Development and psychometric properties of a questionnaire on environmental literacy in Indonesian secondary school.

Páginas: 267-282 Recibido: 2022-12-06 Revisado: 2022-12-14 Aceptado: 2023-03-25 Preprint: 2023-04-21 Publicación Final: 2023-09-01 www.revistascientificas.us.es/index.php/fuentes/index DOI: https://doi.org/10.12795/revistafuentes.2023.22642 Development and psychometric properties of a questionnaire on environmental literacy in Indonesian secondary school Desarrollo y propiedades psicométricas de un cuestionario sobre alfabetización ambiental en la escuela secundaria de indonesia Mujib Mujib Universitas Islam Negeri Raden Intan Lampung (Indonesia) Mardiyah Mardiyah Universitas Islam Negeri Raden Intan Lampung (Indonesia) Suherman Suhermen University of Szeged (Hungary) Abstract Concurrently, the amount of research dedicated to mitigating the effects of environmental issues is increasing. Environmental Education has emerged as one of the most promising research fields in preventing and addressing environmental problems. Nevertheless, it is important to note that environmental literacy is not limited by age, education level, or occupation, particularly in the Indonesian context. Therefore, this study aimed to evaluate the psychometric properties of an environmental literacy questionnaire for secondary school students. The sample consisted of 1,021 students from both public and private schools in Indonesia, ranging from grades 7-9. A three-factor model, including environmental consciousness, environmental anxiety, and environmental awareness, was validated using confirmatory factor analysis (CFA). Model fit was assessed using a variety of fit indices, such as the chi-squared test, comparative fit index (CFI), Tucker-Lewis Index (TLI), Good-Fit Index (GFI), root mean square approximation error (RMSEA), standardized root mean square residual (SRMR), and KMO index. Cronbach Alpha and McDonald's Omega were used to evaluate the reliability of the three factors, with values ranging from 0.66 to 0.78 and 0.68 to 0.79, respectively. The results indicate that the environmental literacy questionnaire is a reliable and valid tool for assessing the environmental literacy of secondary school students in the Indonesian context. Resumen Al mismo tiempo, la cantidad de estudios sobre la mitigación de los efectos de los problemas ambientales está creciendo. La Educación Ambiental es uno de los campos de investigación más prometedores en términos de prevención y consecuencias de los problemas ambientales. Sin embargo, la alfabetización ambiental no está limitada por la edad, el nivel educativo o la ocupación, especialmente en el contexto indonesio. Este estudio tuvo como objetivo evaluar psicométricamente un cuestionario de alfabetización ambiental para la educación secundaria. La muestra del estudio estuvo compuesta por 1.021 estudiantes de 7º a 9º grado de escuelas secundarias seleccionadas de escuelas públicas y privadas en Indonesia. La validez de la estructura de tres factores del modelo fue evaluada mediante análisis factorial confirmatorio (CFA), a saber, la conciencia ambiental, la ansiedad ambiental y la conciencia ambiental. Después del CFA, se utilizaron índices de ajuste para evaluar el ajuste del modelo, como la prueba de chi-cuadrado, el índice de ajuste comparativo (CFI), el índice de Tucker-Lewis (TLI), el índice de buen ajuste (GFI), el error de aproximación de la raíz cuadrada media (RMSEA), el residuo cuadrático medio estandarizado (SRMR) y el índice KMO. Los valores de confiabilidad para los tres factores oscilaron entre 0,66 y 0,78 con Alfa de Cronbach, mientras que los de Omega de McDonald oscilaron entre 0,68 y 0,79. El cuestionario de alfabetización ambiental es un instrumento viable para evaluar la alfabetización ambiental de los estudiantes de secundaria en el contexto indonesio. Palabras clave / Keywords alfabetización ambiental, escuela secundaria, validación psicométrica, análisis factorial confirmatorio, mapa de Wright, psicométrico, educación, desarrollo environmental literacy, secondary school, psychometric validation, confirmatory factor analysis, wright map, psychometric, education, development First Online 268 ··· REVISTA FUENTES, 25(3); Páginas 267-282 www.revistascientificas.us.es/index.php/fuentes/index 1. Introduction Every day, more and more individuals are becoming aware of and affected by environmental challenges. The media regularly reports on the repercussions of global warming and ozone layer depletion, which are felt worldwide. Meanwhile, research into mitigating the effects of environmental issues is on the rise. Some studies aim to eliminate the effects of environmental problems, while others concentrate on preventing them from occurring in the first place, as this approach is simpler and more effective. Environmental Education is one of the most promising research fields for preventing environmental