Survey

Information collected with the questionnaire for monitoring the implementation by UNESCO Member States of the 1974 Recommendation concerning Education for International Understanding, Co-operation and Peace and Education relating to Human Rights and Fundamental Freedoms and from 2026, the 2023 Recommendation on Education for Peace, Human Rights and Sustainable Development is used for the construction of the global indicator. For each of the four components of the indicator (policies, curricula, teacher education, and student assessment), a number of criteria are measured, which are then combined to give a single score between zero and one for each component. Only information for primary and secondary education are used for calculation of indicator 4.7.1/12.8.1/13.3.1.

Laws and policies

The following questions are used to calculate the policies component of the indicator:

A2: Please indicate which global citizenship education (GCED) and education for sustainable development) ESD themes are covered in national or sub-national laws, legislation or legal frameworks on education.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) and two levels of government (national and sub-national) = 16 responses.

Response categories are no = 0, yes = 1, unknown, which is treated as zero, and not applicable, which is ignored. Blanks are also treated as zeros.

If more than half of responses are unknown or blank the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = simple mean of the 0 and 1 scores, excluding not applicables (i.e., if eight of the 16 responses are ‘not applicable’, the sum of the 0 and 1 scores is divided by 8 to get the mean and not by 16).

A4. Please indicate which GCED and ESD themes are covered in national or sub-national education policies, frameworks or strategic objectives.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) = 8 responses.

Response categories are no = 0, yes = 1, and unknown (treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

A5. Please indicate whether national or sub-national education policies, frameworks or strategic objectives on education provide a mandate to integrate GCED and ESD.

There are two levels of government (national, sub-national) and five areas of integration (curricula, learning objectives, textbooks, teacher education, and student assessment) = 10 responses.

Response categories are no = 0, yes = 1, unknown (treated as zero), and not applicable, which is ignored. Blanks are also treated as zeros.

If more than half of responses excluding not applicables are unknown or blank, the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = simple mean of the 0 and 1 scores, excluding not applicables (i.e., if five of the 10 responses are ‘not applicable’, the sum of the 0 and 1 scores is divided by 5 to get the mean and not by 10).

E1a. Based on your responses to questions in the previous section (laws and policies) please indicate to what extent global citizenship education (GCED) and education for sustainable development (ESD) are mainstreamed^[2] in education laws and policies in your country.

There are two levels of government (national, sub-national) = 2 responses.

Response categories are not at all = 0, partially = 1, extensively = 2, unknown (treated as zero), and not applicable, which is ignored. Blanks are also treated as zeros.

If more than half of responses excluding not applicables are unknown or blank, the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = half the simple mean of the 0, 1 and 2 scores, excluding not applicables (i.e., if one of the two responses is ‘not applicable’, the sum of the 0, 1 and 2 scores is divided by 2 to get half the mean and not by 4). The score is half the mean in order to ensure it lies between 0 and 1 as do the scores for the other three questions in this section.

Policy component score = simple mean of the scores for questions A2, A4, A5 and E1a. Where a question score could not be calculated because too many responses were unknown or blank, the component score is not calculated and is reported as not available.

Curricula

The following questions are used to calculate the curricula component of the indicator:

B2: Please indicate which global citizenship education (GCED) and education for sustainable development (ESD) themes are taught as part of the curriculum.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) = 8 responses.

Response categories are no = 0, yes = 1, and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

B3. Please indicate in which subjects or fields of study GCED and ESD are taught in primary and secondary education.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) and twelve subjects in which they may be taught (arts; civics, civil or citizenship education; ethics/moral studies; geography; health, physical education and sports; history; languages; mathematics; religious education; science; social studies and integrated studies) = 96 responses.

Response categories are no = 0, yes = 1, and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank the question score is not calculated.

Note that responses to ‘other subjects, please specify’ in the question are ignored. If appropriate, during quality assurance answers in this category may be recoded to one of the other 12 subjects.

Question score = simple mean of the 0 and 1 scores.

B4. Please indicate the approaches used to teach GCED and ESD in primary and secondary education.

There are four teaching approaches (GCED/ESD as separate subjects, cross-curricular, integrated, whole school) = 4 responses

Response categories are no = 0, yes = 1, and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

E1b. Based on your responses to questions in the previous section (curricula) please indicate to what extent global citizenship education (GCED) and education for sustainable development (ESD) are mainstreamed^[3] in curricula in your country.

There are two levels of government (national, sub-national) = 2 responses.

Response categories are not at all = 0, partially = 1, extensively = 2, unknown (treated as zero), and not applicable, which is ignored. Blanks are also treated as zeros.

If more than half of responses excluding ‘not applicables’ are unknown or blank, the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = half the simple mean of the 0, 1 and 2 scores, excluding ‘not applicables’ (i.e., if one of the two responses is ‘not applicable’, the sum of the 0, 1 and 2 scores is divided by 2 to get half the mean and not by 4). The score is half the mean in order to ensure it lies between 0 and 1, as do the scores for the other three questions in this section.

