Abstract
The development process of female football players involves understanding the pathways and experiences acquired during childhood and adolescence. In Brazil, instruments designed to assess these early sport experiences are scarce. Therefore, the present study aimed to construct and gather validity evidence for the Questionnaire on Sports Experiences in Women’s Football (QEEF-Fem), a psychometric instrument grounded in the Personal Assets Framework (PAF), to evaluate the sport experimentation phase of young female athletes. The research was conducted in four stages: face validity, content validity, construct validity, and internal consistency. Initially, 45 items were developed and evaluated by athletes and experts. After two rounds of content validation, 38 items remained. For construct validation, 391 female athletes aged between 12 and 17 years (mean age 14.6 ± 1.8 years), representing 15 professional clubs from the southern region of Brazil, were recruited. Psychometric methods were employed to confirm and validate the QEEF-Fem, including Exploratory Factor Analysis (EFA) with polychoric correlations, Bartlett’s test of sphericity, and Kaiser–Meyer–Olkin measures. Internal consistency was assessed using Cronbach’s alpha, McDonald’s omega, and composite reliability (CR). Results indicated satisfactory factor loadings across the three dimensions of the PAF: Personal Engagement in Activities, Quality Social Dynamics, and Appropriate Settings. Reliability indices reached acceptable thresholds for exploratory studies, confirming the stability of the instrument. Internal consistency indices ranged from Cronbach’s α = 0.561 − 0.669, McDonald’s ω = 0.554 − 0.618, and CR = 0.619 − 0.714, indicating borderline to acceptable reliability for exploratory research purposes. All analyses were conducted using JASP (0.95.2.0). It was concluded that the QEEF-Fem provides valid and reliable evidence to assess early sport experiences in female football, reflecting the interaction between personal, contextual, and relational dimensions of athlete development. The QEEF-Fem represents a culturally adapted tool that can be used to investigate developmental trajectories of young female football players in Brazil.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-026-45712-w.
Keywords: Youth sport development, Talent development environments, Sport participation pathways
Subject terms: Health care, Psychology, Psychology
Introduction
Football is among the most widely practiced sports globally and occupies a central position in youth development systems across cultures. In recent decades, research has increasingly examined the environments and experiences that shape athletes’ developmental trajectories. Within this broader landscape, women’s football has expanded rapidly in participation and visibility, accompanied by growing scientific attention1. Understanding player development requires examining the pathways and experiences accumulated during childhood and adolescence2,3, particularly how multiple contexts interact to shape both athletic and personal growth. Contemporary models of youth sport development emphasize diversified and interconnected experiences in the early years, integrating developmental phases rather than prioritizing early specialization4,5.
The Personal Assets Framework (PAF)4 conceptualizes long-term development through three interrelated components: appropriate settings, referring to the physical and competitive environment; personal engagement in activities, encompassing diversified sport participation; and quality social dynamics, reflecting relationships with coaches, peers, parents, and other stakeholders. The interaction of these components is proposed to foster adaptive developmental outcomes6. During childhood and adolescence, deliberate play (informal, intrinsically motivated activity) and deliberate practice (structured training) jointly shape skill acquisition and psychosocial development. The model posits that deliberate play should predominate during the sport experimentation phase, facilitating autonomy, enjoyment, and broad motor and cognitive development6. Early diversified engagement has also been linked to enhanced decision-making, motor coordination, and tactical learning7.
However, empirical findings remain mixed. While some evidence suggests greater diversification in girls’ early sport participation3,8, studies of elite performers indicate higher volumes of early sport-specific practice among those reaching professional levels9. These tensions underscore the need for context-sensitive investigation of developmental pathways.
In Brazil, youth development in women’s football faces structural constraints, including limited infrastructure, inconsistent youth programming, and persistent social barriers to female participation10. Evidence from Brazilian handball, analyzed through the PAF lens, suggests that strong social dynamics may partially compensate for precarious environmental conditions during early development (Lima et al., 2022)24, highlighting the interdependence between relational and contextual factors in sustaining engagement.
Despite the growth of women’s football, systematic analyses of the developmental environments of Brazilian female players remain limited. Existing assessment tools have largely been developed in male populations or within sociocultural contexts that differ substantially from Brazilian women’s football. Consequently, gender-specific and contextual dimensions of development may be insufficiently captured. Similar limitations have been reported internationally; even in England, female talent development environments remain comparatively underexamined, and stakeholder perspectives are often incompletely integrated2.
The Participation History Questionnaire (PHQ)11 illustrates this limitation. Developed and validated in male, English-speaking samples, it reflects assumptions that may not align with the structural and cultural realities of Brazilian women’s football. Such instruments risk overlooking the specific constraints, opportunities, and social configurations shaping female developmental pathways. There remains a need for culturally grounded, gender-sensitive measures capable of capturing early sport experiences within women’s football.