problems and their consequences. Human actions are primarily responsible for environmental issues (Saribas et al., 2017), and therefore humans must be responsible for solutions and processes (Suherman et al., 2020). Thus, boosting environmental awareness and concern among the population is crucial in finding solutions. Environmental education, a term first introduced at the Tbilisi Congress in 1977, can be defined as the process of educating individuals to obtain information and skills necessary to improve motivation and attitudes towards proposing ideas for solutions to environmental problems (UNESCO, 1978). Environmental education is important as it leads to a higher level of environmental literacy among individuals (Atabek-Yiğit et al., 2014). An environmentally literate person, according to Janmaimool & Khajohnmanee (2019a), understands the relationship between nature and social systems, acknowledges human attitudes and the impact of technological development on the environment, and is aware that environmental knowledge can be acquired throughout life. Knowledge, attitude, behaviour, and consciousness are the four components of environmental literacy, all of which an individual must possess to be deemed environmentally literate (Harring & Jagers, 2018). Students can learn environmental literacy through their activities (Arsyad & Villia, 2022; Corebima et al., 2017). The foremost way to address environmental issues is through environmental education and environmental literacy. From this perspective, it is evident that environmental literacy must be fostered. Environmental literacy in Indonesia is a growing concern as the country faces various environmental issues, such as deforestation, air pollution, and plastic waste, among others. According to a study conducted by Sudarmadi et al. (2001), Indonesian students lack knowledge and understanding of environmental issues, resulting in low environmental literacy levels. The study involved 434 high school students from three different regions in Indonesia, and the results showed that the average score for environmental literacy was only 46.2 out of 100. One of the main problems with environmental literacy in Indonesia is the lack of education and awareness about environmental issues. According to a report by the Indonesian Ministry of Environment and Forestry, only a few schools in the country integrate environmental education into their curricula (UNESCO, 2016). Furthermore, the level of environmental literacy among students is low, and there is a lack of teacher training in environmental education (Safitri et al., 2020). The study also noted that the current environmental education curriculum in Indonesia is insufficient to promote environmental literacy. Another issue is the lack of public awareness and participation in environmental protection. Many young people in Indonesia do not identify themselves as environmentalists and often lack the knowledge, resources, and opportunities to act on their attitudes in their daily lives (Parker et al., 2018). Additionally, there is a lack of public participation in environmental decision-making processes, such as environmental impact assessments and public hearings, which hinders the implementation of effective environmental policies and regulations (Swangjang, 2018). In summary, the problems of environmental literacy in Indonesia include the lack of education and awareness about environmental issues, shortage of qualified teachers, inadequate funding and resources, lack of public awareness and participation, and inadequate implementation of environmental policies and regulations. These issues require urgent attention and action from the government, education sector, and civil society to improve environmental literacy and promote sustainable development in Indonesia. Research has been conducted on the development and validation of environmental literacy at various scales. For instance, Erdoğan (2003) developed a scale to assess the attitudes of college students toward environmental issues, and Dunlap et al. (2000) enhanced the New Environmental Paradigm scale. Berberoglu & Tosunoglu (1995) created the Environmental Attitude Scale (EAS) after conducting a validity and reliability study with Turkish university students, identifying four variables and forty-seven items. Teksöz et al. (2010) originally created the Environmental Literacy Scale at Michigan State University, and Leeming et al. (1995) established the Children's Environmental Attitudes and Knowledge Scale. Alp et al. (2006) administered a Turkish version of the measure to sixth, eighth, and tenth-grade students to assess their environmental attitudes and knowledge, as well as the influence of grade and gender on environmental attitudes. Naim & Sağlam (2006) created the Environmental Attitude Scale for high school students. First Online 269 ··· REVISTA FUENTES, 25(3); Páginas 267-282 