Physical literacy (PL) is a concept that contributes to the ability to perform physical activity (1). The International Physical Literacy Association defines PL as “the motivation, confidence, physical competence, knowledge, and understanding to value and take responsibility for engagement in physical activities for life” (2). It encompasses four conceptually interrelated domains, including physical (physical competence and skills, etc.), psychological or emotional (confidence, motivation, etc.), social (ethics, social skills, etc.), and cognitive (knowledge, strategy, etc.) (3–5). A previous study indicated that PL, being related to physical activity, may provide opportunities for health promotion and disease prevention (1). In school-age children, PL is positively associated with not only physical activity (6) but also health indicators such as body composition, fitness, blood pressure, and Health-Related Quality of Life (4, 7). A holistic exercise and education intervention that addressed the physical, behavioral, emotional, and cognitive components of physical activity was shown to improve PL and physical activity behavior among inactive adults (8), suggesting that PL acquisition enhances physical activity.

In Japan, the concept of PL has recently started gaining recognition. For the first time, the Third Basic Sports Plan, formulated in 2022 (9, 10), mentioned PL as a goal for fostering children’s lifelong engagement in sports and physical activity. Additionally, the Japan Sports Association (JSPO), organization responsible for promoting sports in Japan, has initiated a PL-related research project in 2021 (11). However, despite these advancements, the development and implementation of PL in Japan remain in their early stages compared to PL-leading countries such as Canada and Australia.

Previous studies have attempted to develop assessment tools for PL (12, 13). The first assessment tool for adults was used to assess the PL of physical education teachers (14). The United Nations Educational, Scientific, and Cultural Organization (UNESCO) promotes Quality Physical Education (QPE) (15). Previous studies in the field of education have suggested that QPE and PL are closely aligned, with the latter contributing positively to achieving QPE goals (16). It was thought that physical education teachers needed to understand the concept of PL for students to live active and healthy lives. Given that PL may promote physical activity, an assessment tool was also developed for use in young adults, such as university students. A previous study identified seven instruments for measuring PL in adults, three of which are original tools (13). Of these three, two target young adults, while one, the Perceived Physical Literacy Instrument (PPLI), was designed to assess physical education teachers (14). More recently, the Perceived Physical Literacy Questionnaire (PPLQ) has been validated as a PL assessment tool for adults in Germany (17). Currently, there is no validated PL assessment tool for Japanese adults; this highlights a need for future research to develop tools suitable for diverse populations and languages.

The International Sports and Culture Association (ISCA) proposed the Physical Literacy for Life self-assessment tool (PL4L) to enable the general European public to self-assess their PL levels. It is available on the ISCA website for public viewing, response, and feedback (18). The PL4L was developed based on the PL concept and consists of four domains (physical, emotional, cognitive, and social). Originally in English, it has been translated into multiple languages, including Slovenian, Spanish, Bulgarian, and French, and can be easily completed online. Although this tool is accessible and useful, to our knowledge, no published studies have examined its factor structure or internal consistency in any language. Furthermore, the potential correlation between its psychometric properties and physical activity has not yet been examined. Therefore, this study aimed to examine the factor structure and internal consistency of the PL4L in adults and its association with the stages of the change model for participation in regular physical activity.

The Transtheoretical Model represents the phases through which individuals progress when making conscious changes to their health behavior (19). These stages reflect variations in awareness and actions related to sustaining health behaviors. Therefore, by evaluating the relationship between the stage of change model for physical activity participation and PL, this study seeks to deepen the understanding of how PL impacts both the amount of physical activity and an individual’s awareness and readiness to engage in such behaviors. Thus, this study offers a valuable framework for assessing how PL influences habitual health behaviors.

The participant characteristics are presented in Table 1. The averages of age and BMI were 41.8 ± 13.2 years and 22.1 ± 3.90 kg/m². The proportion of participants with ≧ 16 years of education was 54.5%. In the stages of change model for participation in regular physical activity, the action stage recorded the lowest rate (5.6%) and the maintenance stage recorded the highest rate (26.5%). Table 2 depicts the scores for each item and the total PL scores. The mean for each PL item ranges from 2.0 to 2.3. The scores for each domain were as follows: physical domain 12.8 ± 2.1, emotional domain 8.5 ± 1.9, cognitive domain 6.1 ± 1.5, and social domain 6.1 ± 1.4; the total score was 33.5 ± 5.7.

In the confirmatory factor analysis, the structure of the PL4L was validated to conform to the assumed structure of the original version. The fit between the initial model and the data was as follows: GFI = 0.936, AGFI = 0.911, CFI = 0.955, RMSEA = 0.066, and AIC = 570.045. The modification indices were examined to guide the search for a model with a better fit. Although each fit index showed acceptable values, the modification indices between the error terms of Physical 5 (Coordination and Balance) and Physical 6 (Object Manipulation Skills) showed a potential for improving the model. The model was adjusted by assuming error covariance between these two items. Figure 1 depicts the modified factor model of the PL4L in Japanese adults. It provided an improved fit to the data: GFI = 0.955, AGFI = 0.937, CFI = 0.970, RMSEA = 0.053, AIC = 435.124. The factor loadings between domains and items ranged from 0.56 to 0.85, all of which were significant (p < 0.001) (27). In the stratified analysis, the model demonstrated acceptable fit in both, men and women subgroups. The fit indices for the men model were GFI = 0.937, AGFI = 0.912, CFI = 0.970, and RMSEA = 0.058, while those for the women model were GFI = 0.944, AGFI = 0.921, CFI = 0.962, and RMSEA = 0.053. Factor loadings ranged from 0.56 to 0.86 in both subgroups, with no items falling below 0.40. The CFA figures for each model are provided as Supplementary file 2.

