Social entrepreneurship plays an important role in economic growth and community empowerment. This phenomenon has been demonstrated to positively impact quality of life, contribute to the alleviation of poverty, and promote gender equality [1-4]. The concept of social entrepreneurship is based on a triple bottom line model, which integrates the dimensions of people, profit, and planet [5, 6]. The primary objective of social entrepreneurship is to address social issues through the framework of entrepreneurship [7, 8].

A critical component of fostering an increase in the social entrepreneurial landscape is the enhancement of social entrepreneurial intention. Social entrepreneurial intention is defined as a person’s interest and conviction accompanied by a comprehensive plan to establish a social enterprise [9]. It is also defined as an individual's belief and readiness to start a social enterprise [10]. Social entrepreneurial intention has been identified as a significant predictor of behavior, particularly in terms of goals, plans, and goal-directed behavior [11].

The importance of social entrepreneurship has been previously demonstrated in the context of addressing social challenges and promoting sustainable innovation to address issues related to social welfare, health, education, and the environment [12-14]. However, there is a need for further research to develop a comprehensive model that captures the essential elements of social entrepreneurial intention. Research on the social entrepreneurial intentions of university students is of significant importance, as students have the potential to become agents of change [15, 16]. It is therefore hoped that they can comprehensively solve various social problems and formulate policies that can be recommended to policymakers [17].

The social entrepreneurial intention is based on theories that are relevant to the context of intention and entrepreneurship. The development of entrepreneurial intention theory was initiated by the conceptualization of Shapero and Sokol's concept of entrepreneurial event theory [18]. Entrepreneurial intention theory is a theoretical framework that aims to improve the understanding of entrepreneurial intentions. This theory proposes that intentions are formed through a combination of rational and intuitive thinking processes. The process is influenced by several contextual factors, including social, political, economic, and historical elements. Furthermore, research has demonstrated that personality traits and abilities are significant factors in the development of entrepreneurial intentions [15].

The theory of behavioral intention [19] is a specific theoretical framework that focuses on individual intentions regarding specific behaviors, which are then referred to as “behavioral intentions.” Ajzen’s theory offers a general concept, yet it is frequently applied in entrepreneurship because its principles can explain the mechanisms underlying individual intentions. However, a more precise theory of intention is needed in the domain of social entrepreneurship, which led to the development of social entrepreneurial intention formation [20]. This theoretical framework can then provide a detailed explanation of how social entrepreneurial intention is constructed.

The Mair and Noboa theory has been developed and extensively referenced in various studies, particularly in the context of students, as it is most relevant to social entrepreneurship. Research on intentions is particularly relevant for student respondents, as they are entering a phase of career choice and preparation. It is assumed that they may be encouraged to become social entrepreneurs [21]. Furthermore, social entrepreneurship is predominantly driven by young people, as indicated by data from the Global Entrepreneurship Monitor (GEM), which reports that the majority of entrepreneurs are students. This finding is further substantiated by data from the British Council, which indicates that the majority of social entrepreneurs are individuals aged 18 to 34 [22, 23].

A review of the literature reveals that several studies have been conducted on social entrepreneurial intention in Indonesia. However, a review of the extant literature indicates that no new measurement tools have been developed, nor have adapted instruments employed, in Indonesia. A number of studies employ adapted instruments in English; however, these studies do not explicitly explain whether the instruments used were adapted or used initially in English. For instance, this issue has been identified in the research conducted by several previous researchers [24-26].

The absence of standardized instruments to measure social entrepreneurial intentions poses a significant challenge for Indonesian researchers, forcing researchers to utilize instruments developed in different contexts and countries, which can introduce variability and inconsistencies into the research findings. Sperber [27] explains that different research instruments require a process of adaptation first. The cross-cultural adaptation process is of critical importance in ensuring the equivalence of the measuring instrument with the original instrument [28]. Consequently, this study endeavors to address the identified knowledge gap.

A review of the extant literature identifies at least three instruments most frequently used to measure entrepreneurial intention among students. These instruments include the three-item Hockerts’ social entrepreneurial intention measurement tool [29], the six-item Linan and Chen’s entrepreneurial intention measurement tool [30], and the eight-item Ip’s social entrepreneurial intention scale [31].

In the context of research on the social entrepreneurship intentions of Indonesian students, the most suitable measurement instrument for the adaptation and validation process was the one developed by Ip (2017) [31]. The selection of this measurement was based on several reasons: the instrument was developed in Asia (Hong Kong), which is significant in the context of cultural and educational systems that are generally similar across the region. Secondly, the scale was developed and validated in the specific context of university students, indicating its suitability for use in this particular setting. Thirdly, the instrument was developed from an established entrepreneurial intention scale [32] and a credible intention theory [20]. This suggests that the theoretical construct is relatively robust. The selection of instruments is both contextually and theoretically consistent and takes into account good psychometric properties.