DOI: 10.12795/revistafuentes.2023.22642 Although research on environmental literacy has been conducted at different levels, it is not limited by age, education level, or occupation, as protecting the environment is everyone's responsibility. Nurhidayati et al. (2022) even developed a scale to measure students' concern for the local potential of natural resources in their environment. However, most of the previous research was conducted on university students, prospective teachers, and elementary or high school students, and developed in Western and Middle Eastern countries to assess their attitudes, knowledge, or environmental literacy. This study aims to develop an inventory of environmental literacy using Confirmatory Factor Analysis (CFA) in junior high school students in the Indonesian context, which is a novel contribution to the existing literature. 2. Literature Review 2.1. Environmental Literacy The definition of environmental literacy states that an environmentally literate person has a basic skill, understanding, and feeling about the human-environment relationship. Such a person must comprehend the interrelationship between natural and social systems, the unity of humans and nature, how technology affects decision-making on environmental issues, and that learning about the environment is a lifelong endeavour. According to Cristovão et al. (2022), environmental literacy refers to "the knowledge, skills, values, and attitudes necessary to understand and participate in activities aimed at solving environmental problems and creating sustainable societies" (p. 77). It requires a deep understanding of environmental issues, their causes, and potential solutions, as well as the ability to critically evaluate and communicate environmental information. Environmental literacy also involves a sense of responsibility and motivation to take action to protect the environment and promote sustainable practices. Thus, "the need for capable awareness, knowledge, skills, and attitude funds to incorporate appropriate environmental considerations in making decisions for consumption, lifestyle, career, and citizenship both individually and in groups." Kelani argues that "Environmental literacy is the 'knowledge necessary to comprehend relatedness, and an attitude of care or stewardship” (Kelani, 2017). Additionally, "Environmental literacy is essentially the capacity to perceive and interpret the relative health of the environmental systems and to take appropriate action to maintain, restore or improve the health of those systems" (Lloyd-Strovas et al., 2018). It can be concluded from these views that environmental literacy is a skill possessed by a person or people in understanding and interpreting action and then acting efficiently and effectively in accordance with the environmental context. 2.2. Assessment of an Environmental Literacy There are numerous dimensions to the assessment tools used to measure environmental literacy. In this perspective, we have developed three dimensions: environmental consciousness level, environmental anxiety level, and environmental commitment level. The environmental consciousness level refers to personal environmental awareness demonstrated by support for environmental sustainability based on values centered around a philosophy of life that prioritizes environmental values and the natural environment (Arifin et al., 2021). Examples of instruments in this category include statements like "I believe that the government should support renewable energy sources (solar, wind, water, geothermal)," "Environmental education should be provided from the beginning of basic education to promote environmental awareness," and "I will use a recycling box if available" (Atabek-Yiğit et al., 2014). The environmental anxiety level can be described as a range of subjective and usually unpleasant sensations regarding environmental problems (Gao et al., 2021), such as worry, stress, and anger (Gao et al., 2021). Indicators of this level include statements like "I think we will not find a place for a picnic in a few generations," "I think everyone should plant a tree in their life," and "I think plants should be saved for future life" (Atabek-Yiğit et al., 2014). Finally, the environmental commitment level refers to an individual's dedication and engagement in pro-environmental behaviors and practices, such as reducing energy consumption, recycling, and reducing waste. It encompasses attitudes, beliefs, and behaviors that contribute to sustainable living and environmental protection (Grilli & Curtis, 2021). Indicators of this level include statements like "I want to learn about environmental issues," "I prefer to buy environmentally friendly goods rather than economical ones," and "I prefer to use public transportation rather than private transportation to protect the environment" (Atabek-Yiğit et al., 2014). First Online 270 ··· REVISTA