Accordingly, this study aimed to develop and gather validity evidence for the Questionnaire on Sports Experiences in Women’s Football (QEEF-Fem), grounded in the PAF. The instrument was designed to assess three central dimensions: personal engagement, quality social dynamics, and appropriate settings, among Brazilian female players aged 12–17 years. We hypothesized that the QEEF-Fem would demonstrate a theoretically coherent factorial structure consistent with the PAF and adequate internal consistency for exploratory research. By providing a context-sensitive psychometric tool, this study seeks to advance the empirical assessment of the sport experimentation phase and inform evidence-based practice in women’s football.
Methods
Study design
This study employed a psychometric design to develop and gather validity evidence for a questionnaire grounded in the Personal Assets Framework (PAF)4. Although participants were aged 12–17 years at data collection, the instrument was designed as a retrospective measure to capture perceptions of experiences during the sport experimentation phase. The questionnaire uses Likert-type response scales assessing agreement, satisfaction, or frequency across three PAF dimensions: Personal Engagement (diversified sport participation), Quality Social Dynamics (relationships with coaches, peers, parents, and administrators), and Appropriate Settings (physical and competitive characteristics of the practice environment).
Validation procedures followed the Standards for Educational and Psychological Testing12. Consistent with contemporary recommendations for construct validation13, the process integrated theoretical (content validity), empirical (face validity), and statistical (construct validity and internal consistency) evidence.
Item development was guided by an analytical matrix derived from the PAF. An initial pool of 45 items was generated (Personal Engagement: 11; Quality Social Dynamics: 15; Appropriate Settings: 19). Higher scores indicate more positive developmental experiences, and no items required reverse coding. Validation proceeded in four stages: I) face validity to assess clarity and comprehensibility; II) content validity through expert evaluation of linguistic and theoretical adequacy; III) construct validity via exploratory factor analysis (EFA); and IV) internal consistency assessed using Cronbach’s alpha, composite reliability (CR), and McDonald’s omega.
Face validity
Face validity assessed whether the appearance of the questionnaire is understandable and appropriate for its users. However, the literature recommends that this step be accompanied by content validity14. For this validation stage, the questionnaire was administered to a pilot group of 45 young female players from a football club in the state of Paraná who are part of the youth categories of a professional club, aged between 12 and 17 years. Higher scores are considered positive.
Content validity
Content validity is understood as fundamental for assessing the degree of representativeness of the instrument in relation to the theoretical constructs it proposes to measure15. This stage was conducted with the participation of experts selected for convenience, similar to that carried out in the study by Tozetto et al.16, based on the following criteria: a) Hold a PhD in Physical Education,b) Work as a university professor; c) Have publications on youth training in sports; d) Have publications focused on youth training and development in football; and e) Have previous experience with other validation processes.
The experts were contacted by email, accompanied by an invitation and a brief description of the research. Thus, ten experts (n = 10) participated in this stage. Responses to the items were obtained on a five-point Likert scale, ranging from 1 to 5 (1 = not relevant; 5 = relevant). Based on the 45 questions initially developed, the Content Validity Index (CVI) was used, considering questions above 0.80 to be acceptable17.
Construct validity
Construct validity seeks to establish a correspondence between the proposed theoretical structure and the empirical hypothesis investigated18. Thus, in the present study, construct validity was examined using data from 391 female athletes, allowing the evaluation of the relationship between the theoretical constructs of the Personal Assets Framework and the proposed measurement model. To this end, as it is a new instrument adapted to a specific culture, EFA was employed. Thus, we sought to verify the consistency between the theoretical basis and the empirical evidence obtained, consolidating the evidence of construct validity13.
Internal consistency of items
Internal consistency was assessed based on the stability of responses, following recommendations (Hayes and Coutts, 2020)26. The objective was to verify whether the set of responses obtained reflects the same underlying theoretical constructs (Hayes and Coutts, 2020). Thus, Cronbach’s alpha was used to measure item homogeneity, CR to assess latent construct consistency, and McDonald’s omega to provide a more accurate reliability estimate based on factor loadings. The results range from 0 to 1, with values closer to 1 indicating greater internal consistency among the items18.
Data collection procedures
The athletes answered the questionnaire in person and collectively during visits to the clubs. The following inclusion criteria were adopted: I) being between 12 and 17 years old (considering the sports specialization phase); II) athletes linked to football teams in the states of Paraná, Santa Catarina, and Rio Grande do Sul that competed in competitions organized by the respective state federations. Thus, the data collection period took place between June and September 2025, with the aim of applying the questionnaire and understanding the contexts of women’s football practice. The sample included 391 athletes representing 15 professional football clubs. The clubs were selected by convenience sampling, based on accessibility and institutional availability to participate. This depended on formal authorization and logistical feasibility, resulting in the inclusion of clubs that were accessible at the time of data collection.
