Evidence of validity and factorial invariance of the cultural competence measurement scale (EMCC-14) among students of healthcare sciences in Panama Authors: Janeth Agrazal-García, Ricaurte Tuñón, Victor Pedrero Published: 2025 ABSTRACT IntroductionThe assessment of cultural competence is crucial for preparing health science students in Latin America to provide equitable and culturally responsive care. The Cultural Competence Measurement Scale (EMCC-14), originally developed in Spanish within the region, has shown strong psychometric properties among healthcare professionals; however, its validity has not yet been examined in student populations. ObjectiveTo analyze the validity, reliability and factorial invariance of the EMCC-14 cultural competence scale in health science students in Panama.Materials and methodsA convenience sample of 565 students from nursing, medicine, pharmacy, psychology, and dentistry programs completed the EMCC-14 scale and a sociodemographic questionnaire. The validation process was conducted in two phases: (a) evaluation of the linguistic and cultural adequacy of the scale and (b) analysis of the internal structure through confirmatory factor analysis (CFA), estimation of reliability for the total scale and its dimensions, and assessment of factorial invariance across academic subgroups.ResultsNo linguistic adjustments were needed. The CFA confirmed the original three-factor structure (χ²/df = 3.24, CFI = 0.943, TLI = 0.930, RMSEA = 0.063, 90% CI [0.054–0.072], SRMR = 0.046). The reliability indices were satisfactory for the overall scale (α = 0.867, ω = 0.866) as well as for the knowledge (α = 0.835, ω = 0.836) and skills (α = 0.837, ω = 0.837) dimensions. However, the sensitivity dimension showed lower reliability (α = 0.653, ω = 0.653). Measurement invariance was supported between medical and nursing students.ConclusionsThe EMCC-14-PAN version for health science students showed adequate evidence of validity and reliability, similar to the original scale designed for health professionals. This version is a useful instrument for studying the cultural competence of students in health sciences.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12909-025-08190-8. Introduction The processes of globalization and the trend of greater migration between and even within countries represent a challenge for health professionals, who must provide care and health care to a culturally diverse population [1]. Owing to their geographic, cultural, ethnic, age, language, gender expression, education, socioeconomic, disability or rurality conditions, culturally diverse populations experience unequal health outcomes than do members of dominant cultural groups [2]. Culturally competent health care that is sensitive to cultural and social diversity is an essential factor in reducing health inequalities [3, 4]. Latin America is considered one of the most culturally diverse regions in the world, with a great variety of ethnic, social and cultural groups. Many of these groups experience social and economic disparities, which have negative consequences for the health of these populations [5]. Panama, a Latin American country, has historically been characterized as a multicultural and diverse country. In 2023, the total population in Panama was 4,064,780 inhabitants, of which 17.2% corresponded to indigenous ethnic groups, 31.7% were self-declared Afro-descendants and 6.1% were foreigners [6]. This multicultural characteristic of Panama’s population represents a challenge and highlights the need to train future health professionals in cultural competence. There are different theoretical perspectives on cultural competence [7]. One of them is the Multidimensional Model of Cultural Competence proposed by Sue and Sue [8], which provides a broad and flexible framework for understanding this construct. This model has been widely applied in different countries and has also been used in studies conducted in Latin America [7, 9–13]. From this perspective, cultural competence is conceptualized as a dynamic, multidimensional, and multilevel process. First, it is dynamic because its development involves an active, evolving, and ongoing process of refinement that enables professionals to progress from initial to more advanced stages of competence. Second, it is multidimensional, as it requires health professionals and institutions to acknowledge their own biases and preconceived beliefs, to understand both their own cultural contexts and those of others, and to integrate these perspectives into the care of diverse groups. Finally, it is multilevel, insofar as competence should not be confined to individual practice but must also be reflected in institutional and social policies that promote and protect diversity. This framework is consistent with other approaches used in the region, such as Intercultural Competence [14, 15]. Despite certain differences, both perspectives view competence as a progressive process, emphasizing the importance of self-awareness of one’s own cultural background and respect for the diversity of others. Cultural competence training for students and future health professionals is essential to ensure culturally respectful, equitable, and