FUENTES, 25(3); Páginas 267-282 www.revistascientificas.us.es/index.php/fuentes/index 3. Method The study was conducted on secondary school students in Lampung province, Indonesia. Sampling for the study used random sampling. The sample of 1021 consisted of grades 7 to grade 9 with an age range of 11- 16 years (M = 13.69; SD = 0.963). The research was conducted using a google form, due to the pandemic situation. The characteristics of the research sample are as follows: Table 1 Characteristics of the research sample Characteristics Frekuensi n % Gender Female 623 61.3 Male 395 38.7 Grade 7 294 28.8 8 394 38.6 9 333 32.6 School Type Public 332 32.5 Private 689 67.5 Place Type City 384 37.6 District 637 62.4 Ethnic Java 633 62.0 Lampung 206 20.2 Sunda 24 2.4 Batak 54 5.3 Padang 22 2.2 Bugis 3 0.3 Others 79 7.7 3.1. Instrument This study adapted the ELSA instrument developed by Atabek-Yiğit et al. (2014). This instrument was developed in the Indonesian context with 3 scales/inventories namely Environmental Consciousness Level of 9 items, Environmental Anxiety Level of 6 items, and Environmental Commitment level of 5 items with a total of 20 items of questions. Alternative responses use five Likert scales, including strongly agree with a value of 5 to strongly disagree with value of 1. There are some instruments to evaluate environmental literacy. The instruments are in Table 2. Table 2 Instruments of environmental literacy Study Instrument Number of items Psychometric properties (Atabek-Yiğit et al., 2014) Environmental Literacy Scale for Adults (ELSA) 29 items Cronbach-Alpha for the whole scale = 0.881; 1st dimension = 0.807; 2nd dimension = 0.765; 3rd dimension = 0.715. (N = 332) (Lloyd-Strovas et al., 2018) Environmental Literacy Instrument (ELI) The ELI instrument consists of 40 items, with 10 items in each of the four domains: knowledge, attitudes, behaviors, and skills. Cronbach's alpha coefficients for the environmental knowledge, attitudes, and behaviors subscales were 0.84, 0.82, and 0.71, respectively. (Tuncer Teksoz et al., 2014) ELQ: environmental literacy questionnaire The items were 12 assess four components (knowledge, attitude, N = 648. Data is valid and reliable based on Rasch measurement. First Online 271 ··· REVISTA FUENTES, 25(3); Páginas 267-282 DOI: 10.12795/revistafuentes.2023.22642 Study Instrument Number of items Psychometric properties attitude towards environmental responsibility and concern). (Negev et al., 2008) The Middle School Environmental Literacy Instrument Three dimensions including environmental knowledge, attitudes, and behavior. N = 3121. Validity and reliability used Cronbach’s (Szczytko et al., 2019) The Environmental Literacy Instrument for Adolescents (ELI-A) Three dimension, namely, Ecological knowledge, Hope, Behaviour N = 665. Data analysis using CFA and SEM. The above researchs lend support to the construct validity as a measurement of the scale. However, there is a paucity of research into environmental litracy in Indonesia, particularly in the context of secondary education. Recent research on environmental literacy has primarily focused on adolescents, teachers, and western populations (De Leyn et al., 2022). Unfortunately, we didnt see of the environmental literacy inventory in the Indonesian context. This is prior reasons that we need develope and looking for psychometric properties of instrument in Indonesian context. 3.2. Research Design The research has 3 stages, the first is to translate the original English questionnaire by experts, in this case it will be validated by mathematics and language experts, namely Ph.D. students in the UK, Japan, and Hungary. The next step is to distribute the questionnaire to junior high school students in Lampung province. The final stage is to analyze the questionnaire results using CFA. 3.3. Data Analysis Data analysis was conducted using SPSS Version 26.0 and Jeffreys's Amazing Statistics Program (JASP) version 0.14.1. Confirmatory Factor Analysis (CFA) was used to check the model fit in the measurement model (Jomnonkwao & Ratanavaraha, 2016), and JASP version 0.14.1 was used for the analysis. Following CFA, several fit indices were used to evaluate the model fit, including the chi-squared test, comparative fit index (CFI), Tucker-Lewis Index (TLI), Good-Fit Index (GFI), root mean square approximation error (RMSEA), standardized root mean square residual (SRMR), and KMO index (Kline, 2015). The chi-square statistic, including degrees of freedom and p-values, is represented mathematically. According to Kline (2015), the “Chisquare test statistic” is extremely sensitive to sample size, with statistically significant chi-square values being more frequently observed with larger samples. Therefore, the CFI value, which is insensitive to sample size, was considered. Values greater than 0.90 indicate that the model is fit and acceptable. Furthermore, a