The sample size was determined based on widely adopted psychometric guidelines for EFA, which recommend a minimum ratio of 10 participants per item to ensure stable factor solutions19. Considering that the initial version of the instrument comprised 38 items, the final sample of 391 athletes exceeded this recommendation. This sample size was therefore considered adequate for exploratory purposes and allowed for potential data loss due to incomplete responses or exclusion criteria.
Ethical aspects of research
This project was submitted to the Human Research Ethics Committee (CEPSH) of the Federal University of Santa Catarina and received a favorable opinion (Process No. 7,517,417). All ethical procedures were followed, with the signing of the Free and Informed Consent Form (FICF) by those responsible and the Free and Informed Assent Form (FIAF) by the participants. Since the participants were minors, written informed consent was obtained from their parents or legal guardians, and assent was obtained from all participants before data collection. Therefore, all procedures performed in this study were conducted in accordance with the relevant CEPSH guidelines and regulations.
Data analysis
The data were organized in a Microsoft Excel spreadsheet and analyzed using JASP 0.95.2.0 software. Face validity was assessed using Aiken’s V coefficient20. Content validity was assessed using the content validity index (CVI), considering items with a score of 4 or 5 by at least 80% of the experts15. The construct validity stage was examined using EFA, considering adjustment estimators with three factors, using the Weighted Least Squares (WLS) factor method with oblique rotation and a polychoric correlation matrix, in order to evaluate the factor structure of the instrument21. Items were excluded based on the following criteria: (1 factor loadings < 0.30, (2 cross loadings between factors; and (3 the balance between statistical stability and theoretical relevance. The model was verified using Kaiser–Meyer–Olkin (KMO) ≥ 0.600 and Bartlett’s test (p < 0.001), indicating sampling adequacy and matrix factorability. Additionally, all items presented Skewness and Kurtosis values within the established limits (Skewness between − 3 and + 3 and Kurtosis between − 10 and + 10), supporting the adequacy of the data distribution for factor analysis18. Thus, Cronbach’s alpha (α), McDonald’s omega (ω), and CR were used to assess data reliability, adopting values ≥ 0.600 as acceptable for exploratory studies22 (Fig. 1).
Fig. 1.
Process of constructing and gathering validity evidence.
Results
Face validity
The process of obtaining evidence of the instrument’s validity began with the face validation stage, based on the application of the questionnaire to 45 football players aged between 12 and 17 years, members of the youth categories of a professional women’s football team. Table 1 describes the results obtained in the face validity stage, considering the responses on a Likert scale from 1 to 5, and presents the mean values, standard deviation, and Aiken’s V coefficient. Considering the responses obtained in the seven items evaluated, a positive assessment of the participating athletes was observed.
Table 1.
Face Validity of the questionnaire, considering a scale from 1 to 5.
| Mean (SD) | CV (%) | Aiken’s V | Qualitative | |
|---|---|---|---|---|
| Item 1—Clear and understandable questions | 4.29 (0.78) | 18.18 | 0.82 | Very high |
| Item 2—Evaluation of experiences | 4.42 (0.71) | 16.06 | 0.85 | Very high |
| Item 3—Difficulty in understanding | 1.89 (1.18) | 62.43 | 0.78 | High |
| Item 4—Confusion or ambiguity | 1.91 (1.26) | 65.97 | 0.77 | High |
| Item 5—Appropriate language | 4.31 (1.05) | 24.36 | 0.82 | Very high |
| Item 6—Enjoyed answering | 4.62 (0.85) | 18.40 | 0.90 | Very high |
| Item 7—Willingness to answer again | 4.42 (1.27) | 28.73 | 0.85 | Very high |
1 = Do you think the questions in the questionnaire are clear and understandable? 2 = Do you think the questions assess the sporting experiences you have had in football and futsal? 3 = Did you find it difficult to understand the questionnaire? 4 = Did you find any questions confusing or ambiguous? 5 = Do the questions use language appropriate for your age? 6 = Did you enjoy answering the questionnaire? 7 = Would you answer this questionnaire again? With responses considered on a scale of 1 = not relevant; 5 = relevant. Responses were organized on a Likert scale ranging from 1 to 5 (1 = not relevant; 5 = relevant). SD = standard deviation; CV = coefficient of variation.
Content validity
Content validity was assessed by 10 PhD in Physical Education. The CVI was used, considering questions above 0.80 to be acceptable. Thus, of the 45 questions previously prepared, 3 questions had an average CVI below 0.80 and were removed (average CVI = 0.73) in the first round of assessment. Another 15 questions underwent a new round of validation after textual adjustments suggested by the experts (average CVI = 0.86). In the second round of evaluation, the experts suggested the exclusion of 4 questions (average CVI = 0.73), leaving the questionnaire with 38 questions (Table 2).
Table 2.