high-quality care [16–18]. International reviews confirm the effectiveness of cultural competence interventions in improving knowledge, skills, and attitudes among health science students [19–22]. In this context, the active integration of cultural competence training into academic curricula has been acknowledged as essential, whether through specific coursework or through experiential approaches, such as cultural immersion [17, 20–22]. However, evidence from Latin America regarding cultural competence training among health science students remains scarce [18]. Findings from Chile suggest that students progress through stages of competence, a process supported by contact with culturally diverse users and by modeling during training [14]. However, findings also indicate that students in later years may report lower levels of competence [1], underscoring the importance of sustained curricular integration. A key aspect for understanding the level of cultural competence and evaluating the effects of interventions on health science students is to have scales for measuring cultural competence that are valid and reliable in the local context. Various systematic reviews of scales designed to measure cultural competence in health sciences professionals and students have been identified in the literature. Most of these scales were originally developed in the English language and have established different evidence of validity and reliability [23–25]. To the best of our knowledge, only three of these scales have been adapted into Spanish: the Cultural Capacity Scale (CCS) [26], the Cultural Competence Assessment (CCA) [27] and the Transcultural Self-Efficacy Tool (TSET) [28]. The Cultural Capacity Scale (CCS) [26] was validated among nursing students in Chile and has a one-dimensional factor structure, which is consistent with its original version. This structure suggests that it is not possible to differentiate the three components that make up cultural competence (sensitivity, skills and knowledge), which represents an important limitation, since these dimensions could present different patterns of behavior; therefore, their distinction is essential for an accurate assessment. Additionally, while the scale has been exclusively tested among nursing students, there is no evidence of validity among other student populations in the health field. Unlike the CCS, the Cultural Competence Assessment (CCA) was validated in Spain in a sample composed of various health professionals [27]. However, its factor structure showed substantial differences with respect to the original version, which suggests some instability of the instrument. While the original version of the scale has two dimensions (cultural sensitivity and culturally competent behaviors), in the adaptation to Spanish, four dimensions were identified, and the items were distributed in a manner inconsistent with the original scale. Another scale adapted to Spanish is the Transcultural Self-Efficacy Tool (TSET), which, although it maintains its three dimensions in the translated version, has only been validated among nursing students in Spain [28]. A notable limitation of this scale is its length, since it consists of 83 items, which hinders its routine use in academic or clinical contexts. In addition to the scales mentioned above, the literature review identified the Cultural Competence Measurement Scale (EMCC-14, by its Spanish acronym: Escala de Medición de la Competencia Cultural), developed in Chile, is the only instrument originally developed in Spanish within the Latin American region [9]. The EMCC-14 was previously validated in a diverse sample of health care professionals, including nurses, physicians, midwives, physical therapists, and nutritionists [9]. The construct of cultural competence measured by this scale comprises three dimensions (sensitivity, knowledge and behaviors) following the Multidimensional Model of Cultural Competence [8]. The presence of these dimensions in the instrument was confirmed through confirmatory factor analysis. Additionally, invariance analyses were considered to ensure that their functioning was equivalent across the different groups of professionals analyzed (nurses, physicians, midwives, physical therapists, and nutritionists). The reliability of the scale, both at a general level and by dimension, was adequate. This scale has shown stability in its factorial structure in the population of nurses in Mexico [10] and mental health professionals in Brazil [29]. Furthermore, its reliability has been consistently supported across different studies, with alpha coefficients ranging from 0.7 to 0.9 [9, 10, 29, 30]. Beyond its psychometric properties, there are further reasons for selecting the EMCC-14 scale in this study. First, this scale is based on the Multidimensional Model of Cultural Competence [8]. Second, the scale was created in Chile, a Latin American country whose official language is Spanish. This is an important consideration, since few scales are available that measure cultural competence in Spanish-speaking environments [25–28]. Furthermore, Spanish is the official language in Panama, where the current study was carried out. Although this scale was