GFI value higher than 0.85 indicates a good fit (Hair et al., 2010), while an RMSEA with a range between 0.03 and 0.08 is considered a good model fit. To assess reliability, alpha reliability and composite Cronbach alphas were examined. “Internal consistency (Crbα; Cronbach's alpha)” and “composite reliability (ω; McDonald's omega coefficient; (Raykov, 1997)” were used as measures of reliability. As previously stated by Habók & Magyar (2018), values greater than 0.70 indicate favourable results for empirical research. In addition, construct reliability (CR) should be > 0.70, and the average variance extracted (AVE) should be more than 0.50. Lower values are acceptable when the CR value is greater than 0.60; however, “lower values are not acceptable when the CR value is less than 0.60” (Fornell & Larcker, 1981). Discriminant validity was assessed using HTMT, with a threshold value of 0.85 being acceptable (Kline, 2015). 4. Results 4.1. EFA EFA is commonly used when the relationship between observed instrument variables is unclear (Glynn et al., 2011). As ELSA is a three-factor instrument specifically designed for an Indonesian context and based on aspect-based factors, EFA was used to analyze the students' responses to the questionnaires in this study. The results of the EFA were consistent with the average Bartlett's test of sphericity (Chi-square = 779.719; DF = 132; p < 0.001) and the KMO measure of sampling adequacy (KMO = 0.886), which indicated that the instrument was reliable and provided a high-quality sample for further analysis. These results were consistent with previous research (Field, 2013; Rytkönen et al., 2007; Suherman & Vidákovich, 2022a). First Online 272 ··· REVISTA FUENTES, 25(3); Páginas 267-282 www.revistascientificas.us.es/index.php/fuentes/index 4.2. Reliability Reliability measures the internal consistency of respondents' answers across questions in an instrument. Since all the items in this research instrument are intended to capture the same underlying construct, respondents' scores should be correlated with each other (Wieland et al., 2017). Cronbach's alpha is used to assess internal consistency, while McDonald's Omega is used to determine composite reliability. The results of the analysis are presented in the following table: Table 3 Consistency reliability and composite reliability Factors 𝜶 𝝎 Environmental Consciousness Level 0.77 0.78 Environmental Anxiety Level 0.78 0.79 Environmental Commitment Level 0.66 0.68 Note. 𝛼 , Cronbach’s alpha; 𝜔 , McDonald’s coefficient omega The results of Table 3, show that the Cronbach Alpha range is 0.66 to 0.78 while at McDonald's Omega, it is in the range of 0.68 and 0.79 for the three factors. This indicates that within these ranges, reliability is acceptable. 4.3. Construct Validity 4.3.1. Convergent Validity Convergent validity is a measure of the degree of correlation between multiple variables within the same construct of an instrument. This means that convergent validity is established when the variables within a factor are strongly related. To achieve convergent validity, we need to evaluate the convergent reliability (CR), factor loadings, and average variance extracted (AVE) (Ab Hamid et al., 2017). Typically, as the sample size decreases, the loading score requirement increases. However, regardless of sample size, it is recommended to have loading scores greater than 0.5 for each element. Each composite factor's AVE threshold should be better than 0.5, and the CR threshold should be greater than 0.70 (Hair Jr et al., 2021). Even if the AVE value is less than 0.5 and the CR is greater than 0.6, the convergent validity of the construct still meets the minimum requirements (Fornell & Larcker, 1981; Malhotra & Dash, 2011). The AVE and CR values were obtained using the major validity instrument (Gaskin & Lim, 2016), and EFA was used to calculate the factor loadings. The results of the item loading score values determined by the convergent validity test using CR and AVE are displayed in Table 4. Table 4 Convergent validity Factors CR AVE Environmental Consciousness Level 0.80 0.35 Environmental Anxiety Level 0.79 0.39 Environmental Commitment level 0.69 0.37 Table 4 shows that the CR scores for the questionnaire's three latent components range from 0.69 to 0.80 respectively, while AVE values range between 0.35 and 0.39. Although there are low AVE values, the convergent validity and reliability can still be established based on CR alone, even if AVE is often too strict. This is because AVE is a measure of the degree of correlation between two variables (Fornell & Larcker, 1981; Malhotra & Dash, 2011). 