CVI Assigned by experts during the content validity stage.
| Item | 1st Round | 2nd Round | ||||||
|---|---|---|---|---|---|---|---|---|
| LC | PP | TR | CVImean | LC | PP | TR | CVImean | |
| Q1 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q2 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q3 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q4 | 0.78 | 0.67 | 0.67 | 0.71* | X | X | X | – |
| Q5 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q6 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q7 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q8 | 0.78 | 1.00 | 1.00 | 0.93 | – | – | – | – |
| Q9 | 0.78 | 1.00 | 1.00 | 0.93 | – | – | – | – |
| Q10 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q11 | 0.67 | 1.00 | 1.00 | 0.89 | 1.00 | 0.90 | 0.90 | 0.93 |
| Q12 | 0.89 | 1.00 | 0.89 | 0.93 | – | – | – | – |
| Q13 | 0.67 | 0.89 | 0.89 | 0.82 | 0.60 | 0.80 | 0.80 | 0.73* |
| Q14 | 0.78 | 0.89 | 0.78 | 0.82 | 0.90 | 1.00 | 1.00 | 0.97 |
| Q15 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q16 | 0.67 | 1.00 | 1.00 | 0.89 | 0.90 | 0.90 | 0.90 | 0.90 |
| Q17 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q18 | 0.89 | 0.78 | 0.89 | 0.85 | 1.00 | 0.90 | 0.90 | 0.93 |
| Q19 | 0.78 | 0.89 | 0.89 | 0.85 | 0.90 | 0.90 | 0.90 | 0.90 |
| Q20 | 0.67 | 1.00 | 1.00 | 0.89 | 0.90 | 1.00 | 1.00 | 0.97 |
| Q21 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q22 | 0.89 | 0.89 | 0.89 | 0.89 | – | – | – | – |
| Q23 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q24 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q25 | 0.78 | 0.89 | 0.89 | 0.85 | 0.90 | 0.90 | 1.00 | 0.93 |
| Q26 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q27 | 0.89 | 0.89 | 1.00 | 0.93 | – | – | – | – |
| Q28 | 0.78 | 1.00 | 0.89 | 0.89 | 0.90 | 1.00 | 1.00 | 0.97 |
| Q29 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q30 | 0.89 | 1.00 | 0.89 | 0.93 | – | – | – | – |
| Q31 | 0.78 | 0.89 | 0.89 | 0.85 | 0.80 | 0.70 | 0.70 | 0.73* |
| Q32 | 0.67 | 0.89 | 0.89 | 0.82 | 0.80 | 0.90 | 0.80 | 0.83 |
| Q33 | 0.78 | 0.89 | 0.89 | 0.85 | 1.00 | 0.90 | 0.90 | 0.93 |
| Q34 | 0.78 | 1.00 | 1.00 | 0.93 | 0.9–0 | 1.00 | 1.00 | 0.97 |
| Q35 | 0.78 | 0.89 | 0.89 | 0.85 | 0.50 | 0.70 | 0.70 | 0.63* |
| Q36 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q37 | 0.44 | 0.89 | 0.78 | 0.70* | X | X | X | – |
| Q38 | 0.89 | 0.89 | 1.00 | 0.93 | – | – | – | – |
| Q39 | 0.56 | 1.00 | 0.78 | 0.78* | X | X | X | – |
| Q40 | 0.89 | 1.00 | 1.00 | 0.96 | – | – | – | – |
| Q41 | 0.78 | 0.89 | 0.78 | 0.82 | 0.60 | 0.70 | 0.70 | 0.67* |
| Q42 | 0.89 | 0.89 | 0.78 | 0.85 | 0.80 | 0.90 | 0.90 | 0.87 |
| Q43 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
| Q44 | 0.78 | 1.00 | 1.00 | 0.93 | 1.00 | 1.00 | 1.00 | 1.00 |
| Q45 | 1.00 | 1.00 | 1.00 | 1.00 | – | – | – | – |
LC = Language Clarity; PP = Practical Pertinence; TR = Theoretical Relevance; X = Eliminated Questions; * = CVI < 0.80.
Construct validity
The Shapiro–Wilk test was applied to the 38 items included in the EFA, confirming the asymmetry between the data. All items presented Skewness and Kurtosis within the established limits. The coefficients of variation (CV) indicated a predominance of low values among the items, revealing greater homogeneity in the responses. Some items presented a higher CV, evidencing greater heterogeneity. The data indicated are represented in Table 3.
Table 3.