originally designed to measure cultural competence in health professionals, it has been used with dental graduate students [31] and nursing students [32–34]. Even though it has been used previously among health science students, there is no research that provides evidence of its validity and reliability in this group. This aspect is essential because the use of scales in another context and group for which it was not originally designed may distort or invalidate the result [35, 36]. This study aims to analyze the validity, reliability, and factorial invariance of the EMCC-14 scale among health science students in Panama. Methods Study design An instrumental cross-sectional study [37] of the linguistic and psychometric adaptation of the EMCC-14 [9] scale among health science students. The validation process was carried out in two phases: (a) the linguistic adaptation of the scale to the context and culture of Panama to obtain an equivalence between the original scale in Spanish and the target scale in the same language but a different cultural environment [36, 38]; (b) an analysis of the evidence of the validity of the internal structure of the EMCC-14 scale, evaluated by confirmatory factor analysis (CFA), and the calculation of the reliability of the general scale and its dimensions; and (c) an analysis of invariance for establishing the equivalence of the factorial structure of EMCC-14 between subpopulations of students. Research setting and participants This study was conducted in Panama from April to December 2024. Nursing, medicine, dentistry, pharmacy and psychology students from public and private universities in Panama were invited to participate in the study. The inclusion criterion for participation in the study was having had experience caring for patients or healthy people during their academic training. Data collection was carried out through an online survey via the SurveyMonkey tool. Data collection was carried out in two stages: the first aimed at the linguistic adaptation of the EMCC-14 scale, and the second aimed at the search for evidence of the validity of the internal structure, calculation of reliability and factorial invariance between groups. In the first stage, 16 students from three universities responded to the survey, and in the second stage, 565 students from six universities responded. In both stages, participation was voluntary, and participants were chosen through convenience sampling. Measuring instruments Cultural competence The EMCC-14 scale was used [9]. This scale consists of 14 items divided into three factors: sensitivity to one’s own prejudices (Items 1–4), cultural knowledge (Items 5–8) and ability to work in culturally diverse environments (Items 9–14). The response options of the scale are provided on a five-point Likert scale: 1-totally disagree, 2-disagree, 3-neither agree nor disagree, 4-agree and 5-totally agree. For a better understanding and interpretation of the total scale and the subdimensions, the scores are scaled on a metric of 0–100; the higher the score is, the greater the degree of cultural competence [39]. English and Spanish versions of the scale are provided in the supplementary file. Sociodemographic aspects and student experience Sociodemographic variables and experience were selected because they were related to cultural competence in previous studies [9, 40]. The structured questionnaire included personal data, such as age, gender (female/male), the career of the health sciences studied (nursing/medicine/pharmacy/psychology/dentistry), name of the university in which you study, level or year of the course you are studying, and experience in caring for diverse cultural groups (Have you had experience in caring for individuals from cultural groups different from your cultural group? YES/NO) and if you have received training in cultural competence in health care (Have you received any training in cultural competence in health care? YES/NO). Data analysis The data analysis was carried out in three stages: (1) linguistic adaptation of the scale to the context and culture of Panama; (2) confirmatory factor analysis (CFA) and reliability of the scale; and c) factorial invariance. The open-access software JASP version 0.95.2 was used in this analysis [41]. Cultural adaptation A group of end users, comprising 16 health science students, was involved as expert judges in the evaluation process. First, the students completed all the questions on the EMCC-14 scale via an online form. Afterward, they provided feedback on the clarity and comprehension of each question and made suggestions for improvement. Please comment on whether the questions were clear or if any were difficult to understand (mention which ones). Would you suggest any modifications? The research team reviewed health science students’ comments and suggestions about the clarity and understanding of the items on the EMCC-14. All the participants reported that the 14 items on the EMCC-14 scale were clear and easy to understand. Only one participant suggested formulating the items in the future verb tense to consider students who did not yet have experience in providing health care to patients. This suggestion was reviewed by the research