Curricula component score = simple mean of the scores for questions B2, B3, B4 and E1b. Where a question score could not be calculated because too many responses were unknown or blank, the component score is not calculated and is reported as not available.

Teacher education

The following questions are used to calculate the teacher education component of the indicator:

C2: Please indicate whether teachers, trainers and educators are trained to teach global citizenship education (GCED) and education for sustainable development (ESD) during initial or pre-service training and/or through continuing professional development.

There are two types of training (initial/pre-service and continuing professional development) and two types of teachers (of selected subjects in which ESD/GCED are typically taught, and of other subjects) = 4 responses.

Response categories are no = 0, yes = 1, and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

C3. Please indicate on which GCED and ESD themes pre-service or in-service training is available for teachers, trainers and educators.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) = 8 responses.

Response categories are no = 0, yes = 1 and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

C4. Please indicate whether teachers, trainers and educators are trained to teach the following dimensions of learning in GCED and ESD.

There are four learning dimensions (knowledge, skills, values, and attitudes/behaviours) = 4 responses.

Response categories are no = 0, yes = 1, and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

C5. Please indicate whether teachers, trainers and educators are trained to use the following approaches to teach GCED and ESD in primary and secondary education.

There are four teaching approaches (GCED/ESD as separate subjects, cross-curricular, integrated, whole school) = 4 responses.

Response categories are no = 0, yes = 1 and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

E1c. Based on your responses to questions in the previous section (teacher education), please indicate to what extent global citizenship education (GCED) and education for sustainable development (ESD) are mainstreamed^[4] in teacher education in your country.

There are two levels of government (national, sub-national) = 2 responses.

Response categories are not at all = 0, partially = 1, extensively = 2, unknown (treated as zero), and not applicable (which is ignored). Blanks are also treated as zeros.

If more than half of responses excluding ‘not applicables’ are unknown or blank, the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = half the simple mean of the 0, 1 and 2 scores, excluding ‘not applicables’ (i.e., if one of the two responses is ‘not applicable’, the sum of the 0, 1 and 2 scores is divided by 2 to get half the mean and not by 4). The score is half the mean in order to ensure it lies between 0 and 1, as do the scores for the other three questions in this section.

Teacher education component score = simple mean of the scores for questions C2, C3, C4, C5 and E1c. Where component question score could not be calculated because too many responses were unknown or blank, the component score is not calculated and is reported as not available.

Student assessment

The following questions are used to calculate the student assessment component of the indicator:

D2: Please indicate whether the global citizenship education (GCED) and education for sustainable development (ESD) themes below are generally included in student assessments or examinations.

There are eight GCED/ESD themes (cultural diversity and tolerance, gender equality, human rights, peace and non-violence, climate change, environmental sustainability, human survival and well-being, and sustainable consumption and production) = 8 responses.

Response categories are no = 0, yes = 1 and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

D3. Please indicate which of the dimensions of learning in GCED and ESD below are generally included in student assessments or examinations.

There are four learning dimensions (knowledge, skills, values, and attitudes/behaviours) = 4 responses.

Response categories are no = 0, yes = 1 and unknown, which is treated as zero. Blanks are also treated as zeros.

If more than half of responses are unknown or blank, the question score is not calculated.

Question score = simple mean of the 0 and 1 scores.

E1d. Based on your responses to questions in the previous section (student assessment), please indicate to what extent global citizenship education (GCED) and education for sustainable development (ESD) are mainstreamed^[5] in student assessment in your country.

There are two levels of government (national, sub-national) = 2 responses.

Response categories are not at all = 0, partially = 1, extensively = 2, unknown (treated as zero), and not applicable, which is ignored. Blanks are also treated as zeros.

If more than half of responses excluding ‘not applicables’ are unknown or blank, the question score is not calculated.

Note that ‘not applicable’ is used where only one level of government is responsible for education.

Question score = half the simple mean of the 0, 1 and 2 scores, excluding ‘not applicables’ (i.e., if one of the two responses is ‘not applicable’, the sum of the 0, 1 and 2 scores is divided by 2 to get half the mean and not by 4). The score is half the mean in order to ensure it lies between 0 and 1, as do the scores for the other three questions in this section.

Student assessment component score = simple mean of the scores for questions D2, D3 and E1d. Where component question score could not be calculated because too many responses were unknown or blank, the component score is not calculated and is reported as not available.

The component scores all lie between zero and one and are presented as a dashboard of four scores. They are not combined to create a single overall score for the indicator. The higher the score, the more GCED and ESD are mainstreamed in the given component. In this way, users can make a simple assessment in which component area more efforts may be needed.