Description of values by mean, standard deviation, coefficient of variation, variation, and Shapiro–Wilk test (p < 0.05).
| Dimensions | Item | Mean ± SD | CV (%) | Variance | Skewness | Kurtosis | Shapiro–Wilk* |
|---|---|---|---|---|---|---|---|
| Personal Engagement in Activities | Q1 | 4.62 ± 0.87 | 18.83 | 0.758 | − 2.679 | 6.923 | 0.497 |
| Q2 | 4.35 ± 0.95 | 21.84 | 0.907 | − 1.540 | 1.814 | 0.703 | |
| Q3 | 4.54 ± 0.93 | 20.48 | 0.873 | − 2.299 | 4.803 | 0.556 | |
| Q4 | 4.14 ± 1.14 | 27.54 | 1.300 | − 1.148 | 0.161 | 0.746 | |
| Q5 | 3.70 ± 1.23 | 33.24 | 1.531 | − 0.580 | − 0.773 | 0.858 | |
| Q6 | 3.67 ± 1.36 | 37.06 | 1.849 | − 0.718 | − 0.714 | 0.833 | |
| Q7 | 4.22 ± 1.19 | 28.20 | 1.429 | − 1.522 | 1.187 | 0.680 | |
| Q8 | 4.23 ± 1.15 | 27.20 | 1.341 | − 1.425 | 0.911 | 0.695 | |
| Q9 | 2.45 ± 1.25 | 51.02 | 1.573 | 0.666 | − 0.512 | 0.867 | |
| Quality Social Dynamics | Q10 | 4.60 ± 0.83 | 18.04 | 0.692 | − 2.115 | 3.500 | 0.536 |
| Q11 | 4.67 ± 0.82 | 17.56 | 0.684 | − 2.816 | 7.646 | 0.454 | |
| Q12 | 4.18 ± 1.33 | 31.82 | 1.791 | − 1.467 | 0.708 | 0.644 | |
| Q13 | 4.20 ± 1.15 | 27.38 | 1.337 | − 1.306 | 0.567 | 0.712 | |
| Q14 | 4.27 ± 0.84 | 19.67 | 0.715 | − 1.156 | 1.230 | 0.776 | |
| Q15 | 4.23 ± 1.03 | 24.35 | 1.078 | − 1.347 | 1.051 | 0.739 | |
| Q16 | 4.54 ± 0.82 | 18.07 | 0.677 | − 2.018 | 3.967 | 0.609 | |
| Q17 | 3.35 ± 1.25 | 37.31 | 1.574 | − 0.247 | − 1.055 | 0.896 | |
| Q18 | 4.11 ± 0.92 | 22.38 | 0.855 | − 0.962 | 0.608 | 0.816 | |
| Q19 | 3.96 ± 1.02 | 25.76 | 1.040 | − 0.836 | 0.093 | 0.841 | |
| Q20 | 4.14 ± 0.93 | 22.46 | 0.866 | − 1.098 | 1.061 | 0.803 | |
| Q21 | 3.57 ± 1.16 | 32.49 | 1.350 | − 0.547 | − 0.523 | 0.886 | |
| Q22 | 3.20 ± 1.50 | 46.88 | 2.250 | − 0.221 | − 1.376 | 0.861 | |
| Q23 | 3.49 ± 1.37 | 39.26 | 1.878 | − 0.594 | − 0.900 | 0.853 | |
| Appropriate Settings | Q24 | 3.92 ± 1.18 | 30.10 | 1.408 | − 0.934 | − 0.051 | 0.817 |
| Q25 | 4.06 ± 0.97 | 23.89 | 0.952 | − 0.957 | 0.408 | 0.820 | |
| Q26 | 4.27 ± 1.12 | 26.23 | 1.265 | − 1.556 | 1.421 | 0.682 | |
| Q27 | 3.51 ± 1.24 | 35.33 | 1.558 | − 0.378 | − 0.966 | 0.884 | |
| Q28 | 4.36 ± 1.02 | 23.39 | 1.054 | − 1.755 | 2.469 | 0.666 | |
| Q29 | 4.17 ± 0.93 | 22.29 | 0.880 | − 1.116 | 0.756 | 0.790 | |
| Q30 | 3.97 ± 1.06 | 26.70 | 1.124 | − 0.906 | 0.114 | 0.830 | |
| Q31 | 3.60 ± 1.39 | 38.61 | 1.951 | − 0.632 | − 0.896 | 0.838 | |
| Q32 | 4.17 ± 0.88 | 21.10 | 0.782 | − 1.006 | 0.817 | 0.805 | |
| Q33 | 3.32 ± 1.28 | 38.55 | 1.654 | − 0.194 | − 1.016 | 0.896 | |
| Q34 | 3.76 ± 1.11 | 29.52 | 1.253 | − 0.593 | − 0.417 | 0.867 | |
| Q35 | 3.79 ± 1.10 | 29.02 | 1.228 | − 0.728 | − 0.171 | 0.862 | |
| Q36 | 3.09 ± 1.18 | 38.16 | 1.406 | 0.054 | − 0.961 | 0.910 | |
| Q37 | 3.53 ± 1.14 | 32.29 | 1.306 | − 0.275 | − 0.978 | 0.889 | |
| Q38 | 1.94 ± 1.22 | 62.89 | 1.495 | 1.150 | 0.171 | 0.758 |
SD = standard deviation; CV = coefficient of variation; * Shapiro–Wilk test (p < 0.05).