The internal consistency coefficients are presented in Table 3. Cronbach’s α coefficient for all 16 items was 0.92. Analysis by domain revealed high internal consistency for the physical (α = 0.89), cognitive (α = 0.83), and social (α = 0.89) domains, and acceptable internal consistency for the emotional domain (α = 0.77). Similarly, McDonald’s ω coefficients also demonstrated good internal consistency, with values of 0.92 for all items, 0.89, 0.84, and 0.87 for the physical, cognitive, and social domains, respectively, while the emotional domain showed acceptable internal consistency (ω = 0.76).

Figure 2 depicts the relationship between PL and the stages of change for participation in regular physical activity. This study identified significant associations between PL and stages of change for participation in regular physical activity, with a coefficient of 0.38. The rest of the goodness-of-fit indices were acceptable (GFI = 0.930, AGFI = 0.907, CFI = 0.950, RMSEA = 0.065).

This study examined the factor structure and internal consistency of the PL4L. Although the initial model revealed an acceptable fit in the CFA, we examined a revised model by considering the error covariance between the error variables of Physical 5 and 6 based on the modification indices. This examination led to a model with an improved fit. The error covariance can only be assumed when it can be interpreted practically (22, 26). In the PL4L, Physical 5 assesses coordination and balance, while Physical 6 evaluates object manipulation skills. Both items reflect the deft movements of the body, which is likely why the assumption of error covariance resulted in a better model fit. Moreover, in the preset model, the path coefficients between each domain and the observed variables were satisfactory overall (27). Based on these results, the PL4L, conforms to the theoretical model presented in the ISCA (18) and previous studies (3–5) and is considered factorially valid. Internal consistency analyses revealed that Cronbach’s α and McDonald’s ω of all items and each domain exceeded the threshold of 0.7, indicating satisfactory internal consistency (28, 29), indicating that the PL4L demonstrates a certain level of reliability for assessing young and middle-aged adults.

Most PL assessment tools have been developed in school settings, primarily targeting children and young adults (30). The development of adult PL assessment tools faced issues due to a disproportionate focus on certain countries (13) and a lack of established tools that cover a wide age range. The PL4L has been translated into multiple languages and is freely accessible online. Although many individuals use the PL4L to self-assess their PL, the scientific basis for its use as an assessment tool, including its psychometric properties such as factor structure and internal consistency, has not yet been published or made publicly available. Importantly, this study provided valuable insights into the factor structure and internal consistency of the PL4L in Japanese adults. It elucidates the tool’s practical utility for users and promotes its dissemination and adoption within the academic community. However, the PL questionnaire may need to be adapted to account for environmental and cultural contexts (31). For example, the Physical Literacy Assessment for Youth-Self (PLAY-Self) for Canadian youth includes a question about whether the respondent is good at sports and activities on ice and snow (32). Some countries receive snowfall in only some areas; in such cases, the level of PL is influenced by the area in which the respondent lives. The PL4L was deemed suitable for responding to questions, regardless of the specific environmental or cultural context. However, future research needs to examine whether PL assessed by the PL4L differs between countries.

Physical activity is an important health promoting behavior (33–35). However, existing strategies to promote regular physical activity are insufficient to change people’s behavior (36), highlighting the need for more innovative approaches. A previous review identified 12 barriers to and 18 facilitators of physical activity among university students (37). These include factors that are part of the four domains of PL that we aim to capture: (1) having the physical skills and fitness to participate in physical activity; (2) having the motivation, enjoyment, and self-efficacy to participate in physical activity; (3) knowledge about the benefits and ways of performing physical activity; and (4) exercising with others. Therefore, PL may serve as a gateway to physical activity, potentially impacting health. Currently, studies have found that interest in PL has moved beyond its traditional roots in physical education and extends to the field of public health (1, 38). This study revealed a positive relationship between PL and the stages of change model for participation in regular physical activity among adults. This result is consistent with that of a prior report on children, indicating that PL is associated with physical activity (6). Our findings suggest that PL is a new determinant of daily physical activity.

This study had several limitations. First, Internet panel surveys have limited generalization potential (39). To address this, the participants were recruited to ensure an equal distribution of age and sex. However, expectations related to incentives may influence the response quality. Additionally, this study employed a convenience sample, which does not fully represent the general population of Japanese adults. It would therefore be beneficial to conduct surveys using random sampling or complete enumeration in future studies. This study examined the factor structure and internal consistency of the PL4L in adults. However, the tool’s test–retest reliability, which evaluates its temporal stability (40), has not yet been confirmed. Future studies should address this limitation and report on the tool’s test–retest reliability. Additionally, this study employed self-report questionnaires, which may introduce social desirability and recall biases (41). It would be beneficial to conduct surveys using random sampling or complete enumeration in future studies. While this study assessed the stages of change model for physical activity based on self-reported data, it is important to note its limitations. Objective measures, such as wearable devices, may provide more accurate estimates of physical activity. These tools could offer a more comprehensive understanding of the role of PL in promoting sustained physical activity. Finally, this cross-sectional study could not identify causal relationships. Future research should investigate whether addressing PL can effectively increase physical activity levels.

This study revealed that the PL4L has a factor structure that is consistent with the concept of PL among Japanese adults. Additionally, PL correlates with the stages of change for participation in regular physical activity. These findings may help evaluate PL and clarify its role in promoting physical activity among adults.