4.3.2. Discriminant Validity Discriminant validity is used to assess the extent to which latent factors differ from each other empirically (Hair et al., 2010; Henseler et al., 2015). In this study, we also used a new criterion for assessing discriminant validity, the Hetero Trait Mono Trait (HTMT). Conceptually and practically, the HTMT value threshold should be below 0.9 and 0.85, respectively (Henseler et al., 2015). Table 5 below presents the results of the study. First Online 273 ··· REVISTA FUENTES, 25(3); Páginas 267-282 DOI: 10.12795/revistafuentes.2023.22642 Table 5 Rasio with HTMT0.85 ECoL EAnL ECL ECoL - 0.66 0.63 EAnL - 0.66 0.63 ECL - 0.84 ECol: Environmental Consciousness Level; EanL: Environmental Anxiety Level; ECL: Environmental Commitment level Table 5 explains that the value of HTMT0.85 establishes discriminant validity less than 0.85. The range of values is 0.66 to 0.84. So discriminant validity is accepted less than 0.85 (Hair et al., 2010; Henseler et al., 2015). 4.4. CFA The aim of this study is to evaluate the measurement model using CFA and JASP. CFA serves to confirm the latent factors in the measurement model, ensuring their adequate operation and achieving the GoF index. This helps researchers identify relationships between latent factors and develop hypotheses in subsequent studies with greater confidence (O’Byrne et al., 2018). To ensure the level of quality, we conducted analyses to determine CR, convergent validity, and discriminant validity, following the recommendations of Chuah et al. (2016). Utilizing the pattern matrix builder plugin developed by Tabachnick et al. (2007), we constructed CFA diagrams in the measurement model to evaluate model fit. The CFA results in the table below confirmed the appropriateness of the model Table 6 Model fit Index Value CFI 0.933 TLI 0.922 RMSEA 0.069 SRMR 0.077 GFI 0.962 Table 6 presents the analysis findings, which indicate that the CFI value is 0.933, the TLI value is 0.922, the RMSEA value is 0.069, the SRMR value is 0.077, and the GFI value is 0.962. To further improve the model fit using CFA, we analyzed the modification indices and identified covariance between items within the same factor with values greater than 0.30. As recommended by Fornell & Larcker (1981), the most suitable modification for the measurement model was the covariation of error terms within the same factor. Table 7 displays the factor loading fit. Table 7 Factors loading of the items Dimentions Items Loading Factors Environmental consciousness level I believe that government should support the renewable energy sources (sun, wind, water, geothermal). 0.532 I, as well as others, have responsibility for the protection of the environment. 0.510 I’m in favour of using solar power in traffic lights and street lamps in order to keep the future generations’ life. 0.559 I’m in favour of using energy sources like solar power and natural gas since the gases given out from stoves are more harmful. 0.470 I would use recycling boxes if there were any. 0.564 I would use e-bill in order to protect the environment. 0.602 I would throw away my garbage if there were nobody there. 0.587 First Online 274 ··· REVISTA FUENTES, 25(3); Páginas 267-282 www.revistascientificas.us.es/index.php/fuentes/index Dimentions Items Loading Factors There is nothing wrong with pouring waste cooking oil into the sink. 0.372 Environmental anxiety level I think we will not find a place to have picnic within a few generation. 0.573 I think everybody should sow a tree in his or her life. 0.570 I think seeds should be kept for the future of life. 0.689 I would throw old newspapers; empty glass-plastic bottles, and cans to recycling boxes. 0.752 I think indiscriminate hunting can cause environmental problems. 0.433 I would warn people if they caused harm to the environment. 0.751 Environmental awareness level When I read a newspaper I pay attention to the topics related to the environment. 0.636 For the protection of environment caused by waste, I watch TV programs that give information about re-use of them. 0.691 I would rather buy environmentally friendly items than economic ones. 0.537 I prefer to use public transportation rather than private transportation to protect the environment. 0.398 Additionally, Figure 1 displays a more accurate model fit. Figure 1 presents an illustration of the CFA diagram that follows the modification index and provides details about the GoF values. The questionnaire has achieved outstanding measurement criteria, as determined by Soeharto & Csapó (2021), who defined the cutoff requirements for fit indices in covariance structure analysis. 