The Bartlett Sphericity test (X2 = 2480.715; df = 153, p < 0.001) and the Kaiser–Meyer–Olkin (KMO) = 0.702 indicated adequacy and interpretability of the correlation matrix of the items. The correlation between the factors in the final model was: factor 1 and factor 2: 0.128; factor 1 and factor 3: 0.384; factor 2 and factor 3: 0.221. Thus, the AFE was constructed based on the three scales recommended by the theoretical structure of the model (Personal Engagement; Quality social dynamics; Context of Practice). Thus, 17 questions had factor loadings above 0.300 are presented at Table 4.
Table 4.
Factor structure of the questionnaire.
| Nº | Questions | Personal engagement in activities | Quality social dynamics | Appropriate settings |
|---|---|---|---|---|
| Q2 | I enjoyed playing football/futsal with other children, without the presence of teachers | 0.364* | − 0.024 | 0.059 |
| Q5 | I was very interested in practicing other sports besides football/futsal | 0.411* | 0.079 | − 0.194 |
| Q6 | Practicing other sports made me more interested in football/futsal | 0.446* | 0.210 | − 0.200 |
| Q7 | I played football with my friends in the street | 0.651* | − 0.092 | 0.137 |
| Q8 | I tried various games and play activities in the street | 0.663* | − 0.080 | 0.115 |
| Q10 | My parents supported me in playing football/futsal | 0.138 | 0.720* | 0.116 |
| Q11 | My parents supported me in practicing sports | − 0.108 | 0.805* | 0.041 |
| Q12 | My siblings supported me in playing football/futsal | 0.178 | 0.587* | − 0.017 |
| Q15 | My Physical Education teachers supported me in practicing sports | 0.239 | 0.497* | 0.035 |
| Q16 | I had a great relationship with my Physical Education teachers at school | 0.308 | 0.429* | − 0.105 |
| Q21 | We had a good relationship with referees during games | 0.294 | 0.311* | 0.066 |
| Q22 | The championships I participated in adapted the field size and goalposts to my age | 0.112 | 0.304* | 0.001 |
| Q23 | The championships I participated in adapted the game rules to my age | 0.254 | 0.333* | 0.039 |
| Q29 | The school sports environments were welcoming | − 0.012 | 0.057 | 0.686* |
| Q30 | The clubs I participated in provided a safe and well-structured sports environment, with new balls, uniforms, water, and locker rooms | 0.146 | 0.069 | 0.507 |
| Q32 | The environments of social projects focused on football/futsal were welcoming | 0.054 | 0.044 | 0.625 |
| Q36 | Women’s football/futsal was valued in my city | 0.137 | 0.122 | 0.312 |
| – | Cronbach’s Alpha | 0.561 | 0.669 | 0.623 |
| – | Composite Reliability | 0.638 | 0.734 | 0.619 |
| – | Mcdonald’s Omega | 0.579 | 0.554 | 0.618 |
*Indicates factor loadings meeting the minimum cutoff value (≥ 0.300).
Significant values are in bold.
Internal consistency
The internal consistency of the unifactorial model was assessed using Cronbach’s alpha (α = 0.745) and McDonald’s omega (ω = 0.705), indicating adequate to good reliability within an exploratory context. For the trifactorial model, Cronbach’s alpha values ranged from 0.561 to 0.669, while McDonald’s omega ranged from 0.554 to 0.618, reflecting borderline to acceptable internal consistency. CR values ranged from 0.619 to 0.714, which are considered adequate for exploratory studies. Overall, these results suggest that the instrument demonstrates reliability levels appropriate for exploratory research, while highlighting the need for further confirmatory analyses in future studies.
Discussion
This study developed and provided initial validity evidence for a psychometric instrument designed to assess the sport experimentation phase of young female football players within the framework of the PAF. Across sequential validation procedures: face, content, construct validity, and internal consistency the resulting 17-item instrument demonstrated acceptable psychometric properties for exploratory use.
Face validity analyses indicated strong agreement regarding clarity and relevance, supporting the instrument’s comprehensibility among athletes aged 12–17 years. High Aiken’s V coefficients suggest that the items were interpreted consistently, reinforcing their practical adequacy. These findings align with previous validation research indicating that perceived clarity enhances respondent engagement and data quality14.
Exploratory factor analysis supported a three-factor structure consistent with the theoretical dimensions of the PAF. Sphericity and sampling adequacy indices confirmed the suitability of the data for factor analysis. Although item Q16 presented cross-loadings, its retention was theoretically justified, and its removal did not substantially improve model fit or reliability estimates. Nonetheless, this item warrants re-examination in future confirmatory analyses. Overall, the factorial solution parallels structures reported in related validation studies (e.g.,23), lending further support to the coherence of the proposed model.