Theoretically, social entrepreneurial intention consists of two dimensions: conviction and preparation [31]. Conviction is defined as the degree of interest and confidence individuals have in establishing social enterprises in the future. Preparation refers to the extent of individuals' plans and readiness to become social entrepreneurs. Consequently, individuals with strong convictions and adequate preparation are more likely to become entrepreneurs in the future. However, the results of IP’s (2017) research indicate contradictory findings. Empirical evidence indicates that social entrepreneurial intention is a unidimensional scale, as evidenced by the eigenvalue score, which exceeds 1 [31]. Based on these research results, this study refers to a unidimensional scale.

In the context of the Indonesian language and culture, the researchers were interested in adapting and validating the social entrepreneurial intention scale based on the aforementioned background. This research is expected to significantly contribute to the development of measurement instruments in Indonesia, especially in the context of social entrepreneurial intentions among university students.

The present study has yielded an Indonesian version of the social entrepreneurial intention scale. The adaptation and validation processes were performed in accordance with established protocols. A cross-cultural adaptation approach was chosen because it enables cross-cultural comparison, which is essential for accumulating scientific evidence on SEI. Epstein noted that when seeking to obtain an adequate measuring instrument, cross-cultural adaptation is a more probable course of action than the creation of a new instrument, as it proves more efficient with respect to time, energy, and financial resources [38].

Ideally, measuring instruments should be tested under conditions as similar as possible to their intended purpose and use, within the same country. This will prove the conceptual equivalence of translations across nations. In the adaptation process, the social entrepreneurial intention scale fulfilled semantic, idiomatic, experiential, and conceptual equivalence according to Beaton et al. ’s guidelines [28].

The adaptation process prioritized semantic equivalence, with experts conducting a review to identify terms that might possess multiple meanings or present translation challenges. Expert analysis indicates that the number of terms difficult to translate is relatively low. Most of these terms have established Indonesian equivalents, thereby facilitating cross-cultural communication. The process also involved achieving idiomatic equivalence. However, some terms proved challenging for respondents to comprehend. These findings emerged from the cognitive interview process, which involved identifying expressions requiring further clarification and seeking equivalent versions to ensure comprehensibility.

Adjustments were made in the context of experiential and conceptual equivalence. These modifications were essential because cross-cultural adjustment was needed; the concept in different cultures did not fully align with the local context of Indonesian social enterprises, which typically feature less complex organizational structures and distinct operational models. For instance, the original item discussing “I will act as a professional manager by getting involved in the management of social enterprise through promotion” was adjusted and simplified to focus on “I will become a professional social entrepreneurship manager”. Consequently, the items have been adapted and simplified. Beaton [28] posits that the concepts and experiences that exist in one country differ from those in another because they are related to the culture or events in that country. Therefore, these two aspects of equivalence are important.

Validity encompasses the collection of pertinent evidence, providing a robust scientific foundation for interpretation [33]. Validity refers to the extent to which empirical evidence and theoretical frameworks support the interpretation of test scores for their intended use. This study incorporates three distinct categories of validity evidence: evidence derived from test content, evidence derived from response processes, and evidence derived from internal structure.

The first validity evidence employed is content validity. Content validity evidence was assessed using the Content Validity Index (CVI). The process under discussion involved a team of five experts, including linguists, methodologists, and social entrepreneurship experts, in accordance with Polit and Beck’s guidelines [35]. This expert review approach is considered highly adequate because it involves an in-depth discussion to ensure conceptual alignment. The calculation of the eight items on the SEI scale resulted in an S-CVI score of 0.92, exceeding the required minimum of 0.78. This indicates that all items are satisfactory. While the final score shows a high level of agreement, this process involved efforts to resolve initial disagreements on several items, which required intensive justification and modification until a satisfactory consensus was reached.

The second piece of validity evidence is response process evidence. A theoretical and empirical analysis of this evidence provides proof of the suitability between the construct and the responses given by participants [33]. The analysis, conducted through cognitive interviews using a verbal probe approach, included the following: first, identifying the meaning of the questions; second, collecting data; third, analyzing and comparing respondents' interpretations of the intended meaning [34]. The result of the analysis demonstrates that the questionnaire is suitable for participants due to its comprehensibility. The document features no substantial amendments, with the exception of the inclusion of comprehensive clarifications for select terminology that has proven to be challenging for the target audience to comprehend.

Following the completion of the adaptation process, the subsequent step involves the validation and reliability assessment of the instrument. The CFA analysis, as described in the result section, indicates that the factor loadings on the original scale (0.67-0.90) [10, 31]. The score obtained suggests that the adaptation process of the measuring instrument, which was carried out according to the established procedure, may not have been exhaustive. However, in terms of scores, it was adequate. Hair [43] posits that the loading factor is higher than 0.50, with an ideal score of 0.70. This variation may also be explained by the process of translation and the inclusion of different research participants.