5. Discussions This study was conducted with the objective of adapting and validating an environmental literacy instrument for the secondary school student population in the Indonesian context. The results indicate that various statistical procedures were performed, leading to the modification of the model through CFA analysis, and ultimately validating the designed measuring instrument. As a result, the study was conducted using statistical procedures to ensure the validity of the instrument. In this study, the researchers developed a 20-item questionnaire to measure the environmental literacy level of junior high school students. The questionnaire was subjected to exploratory factor analysis, which revealed three variables for the measure. The KMO value of 0.886 indicated that the instrument was able to accurately distinguish the three latent factors in the questionnaire. However, two of the twenty items were removed from the questionnaire due to their factor loading values being below 0.3. These items were "Environmental education should be provided from the beginning in order to provide environmental awareness" and "I want to learn about environmental issues". The results from graph 1, which is based on an ethnic/tribal sample in Indonesia, suggest that there is no bias among the participants towards filling in the items of the questionnaire. The sample includes individuals from Javanese (1), Lampung (2), Sundanese (3), Batak (4), Padang (5), Bugis (6), and other (7) ethnicities. This finding indicates that the questionnaire is suitable for use with diverse groups in Indonesia. Furthermore, the level of environmental literacy among the students can be observed from Figure 2, which shows the distribution of their scores. This information is useful in understanding the strengths and weaknesses of the students' environmental literacy and can inform the development of targeted interventions to improve their understanding of environmental issues. The level of environmental literacy possessed by individuals or organizations can help to determine their perspective on environmental issues. It is widely known that one approach to solving environmental problems is to develop innovative technologies and solutions, typically after problems have already arisen. Another approach is to educate the general public to prevent the manifestation of these issues. Individuals with high environmental literacy are better equipped to utilize the latter approach. Additionally, research has shown that children tend to model their behavior after their parents, so environmentally conscious parents are more likely to have environmentally conscious children. Therefore, it is crucial to assess a person's level of environmental literacy. While there are some studies on environmental issues in the global literature, there are relatively few First Online 275 ··· REVISTA FUENTES, 25(3); Páginas 267-282 DOI: 10.12795/revistafuentes.2023.22642 studies that specifically explore the perspectives of students on this topic (Fujitani et al., 2017; Kaya & Elster, 2018; Nunez & Clores, 2017). Morrone et al. (2001) have provided research findings that can contribute to environmental literacy research by presenting a tested and validated survey instrument to measure ecological knowledge, which is one of the components of environmental literacy. In a study conducted by Erdogan & Ok (2011) to measure the environmental literacy level of secondary school students in Turkey, the results showed that 61% of students had a moderate level of environmental literacy despite using a literacy questionnaire. Atabek-Yiğit et al. (2014) developed ELSA to determine the environmental literacy of adults. Figure 1. CFA Model Fit (N = 1021) First Online 282 ··· REVISTA FUENTES, 25(3); Páginas 267-282 www.revistascientificas.us.es/index.php/fuentes/index Tabachnick, B. G., Fidell, L. S., & Ullman, J. B. (2007). Using multivariate statistics (Vol. 5). Pearson Boston, MA. Teksöz, G., Şahin, E., & Ertepinar, H. (2010). Çevre okuryazarlığı, öğretmen adayları ve sürdürülebilir bir gelecek. Hacettepe Üniversitesi Eğitim Fakültesi Dergisi, 39(39), 307–320. Tuncer Teksoz, G., Boone, J. W., Tuzun, O. Y., & Oztekin, C. (2014). An evaluation of the environmental literacy of preservice teachers in Turkey through Rasch analysis. Environmental Education Research, 20(2), 202–227. https://doi.org/10.1080/13504622.2013.768604 UNESCO. (2016). Global Education Monitoring Report 2016: Education for People & Planet: Creating Sustainable Futures for All. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000245656 UNESCO, T. D. (1978). Intergovernmental conference on environmental education. Final Report. Paris. Vergara Morales, J. R., Díaz Larenas, C. H., Tagle Ochoa, T., & Ortiz Navarrete, M. (2022). Adaptación y Validación de la Escala de Percepción del Clima de Evaluación en el Aula (EPCEA). Revista Fuentes. https://doi.org/10.12795/revistafuentes.2022.19232 Wieland, A., Durach, C. F., Kembro, J., & Treiblmaier, H. (2017). Statistical and judgmental criteria for scale purification. Supply Chain Management: An International Journal, 22(4), 321–328. https://doi.org/10.1108/SCM-07-2016-0230 Yeh, F.-Y., Tran, N.-H., Hung, S. H., & Huang, C.-F. (2021). A study of environmental literacy, scientific performance, and environmental problem-solving. International Journal of Science and Mathematics Education, 1–23. https://doi.org/10.1007/s10763-021-10223-9