team and the author of the original scale; however, no adjustments were made, and the original format of the scale was maintained, since its purpose is to measure cultural competence in students with experience in caring for patients during their professional training. Confirmatory factor analysis and reliability For sample size estimation, the number of participants in the original EMCC-14 study was used as a reference [9]. That study included a total of n = 483 participants; therefore, the present study aimed to achieve at least an equivalent sample size. For the statistical analysis, a descriptive analysis was first conducted, including the characteristics of the study participants and the descriptive statistics of the EMCC-14 scale items. Confirmatory factor analysis (CFA) was then performed, and the three-factor model of the original EMCC-14 scale was tested. The maximum likelihood estimation method was used with a robust estimator given the lack of normality of the data [42]. To evaluate the fit of the model, the following goodness-of-fit indices were used: χ² (Chi square), comparative fit index (CFI), Tucker Lewis index (TLI), root mean square error approximation (RMSEA) and standardized root mean square residual (SRMR). A good fit of the model was considered with CFI and TLI values ≥ 0.90, ideally close to 0.95. The SRMR is close to or less than 0.08, and the RMSEA is less than or equal to 0.06, with a maximum of 0.08 [43]. Considering that the chi-square statistic is sensitive to sample size [43], the ratio between the model’s chi-square value and its degrees of freedom was calculated, with values less than 3.0 indicating an acceptable model fit. Afterward, the reliability of the total scale and each of its dimensions was calculated via two approaches: Cronbach’s alpha coefficient (α) and McDonald’s omega coefficient (ω) [44]. These indices are conceptually distinct, while Cronbach’s alpha is based on the interitem correlation matrix, omega is estimated from factor loadings. Although omega is generally considered a more accurate index, particularly when factor loadings vary across items, Cronbach’s alpha remains the most commonly reported measure of reliability and has been used in previous studies with the EMCC-14. Reporting both indices provides a more comprehensive view of reliability and allows comparison of the present results with those of earlier studies using the EMCC-14. Measurement of factorial invariance To determine the equivalence of the instrument factorial structure between subgroups, groups of students from different programs were subjected to factorial invariance tests [45]; this is consistent with the original approach followed by the authors of the instrument [9]. Given the differences in sample size between the different groups of students, factorial invariance was only examined in the groups of medical students (n = 136) and nursing students (n = 246), which had larger sample sizes. Prior to this analysis, CFA was performed on these two groups. The equivalence assessment was performed in several steps by comparing nested CFA models for multiple groups [45, 46]. In the first step, the configural invariance (M1) was measured, an unrestricted model that implies that the number of factors, three in this case, is similar in the different groups. In the second step, the metric invariance (M2) was measured, a model with restrictions to equal factor loadings that seeks to guarantee equivalence of the relationship between the items and the latent factors across the different groups. In the third step, scalar invariance (M3) was measured, which implies equal factor loadings and equal intercepts of the items (means of the items). In the fourth step, strict invariance (M4) was measured, which implies equal factor loadings, equal intercepts of the items, and equal errors. To evaluate the presence of invariance, two conditions must be met. The first is that each model fits adequately; that is, it shows goodness-of-fit indicators (RMSEA, CFI, TLI) within the acceptable ranges for a CFA. Second, the most restrictive model must exhibit a comparable fit to that of the immediately preceding model. For example, to accept a strict level of invariance (M4), the fit of this model should not be substantially worse than the fit of the previous model, in this case, the scalar model (M3). To compare the models and determine their equivalence, the criterion of an increasing CFI (Δ CFI) and RMSEA (Δ RMSEA) was proposed, which states that when the difference between the CFI of a model and the one that precedes it is equal to or less than 0.01 and the RMSEA is less than 0.015, the model is acceptable, and the analysis of the following model continues [47]. The instrument is considered comparable between the groups if there is at least scalar invariance. Results A total of 565 health science students answered the EMCC-14. Among these, 43.5% were pursuing nursing careers, 24% were studying medicine, 14.6% were studying pharmacy, 9% were studying psychology, and 8.6% were studying dentistry. In all careers, there was a predominance of women, with proportions greater than 85% in the psychology, dentistry and nursing programs. The average total age