Internal consistency indices were within acceptable thresholds for exploratory research. While some coefficients were modest (particularly in factors with fewer items) such values are consistent with early-stage scale development and with the influence of item quantity on reliability estimates22. Importantly, reliability estimates did not contradict the theoretical structure, suggesting that the dimensions capture related but distinguishable aspects of developmental experience.
Conceptually, the instrument operationalizes the core mechanisms proposed by the PAF (i.e. personal engagement, quality social dynamics, and appropriate settings) within the specific context of Brazilian women’s football. Prior qualitative work has supported the relevance of the PAF in Brazilian athlete development24, the present findings extend this evidence by providing a structured quantitative measure aligned with the model.
These results also contribute to broader discussions regarding gender-sensitive and culturally grounded assessment in sport. International research has highlighted the relative underrepresentation of female talent development environments2 and the central role of communication and social support in shaping female athletes’ trajectories (Gledhill and Harwood, 2019,25 Hendry et al., 2019)27. Within Brazil, structural constraints and sociocultural barriers further underscore the need for contextually responsive instruments10. The QEEF-Fem addresses this gap by integrating international theoretical frameworks with local developmental realities.
Practically, the instrument may assist practitioners in monitoring perceived developmental experiences, including the balance between structured practice and diversified engagement, as well as relational and environmental conditions within youth programs. However, interpretations should remain cautious given the exploratory design and region-specific sample.
Several limitations warrant consideration. The absence of temporal stability assessment (test–retest) precludes conclusions regarding response consistency over time. Additionally, the sample was restricted to clubs from southern Brazil, limiting generalizability. The relatively small number of items per factor may also constrain content breadth and partially explain moderate reliability estimates. Future research should expand the item pool, conduct confirmatory factor analyses in independent samples, test measurement invariance, assess temporal reliability, and explore item response theory approaches.
Within these boundaries, the QEEF-Fem represents an initial step toward a context-sensitive assessment framework for women’s football. By providing structured evidence aligned with developmental theory, it contributes to strengthening the empirical basis of athlete development research in Brazil.
Conclusion
This study developed and compiled different validity evidence for the Questionnaire on Sports Experiences in Women’s Football (QEEF-Fem), providing robust evidence of its reliability and validity. The results confirm that the QEEF-Fem is a culturally adapted and scientifically rigorous tool, capable of generating accurate information about the developmental trajectories of female athletes. In addition to its methodological contribution, the questionnaire offers practical applications for coaches, pedagogical coordinators, sports clubs, and public policy makers, supporting the planning of training programs and development strategies sensitive to the realities of women’s football in Brazil. The QEEF-Fem represents a groundbreaking advancement in women’s soccer research and practice, strengthening the scientific basis for athlete development and contributing to more inclusive and evidence-based sports policies.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We acknowledge the contribution of all participants in this study. This study was financed in part by the Secretaria Nacional do Futebol e Defesa dos Direitos do Torcedor (SNFDT), Ministério do Esporte—Brazil (Process Number 71000.052809/2024-41), to which we would like to express our thanks. In addition, we would like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the grants awarded to JCSB.
Author contributions
JCCB and PHB wrote the main manuscript; LF, JVN and MM reviewed the entire manuscript; WT and PHB did statistical analysis; PHB supervised the research project.
Funding
This study was financed in part by the Secretaria Nacional do Futebol e Defesa dos Direitos do Torcedor (SNFDT), Ministério do Esporte—Brazil (Process Number 30879720240008-003016), to which we would like to express our thanks. In addition, we would like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the grants awarded to JCSB.
Data availability
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Barreira, J., Lemes, R., and Galatti, L. R. (2022). Trajectories and professional skills of high-level women’s football managers in Brazil. In Women’s Football in Latin America: Social Challenges and Historical Perspectives Vol. 1, 145–162 (Springer International Publishing, Cham, 2022).
- 2.Emmonds, S., Gledhill, A., Mitchell, T., Ivarsson, A. & Sargent-Megicks, B. Expanding our understanding of talent development environments in English female youth football: Player, parent and coach perspectives. J. Sports Sci.10.1080/02640414.2025.2604931 (2026). [DOI] [PubMed] [Google Scholar]
- 3.Machado, G., González-Víllora, S. & Teoldo, I. Contribution of deliberate practice, play, and futsal to the acquisition of decision-making skills in Brazilian professional female football players. Int. J. Sport Exerc. Psychol.22(3), 756–774 (2022). [Google Scholar]
- 4.Côté, J., Turnnidge, J., & Vierimaa, M. A personal assets approach to youth sport. In Routledge handbook of youth sport 243–255 (Routledge, 2016).