Based on the results of the fit indices analysis, it can be seen that the value of the absolute fit indices, as indicated by an SRMR = 0.041, represents an excellent fit. SRMR reflects the average standardized difference between the observed correlation matrix and the correlation matrix predicted by the model. Smaller values indicate a better model fit [28, 29], and this is supported by the GFI value, which shows a score of 0.996 (Excellent fit). This illustrates the proportion of variance-covariance in the data explained by the model [30]. The second standard used is the relative fit indices, which in this study are represented by CFI=0.930 and TLI=0.903 (acceptable fit) [25, 26]. CFI compares the proposed model to a baseline (null) model that assumes there are no correlations among variables, while TLI also compares the proposed model to the baseline model, but applies a penalty for model complexity [44].

The maximum possible Average Extraction Value (AVE) is 0.457. The AVE value is less than 0.5, which is below the optimal range. Hair et al. posited that the optimal AVE value is greater than 0.5 [43], given that the AVE is an indicator of convergent validity, it can be strengthened by other evidence, specifically the factor loading value that has met the requirements. Considering the aforementioned points, it may be concluded that the social entrepreneurial intention scale is acceptable.

To ensure the validity and reliability of the measurement tool, the researcher must implement a validation process capable of anticipating potential biases. The anticipation of bias was undertaken to ensure that the relationship between the constructs [45]. Anticipation of bias was measured by providing a pause between one instrument and another, with the pause consisting of additional questions that were different from the questions in the instrument. This was implemented to prevent participants from identifying the relationship between the predictor and criterion variables, as well as to assess the participants’ level of attention. Abbey and Meloy [37] posit that the attention check was capable of identifying participants who were not paying attention when filling out the questionnaire. Anticipation of bias was also controlled by providing clear instructions and informed consent. The data of participants who completed the questionnaire without anticipation of bias were the only ones considered in this study.

The adaptation and validation processes were carried out to the fullest extent possible; however, this study still has limitations that cannot be avoided. The initial constraint pertains to the sample, which comprises students from three private universities in Indonesia. From a statistical perspective, the sample size can be considered adequate. However, the use of convenience sampling may limit the external validity of the study. The result may not fully represent the student population throughout Indonesia, especially those from public universities, polytechnics, or other institutions with different curricula and environments.

Second, the validity measurement did not reach the maximum value and was lower than the original scale measurements. This phenomenon may be attributed to an imperfect equating process, influenced by factors such as experience and context, each with distinct characteristics. When evaluated from the perspectives of participants in scientific, health-related, and engineering disciplines, the outcomes diverge from those observed among participants in the social sciences. This phenomenon may be attributed to the presence of unfamiliar terminology within the instrument’s items.

Future researchers should prioritize homogeneity among respondents and a more representative sample to address these limitations. In the context of social entrepreneurship, it is imperative to assess the intention to pursue this vocation in a diverse setting. These settings may include academic institutions, community groups, and commercial enterprises.

The limitation is the sampling technique. We used a non-random sampling method, which limits the ability to generalize the findings on the validity and structure of the instrument to the entire population of Indonesian students. Therefore, the result obtained should be interpreted as strong evidence regarding the validity of the instrument itself in a representative sample of students for the context of measuring SEI. Future researchers in Indonesia who wish to use this adapted instrument are advised to contact the corresponding author to obtain a copy of the final Indonesian questionnaire.

The Indonesian version of the instrument was obtained based on the results of the adaptation and validation of the social entrepreneurial intention scale. The adaptation process is carried out through the stages of Beaton, which consist of five stages: forward translation, synthesis of forward translations, backward translation, expert review committee, and pre-final testing. Concurrently, the validation process manifests three distinct manifestations of validation: evidence of the test content, evidence of the process response, and evidence of the internal structure.

The results of the test demonstrate content validity, evidence of process response, and evidence of internal structure, indicating adequate results overall. The results of the content analysis indicate a content validity index (CVI) score of 0.92, which exceeds the minimum threshold of 0.78, thereby substantiating the content's validity. The efficacy of the process response is substantiated by the outcomes of cognitive interviews, encompassing aspects such as comprehension, retrieval, judgment, response, and the adequacy of content. Generally, the participant's response demonstrates a high degree of understanding of the measuring instrument; however, certain inputs related to specific aspects have undergone adjustment. Evidence of internal structure is obtained from the results of Confirmatory Factor Analysis (CFA). This analysis demonstrates factor loading, scale reliability coefficient, and an adequate model fit index. Therefore, it can be assumed that the model of the social entrepreneurial intention scale has adequate data.

The sample primarily consisted of three universities located in a region. Consequently, the findings are not sufficiently representative to draw conclusions about the conditions in other regions. In addition, the findings of the validity measurements did not delineate the maximum value and were lower than the original scale score. Future research should strive to include a more diverse sample in terms of geographic location, institution type, and socioeconomic background. Although the study includes validity evidence from the response process and internal structure, future research could strengthen validity by including evidence from relationships with other variables. Research could test social entrepreneurial intention and other factors, such as empathy, social values, etc.