was 23.01 years (SD = 3.98), the minimum age was 18 years, and the maximum age was 55 years. The average ages were similar across the different programs (see Table 1). Table 1Characteristics of the participants in the studyHealth studentsMaleFemaleAge (years)MeanSDMin–maxNursing (n = 246)36 (14.6%)210 (85.4%)22.783.7918–39Pharmacy (n = 83)21 (25.3%)62 (74.7%)23.676.2719–55Medicine (n = 136)48 (35.3%)88 (64.7%)22.152.1818–34Dentistry (n = 49)6 (12.2%)43 (87.8%)25.872.9123–42Psychology (n = 51)5 (9.8%)46 (90.2%)22.643.3618–35Total116 (20.5%)449 (79.5%)23.013.9818–55Note. n = 565 Table 1Characteristics of the participants in the studyHealth studentsMaleFemaleAge (years)MeanSDMin–maxNursing (n = 246)36 (14.6%)210 (85.4%)22.783.7918–39Pharmacy (n = 83)21 (25.3%)62 (74.7%)23.676.2719–55Medicine (n = 136)48 (35.3%)88 (64.7%)22.152.1818–34Dentistry (n = 49)6 (12.2%)43 (87.8%)25.872.9123–42Psychology (n = 51)5 (9.8%)46 (90.2%)22.643.3618–35Total116 (20.5%)449 (79.5%)23.013.9818–55Note. n = 565 Characteristics of the participants in the study Note. n = 565 The descriptive statistics of the items on the EMCC-14 scale are presented in Table 2. The mean scores of the items ranged from 2.93 to 4.18 on a 5-point scale (1 = totally disagree, 5 = totally agree). The items with the lowest average score were Item 4, “I am aware that my beliefs about patients influence the therapeutic recommendations that I give them”, and Item 3, “I believe that my cultural context influences my attitudes and beliefs about other cultural groups”, with values of 2.93 and 3.06, respectively. The items with the highest average score were Item 5, “The beliefs, values and customs of patients should be valued in health care”, and Item 13, “I make an effort to explain to the patients their medical treatment, even if they believe that the cause of their illness is supernatural,” with values of 4.25 and 4.18, respectively. With respect to asymmetry, all the items presented negative asymmetry; six items presented values close to zero, and the remaining items presented values greater than one, ranging between − 1.10 and − 1.79. With respect to kurtosis, three items had values > 3 (Items 13, 14, 5 and 7), and three items had negative values (Items 1, 3 and 4). Table 2Descriptive statistics of the items of the Panama version of the EMCC-14ItemN°ItemsDescriptive statisticsMeanSDSkewnessKurtosis1I believe that patients who have different beliefs and customs have different expectations and/or needs in healthcare.3.691.28−0.94−0.172I believe that patients’ beliefs, values and customs affect their health.3.751.11−0.960.403I believe that my cultural context influences my attitudes and beliefs about other cultural groups.3.061.23−0.10−1.034I am aware that my beliefs about patients have an influence on the therapeutic recommendations that I give them.2.931.23−0.06−1.085The beliefs, values and customs of patients should be valued in health care.4.250.98−1.793.286I believe that becoming more familiar with patients’ beliefs and customs will help me establish more appropriate treatments.4.091.05−1.421.817I believe each patient has their own concepts of health and illness.4.030.92−1.553.028I believe that patients’ health problems must be understood within their cultural context (beliefs, values and customs).3.910.94−1.061.299I ask patients and their families to express their expectations regarding health and healthcare.3.810.86−0.891.3610I can identify potential barriers to access healthcare services that patients may face.3.980.86−1.342.8211I am able to establish therapeutic goals and/or objectives considering the cultural context (beliefs and customs) of my patients and their needs.3.850.84−1.102.1212I document the patient’s beliefs and customs in their clinical record.3.771.00−0.890.5613I make an effort to explain to the patients their medical treatment, even if they believe that the cause of their illness is supernatural.4.180.93−1.723.5814I am aware of potential difficulties that may arise due to cultural differences between me and the patient.4.060.89−1.523.27 Table 2Descriptive statistics of the items of the Panama version of the EMCC-14ItemN°ItemsDescriptive statisticsMeanSDSkewnessKurtosis1I believe that patients who have different beliefs and customs have different expectations and/or needs in healthcare.3.691.28−0.94−0.172I believe that patients’ beliefs, values and customs affect their health.3.751.11−0.960.403I believe that my cultural context influences my attitudes and beliefs about other cultural groups.3.061.23−0.10−1.034I am aware that my beliefs about patients have an influence on the therapeutic recommendations that I give them.2.931.23−0.06−1.085The beliefs, values and customs of patients should be valued in health care.4.250.98−1.793.286I believe that becoming more familiar with patients’ beliefs and customs will help me establish more appropriate treatments.4.091.05−1.421.817I believe each patient has their own concepts of health and illness.4.030.92−1.553.028I believe that patients’ health problems must be understood within their cultural context (beliefs, values and customs).3.910.94−1.061.299I ask patients and their families to express their expectations regarding health and