more time to sports, to the detriment of the time they dedicate to their studies 15,98 or other obligations, such as taking care of the family 15,95 .In this regard, the supports provided by family, friends, or the community, are essential when practicing sports, especially in the beginning, and for some more severe disabilities, throughout the sporting life 99 .The support from a family prepared for this challenge becomes fundamental for the development of the career of the student-athlete with a disability 100 .
Another important finding of this research was that professional athletes showed the highest values in most items related to their identity as athletes, followed by semi-professional athletes and amateur athletes.This suggests that sports play an important role for individuals with disabilities to stay healthy, build social relationships, increase independence, and achieve personal goals [101][102][103] , which increase together with the level of professionalization as athletes.Previous studies have pointed to a direct relationship between the level of professionalization of able-bodied student-athletes and the importance they grant to their role as athletes, prioritizing it over their status as students 15,98 .However, there have been documented cases of athletes who, after finishing their sporting career, have suffered identity crises that have negatively conditioned their post-sports development 104 .In this way, a greater commitment during the dual career can enrich the athlete's personality and avoid future identity confusion that may hinder their transition to their post-sporting life 105 .
In view of the results of the present research, the hypotheses about the differences in the perception of barriers and athletic identity of student-athletes with disabilities according to the level of professionalization are assumed to be correct.
In the light of the results of this study, student-athletes with disabilities need support measures, management, and policies 106 that create a network of political and institutional support to integrate their student life into their sports cycle, and to avoid difficult transitions to work after the end of the sporting career 107,37 .These must be adapted to their perceived barriers and generate specific measures depending on the type of disability and level of sports.However, as this is a pioneer study about dual career of athletes with disabilities, there are potential limitations, such as the heterogeneity of the sample, disabilities, career stages, sport/education careers, age, inclusion of athletes with disabilities independently from their acquired or congenital disabilities, and different models between para-sports organizations and educational institutions/public and private companies in the European Member States.With regard to future lines of research, it would be interesting to analyze whether www.nature.com/scientificreports/factors such as gender, the stage of their sporting career, and level of education, could affect the self-perception of dual career barriers and athletic identity of student-athletes with disabilities.

Table 1 .
Differences in socio-demographic; time distribution; studying and sporting characteristics depending on type of disability and level of sports professionalization. *p < 0.05; **p < 0.01; ***p < 0.001; -no significant differences.Education: Pre pre-university studies, UD university degree, Post post-university degree.Disability: H hearing, V visual, Ph physical, CP cerebral palsy.

Table 3 .
Differences in dual career barriers; exercise barriers and athletic identity depending on sports level of professionalization.