- 5.Zhao, J. et al. Differences and relationships between talent detection, identification, development and selection in sport: A systematic review. Heliyon10.1016/j.heliyon.2024.e27543 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Côté, J., Turnnidge, J., Murata, A., McGuire, C. & Martin, L. Youth sport research: Describing the integrated dynamics elements of the Personal Assets Framework. Int. J. Sport Psychol.51(6), 562–578 (2020). [Google Scholar]
- 7.Machado, G., González-Víllora, S., Pastor-Vicedo, J. C. & Teoldo, I. Mapping talent pathways: A comparative study of developmental activities and practice structure in Brazilian and Spanish U-18 elite youth male soccer players. Int. J. Sports Sci. Coach.19(5), 2006–2015 (2024). [Google Scholar]
- 8.Ford, P. R., Hodges, N. J., Broadbent, D., O’Connor, D., Scott, D., Datson, N., Andersson, H. A. and Williams, A. M. The developmental and professional activities of female international Football players from five high-performing nations. In Science and Football 236–244 (Routledge, 2023). [DOI] [PubMed]
- 9.Andrew, M. et al. Talent development in female football: Developmental activities of professional players in England. J. Sports Sci.42(10), 853–864 (2024). [DOI] [PubMed] [Google Scholar]
- 10.Machado, G., Gonzalez-Villora, S., & Teoldo, I. (2025). Two genders, different roads? Comparing talent development pathways between female and male elite football players in Brazil. Int. J. Sports Sci. Coach. 17479541251377537.
- 11.Ford, P. R., Yates, I. & Williams, A. M. An analysis of practice activities and instructional behaviours used by youth football coaches during practice: Exploring the link between science and application. J. Sports Sci.28(5), 483–495 (2010). [DOI] [PubMed] [Google Scholar]
- 12.American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. In Standards for educational and psychological testing (American Educational Research Association, 2014).
- 13.Lambert, L. S. & Newman, D. A. Construct development and validation in three practical steps: Recommendations for reviewers, editors, and authors. Organ. Res. Methods.26(4), 574–607 (2023). [Google Scholar]
- 14.Masuwai, A., Zulkifli, H. & Hamzah, M. I. Evaluation of content validity and face validity of secondary school Islamic education teacher self-assessment instrument. Cogent Educ.11(1), 2308410 (2024). [Google Scholar]
- 15.Roebianto, A. D. I. Y. O., Savitri, S. I., Aulia, I. R. F. A. N., Suciyana, A. R. I. E. & Mubarokah, L. Content validity: Definition and procedure of content validation in psychological research. Test., Psychom., Methodol. Appl. Psychol. (TPM)30(1), 5–18 (2023). [Google Scholar]
- 16.Tozetto, W. R., Milan, F. J., Lima, C. O. V. & Milistetd, M. Sport mental health continuum in young Brazilian athletes: Adaptation and psychometric properties. Sci. Rep.14(1), 23194 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Madadizadeh, F. & Bahariniya, S. Tutorial on how to calculating content validity of scales in medical research. Perioper. Care Oper. Room Manag.31, 100315 (2023). [Google Scholar]
- 18.Kline, R. B. Principles and practice of structural equation modeling (Guilford publications, 2023). [Google Scholar]
- 19.Hair, J. F., Black, W. C., Babin, B. J. & Anderson, R. E. Multivariate data analysis 8th edn. (Cengage Learning, 2019). [Google Scholar]
- 20.Aiken, L. R. Content validity and reliability of single items or questionnaires. Educ. Psychol. Meas.40(4), 955–959 (1980). [Google Scholar]
- 21.Lorenzo-Seva, U. & Ferrando, P. J. A simulation-based scaled test statistic for assessing model-data fit in least-squares unrestricted factor-analysis solutions. Methodology19, 96–115. 10.5964/meth.9839 (2023). [Google Scholar]
- 22.Hair, J. F. Jr. et al. Partial least squares structural equation modeling (PLS-SEM) using R: A workbook 197 (Springer Nature, 2021). [Google Scholar]
- 23.Rechenchosky, L. et al. Adaptation and validation of a test for the evaluation of tactical knowledge in football: Test De Conocimiento Táctico Ofensivo en Fútbol for the Brazilian context (TCTOF-BRA). Front. Psychol.13, 849255 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lima, L. A. D., Reverdito, R. S., Scaglia, A. J. & Galatti, L. R. Engagement in athletic career: A study of female Brazilian handball world champions. Int. J. Sports Sci. Coach.18(4), 1056–1066 (2023). [Google Scholar]
- 25.Gledhill, A., & Harwood, C.Toward an understanding of players’ perceptions of talent developmentenvironments in UK female football. Journal of applied sport psychology, 31(1), 105-115(2019).
- 26.Hayes, A. F., & Coutts, J. J. Use omega rather than Cronbach’s alpha for estimating reliability. But….Communication methods and measures, 14(1), 1-24(2020).
- 27.Hendry, D. T., Williams, A. M., Ford, P. R., & Hodges, N. J. Developmental activities and perceptions ofchallenge for national and varsity women soccer players in Canada. Psychology of Sport and Exercise, 43, 210-218(2019).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.