healthcare.3.810.86−0.891.3610I can identify potential barriers to access healthcare services that patients may face.3.980.86−1.342.8211I am able to establish therapeutic goals and/or objectives considering the cultural context (beliefs and customs) of my patients and their needs.3.850.84−1.102.1212I document the patient’s beliefs and customs in their clinical record.3.771.00−0.890.5613I make an effort to explain to the patients their medical treatment, even if they believe that the cause of their illness is supernatural.4.180.93−1.723.5814I am aware of potential difficulties that may arise due to cultural differences between me and the patient.4.060.89−1.523.27 Descriptive statistics of the items of the Panama version of the EMCC-14 Confirmatory factor analysis and reliability The CFA of the EMCC-14 Panamanian version confirmed the structure of the original three-factor scale, with results of the model’s fit indices (χ2/df = 3.24, CFI: 0.943, TLI: 0.930, RMSEA 0.063, 90% CI [0.054–0.072], SRMR: 0.046). The factor loadings were adequate and higher than the 0.45 parameter. The item with the highest factor load was Item 14, “I am aware of potential difficulties that may arise due to cultural differences between me and the patient” (Factor 3), and Item 6, “I believe that becoming more familiar with patients’ beliefs and customs will help me establish more appropriate treatments” (Factor 2), with factorial loads of 0.83 and 0.809, respectively. The items with the lowest factor load were Item 4, “I am aware that my beliefs about patients influence the therapeutic recommendations that I give them,” and Item 3, “I believe that my cultural context influences my attitudes and beliefs about other cultural groups,” both of which are Factors 1, with a factorial load of 0.476 and 0.528, respectively. The correlations between the three factors of the scale were adequate and significant. The correlation between cultural sensitivity and cultural skills was moderate (r = 0.52; p < 0.001), whereas the correlations between cultural sensitivity and cultural knowledge and between cultural skills and cultural knowledge were high (r = 0.72; p < 0.001 and r = 0.74 (p < 0.001), respectively). With respect to the reliability of the scale, adequate results were obtained for the global scale (α = 0.867, ω = 0.866), knowledge (α = 0.835, ω = 0.836), and skills (α = 0.837, ω = 0.837) dimensions. However, the sensitivity dimension was below the commonly recommended threshold of 0.70 (α = 0.653, ω = 0.653) (Fig. 1). Fig. 1Factor loadings of the Confirmatory Factor Analysis (CFA) of the EMCC–14 Scale, Panama version (n = 565). Note: The reliability of the dimensions was as follows: sensitivity (α = 0.653, ω = 0.653), knowledge (α = 0.835, ω = 0.836), skills (α = 0.837, ω = 0.837) and global scale (α = 0.867, ω = 0.866) Factor loadings of the Confirmatory Factor Analysis (CFA) of the EMCC–14 Scale, Panama version (n = 565). Note: The reliability of the dimensions was as follows: sensitivity (α = 0.653, ω = 0.653), knowledge (α = 0.835, ω = 0.836), skills (α = 0.837, ω = 0.837) and global scale (α = 0.867, ω = 0.866) Factorial invariance by group of students As a preliminary step in the factorial invariance analysis, the CFAs were examined by a subgroup of students. In the group of nursing students (n = 246), the five fit indices were adequate for confirmatory factor analysis (χ2/df = 1.76, CFI: 0.948, TLI: 0.936, RMSEA 0.056, 90% CI [0.040–0.071], SRMR: 0.049). The factor loadings were adequate and ranged from 0.502 to 0.916. In the group of medical students (n = 136), only four of the five indices were adjusted, since an RMSEA value greater than 0.08 was obtained (χ2/df = 1.97, CFI: 0.949, TLI: 0.935, RMSEA 0.085, 90% CI [0.064–0.105], SRMR: 0.054). The factor loadings were adequate and ranged from 0.437 to 0.927. The item with the lowest factor loading for both the nursing students and medical students was Item 3 (Factor 1), and the item with the highest factor loading for both groups was Item 14 (Factor 3) (Table 3). Table 3Confirmatory factor analysis (CFA) of the Panama version of the EMCC–14 scale for nursing and medical studentsNursing students (n = 246)Medical students (n = 136)ItemsFactor 1 SensitivityFactor 2 KnowledgeFactor 3 SkillsFactor 1SensitivityFactor 2 KnowledgeFactor 3SkillsItem 10.7220.716Item 20.8230.814Item 30.5020.437Item 40.5480.501Item 50.8800.882Item 60.8440.844Item 70.8290.83Item 80.7600.755Item 90.7800.751Item 100.8110.809Item 110.8050.776Item 120.750.746Item 130.8490.862Item 140.9160.927 Table 3Confirmatory factor analysis (CFA) of the Panama version of the EMCC–14 scale for nursing and medical studentsNursing students (n = 246)Medical students (n = 136)ItemsFactor 1 SensitivityFactor 2 KnowledgeFactor 3 SkillsFactor 1SensitivityFactor 2 KnowledgeFactor 3SkillsItem 10.7220.716Item 20.8230.814Item 30.5020.437Item 40.5480.501Item 50.8800.882Item 60.8440.844Item 70.8290.83Item 80.7600.755Item 90.7800.751Item 100.8110.809Item 110.8050.776Item 120.750.746Item 130.8490.862Item 140.9160.927 Confirmatory factor analysis (CFA) of the Panama version of the EMCC–14 scale for nursing and medical students Factor 1 Sensitivity Factor 3 Skills Table 4 shows the sequential estimation of the factorial invariance between nursing and medical students. The configural invariance (M1) presented good fit values: CFI: 0.925, TLI: 0.908, RMSEA 0.081, 90% CI [0.069–0.092], SRMR: 0.055. The M1 values reflect the same factorial structure in both groups. M2 presented good fit indices: CFI: 0.923, TLI: 0.912, RMSEA 0.062, 90% CI [0.067–0.090], SRMR: 0.062. These indices are similar to those obtained in M1, with a low incremental difference Δ CFI = 0.002 and ΔRMSEA = 0.002, which support the equivalence of the factor loadings in the two groups. The fit indices for the scalar invariance model (M3) were comparable to those observed in the metric invariance model (M2): CFI: 0.918, TLI: 0.913, RMSEA 0.078, 90% CI [0.067–0.090], SRMR: 0.061. The differences between this model and the model above were a minimum Δ CFI = 0.001 and ΔRMSEA = 0.001, and the scalar invariance of the items was accepted. Finally, the strict invariance model (M4) demonstrated adequate fit indices (CFI: 0.909, TLI: 0.910, RMSEA: 0.080, 90% CI [0.069–0.090], SRMR: 0.061), with minimal changes relative to the previous model, and within the recommended thresholds for CFI and RMSEA. Accordingly, strict invariance was accepted. Table 4Invariance analysis of the EMCC-14 and Panama 2024 versionsNivelχ2dfCFITLISRMRRMSEAΔCFIΔ RMSEAM1331.8971480.9250.9080.0550.081M2347.3021590.9230.9120.0620.0790.0020.002M3369.8861700.9180.9130.0610.0780.0010.001M4407.4231840.9090.9100.0610.0800.0090.002Note. M1: Configural; M2: metric; M3: scalar; M4: strict Table 4Invariance analysis of the EMCC-14 and Panama 2024 versionsNivelχ2dfCFITLISRMRRMSEAΔCFIΔ RMSEAM1331.8971480.9250.9080.0550.081M2347.3021590.9230.9120.0620.0790.0020.002M3369.8861700.9180.9130.0610.0780.0010.001M4407.4231840.9090.9100.0610.0800.0090.002Note. M1: Configural; M2: metric; M3: scalar; M4: strict Invariance analysis of the EMCC-14 and Panama 2024 versions Note. M1: Configural; M2: metric; M3: scalar; M4: strict Discussion The objective of this research was to provide evidence of the validity and reliability of the EMCC-14 scale designed for health professionals [9] in a specific group of health science students (nursing, medicine, dentistry, psychology and pharmacy) in Panama. The findings of this study indicate that the Panamanian adaptation of the EMCC-14 scale for health science students retains the internal structure of the original instrument. A moderate to strong association was evidenced between the dimensions and adequate reliability of the scale, both globally and by the dimensions. Configural, metric, scalar and strict invariance were identified among the group of medical and nursing students. Linguistic–cultural equivalence was verified through surveys of end users (students), who reported that the items on the EMCC-14 scale were understandable and clear; this finding supported the full use of the scale in Panama. These data are relevant because, even when scales are in the same language, it is still necessary to assess linguistic adequacy [38]. Confirmatory factor analysis among health science students demonstrated an adequate fit [43] and confirmed a three-factor structure that is consistent with theoretical frameworks and with the original scale [8, 9]. This result is noteworthy, as it demonstrates that the EMCC-14 scale maintains the same internal structure when applied to health science students as when used with the group of health professionals for whom it was originally designed. Notably, the three identified factors showed adequate correlations. This finding is consistent with the theoretical basis of the scale, which conceptualizes cultural competence as a multidimensional process [8]. In this model, health professionals progressively develop cultural sensitivity, that is, awareness of their own culture, beliefs, and values and recognition of how these may generate biases, stereotypes, and prejudices, along with respect to and appreciation for cultural diversity. They also integrate knowledge, which involves understanding the perspectives of culturally diverse groups and their sociohistorical contexts, and they acquire the communication and intervention skills necessary for working effectively with culturally diverse populations [8]. In this study, the reliability of the scale was assessed via two indices, Cronbach’s alpha (α) and McDonald’s omega (ω). The coefficients obtained for the general scale were adequate [44], consistent with the results reported in the original validation [9] and in subsequent studies conducted in Colombia [48], Mexico [10], and Brazil [29]. At the dimensional level, the knowledge and skills dimensions showed satisfactory values [9]. However, the sensitivity dimension presented a reliability coefficient of 0.653, which is slightly below the commonly recommended threshold of 0.70. Reliability reflects the proportion of variance attributable to measurement error, whereas a coefficient of 0.70 indicates that approximately 30% of the variance is error, and a value of 0.65 suggests approximately 35% [44]. From this perspective, although the reliability of the sensitivity dimension falls marginally short of conventional standards, it does not constitute a major psychometric limitation. Nonetheless, scores from this subscale should be interpreted with greater caution than those from the other dimensions of the EMCC-14. Moreover, the results confirmed the measurement invariance of the EMCC-14 scale between medical and nursing students. This result demonstrates that cultural competence is similarly understood across students from distinct health-related disciplines, thereby supporting the validity of cross-group score comparisons in future studies [45, 46]. Together, these findings are important because the population assessed in this study (health science students) differs from that of the original validation (health professionals), thereby extending the applicability of the EMCC-14 to a new context. The evidence of validity and reliability obtained in the Panamanian context is particularly relevant in view of the current challenges of interprofessional training, which promotes collaboration across disciplines as a strategy to achieve universal health coverage and a patient-centered approach to care [49]. Implications for practice This research is relevant since it provides evidence of the validity and reliability of the Panamanian version of the EMCC-14 scale in specific populations of nursing, medical, dentistry, psychology and pharmacy students, which allows its application in practice. Cultural competence is fundamental in the training of students, given its positive effect on the quality of care and the satisfaction of health system users. This scale is useful in the Latin American context and in Panama, given that they are multicultural and diverse societies, where cultural competence acquires essential value for health care and attention. Similarly, the invariance of the scale in nursing and medical students allows comparative analysis of cultural competence among these groups. Importantly, there is a research gap in this area, which underscores the need for more studies that allow its implementation in practice. Limitations and future research Although this study has relevant results, it is not without limitations. First, convenience sampling was used through an online call. This type of nonprobability sampling implies limited statistical inference and generalizability to the general population of health science students [50]. Second, although the sample size was relatively large, it was not homogeneous across all health science student groups, which prevented the testing of factorial invariance across all disciplines. Establishing factorial invariance is essential, as it ensures that groups conceptualize and interpret the construct, items, and response scales in the same way [45, 46]. The presence of this attribute suggests that there is no bias between groups and that the scores obtained are the result of the presence or absence of the construct and not of the group to which they belong. Third, as a cross-sectional study, longitudinal invariance was not measured, which prevents us from demonstrating that the scale measures the same construct over time. Longitudinal invariance is useful to guarantee that the changes in the scores are attributed to real changes in the measured trait and not to alterations in the functioning of the scale [43, 45]. Fourth, this study did not examine the relationships between cultural competence and other variables previously identified in the literature, such as empathy [51, 52], social identity [10], formal training in cultural competence [19, 53], or experience and contact with culturally diverse groups [14, 52, 54]. These limitations should be considered in future studies to obtain a more complete explanation of the construct of cultural competence in health science students, its determinants and its application in practice. Future research could employ probabilistic and stratified samples, considering variables potentially related to the construct, such as academic discipline, geographic context, contact with diverse groups, and prior training in cultural competence. It would also be important to examine how cultural competence domains relate to observable behaviors. Consistent with theoretical models [55, 56], sensitivity and knowledge may shape attitudes, whereas skills may reflect past behaviors, both of which are strong predictors of subsequent behavior. Moreover, longitudinal designs are needed not only to assess the temporal stability of the scale but also to analyze the determinants of cultural competence, as intervention studies could help strengthen these competencies among health science students. Conclusions This is the first study that reports evidence of the validity and reliability of the cultural EMCC-14 scale in a population of Panamanian health science students. This study confirms that the internal structure of the original three-factor scale, cultural sensitivity, knowledge and skills, is maintained in this new sample. These results provide evidence of the usefulness of this scale for measuring cultural competence in health science students and the equivalence of the cultural competence construct among nursing and medical students, which allows future comparative studies to be carried out. Supplementary Information Supplementary Material 1 Supplementary Material 1