Greening

Curricula

Document preparation

All collected documents are added to a single database in a standardized fashion. Documents are downloaded if found online and converted to PDF if in another format. In many cases, subject curricula are part of a larger document, in which case, relevant subject- and grade-specific material are extracted into separate documents. Documents in the database are named using the following protocol:

“country_state/province_documenttype_region_year_language_grade_knowledgedomain”

Information about each document is stored in a database (one row per document), including document title, year of publication, subject, author, source, and language.

For documents in languages for which there are fewer than three documents in that language (Burmese, Norwegian, Swedish, and Urdu), the documents are machine translated into English using Google Translate.

Keyword selection and analysis

The GCI measures the inclusion of green content in four document types (NCF, grade 3 subject curricula, grade 6 subject curricula, and grade 9 subject curricula). It counts the presence of 13 keywords corresponding to three themes of Environment/Sustainability, Climate Change, and Biodiversity. The selected keywords: 1) best represent the theme, 2) can be translated into all relevant languages, and 3) are sufficiently prevalent in the analysed documents to provide data for measuring components of the GCI (see Table 2 above). Additional sources such as recent UNESCO studies of greening education and the Greening Education Partnership curriculum guidance were also used to identify relevant green keywords.

Each keyword includes its plural and singular as well as the many forms the word may take depending on the language.^{^[6]} Some languages and/or countries employ distinctive language/culture-specific keywords to capture a theme. Thus, each theme includes space for a culture- or language-specific keyword to be added, if appropriate.^{^[7]} The keywords and their translations into 40 languages are reviewed and validated by native speaking experts who are familiar with greening education concepts.

A Python-based application is used to bulk process text files and identify keywords in documents in all the required languages. To be read by the Python application, all the text documents are converted to UTF-8 text format and stored in a local folder. The Python application also requires a two-column spreadsheet with columns for "File Name" and "Language" and a second spreadsheet with columns for “Keyword” and the keyword’s “Language.” These files and the folder location are then loaded into the Python application. The application uses the language file to determine which column from the keyword file to utilize in searching for keywords for each text file. The application then counts relevant keywords in every document (NCF and subject curricula) in the specified language. After completing the keyword search processing, the application outputs a spreadsheet file that contains a row for each curriculum document and columns for every keyword.^{^[8]} This output file becomes the raw data used in the calculation of the GCI.

Calculation of the greening curriculum indicator

After the prevalence of each keyword in each document is determined, keyword counts are compiled into an output spreadsheet which is then used to calculate a country’s GCI score. The following specific steps are taken to calculate a country’s GCI score:

Phase 1) Development of standardized keyword counts

The analysis of the green content of each country’s NCFs and subject curricula is done at the country level.

For the NCF and each grade level (3, 6, and 9), the frequency of keywords belonging to the themes of Environment/Sustainability, Climate Change, and Biodiversity is calculated by summing up the counts of the keywords.
To account for varying document lengths, the number of keywords is standardized for each theme by dividing the keywords counts in that country’s theme with the total number of words in the country’s documents.
This standardized number is then multiplied by 1 million to transform the result into a number that is more easily interpreted (i.e., not a very small decimal). The result is a keyword count per million words for each theme at each grade level and NCF for each country. The standardization calculation is as follows:
- 1,000,000*(Keywords in that theme for a country) / (Total words in documents for that country)

Phase 2) Transformation of standardized keyword counts into an ordinal scale

The distribution of these standardized numbers presents a statistical challenge since it is both zero bounded^{^[9]} and has a long tail.^{^[10]}

To create a more normal distribution, the standardized numbers are transformed into an ordinal scale ranging from 0 to 10 in the following way:
- If there are no keywords, the score is 0, otherwise it ranges from 1 to 10 using a ½ life logarithmic transformation.^{^[11]}
- For the Environment/Sustainability 'core' theme, the maximum score of 10 is achieved with 10,000 standardized keywords. The following formulas are used:
  - >10,000 standardized keywords are assigned a score of 10,
  - <=20 standardized keywords are assigned a score of 1,
  - 0 standardized keywords are assigned a score of 0,
  - Otherwise, 10-log.5(#/10,000)
  - Result multiplied by 10
- For the Climate Change and Biodiversity themes, the maximum score of 10 is achieved with 5,000 standardized keywords, given that these keywords are used less commonly. The following formulas are used:
  - >5,000 standardized keywords are assigned a score of 10,
  - <=10 standardized keywords are assigned a score of 1,
  - 0 standardized keywords are assigned a score of 0,
  - Otherwise, 10-log.5(#/5,000)
  - Result multiplied by 10

Phase 3) Calculating GCIs for federated countries

To calculate the GCI for federated countries (e.g., Australia, Canada, Switzerland, United Kingdom), all of the above mentioned steps are carried out for each sub-national jurisdiction, which results in a number of (sub-national) GCIs. The sub-national GCI scores for the country are then averaged into a national GCI score. The data for all federated countries are then added to the dataset produced in Phase 1.

Phase 4) Final calculation of the GCI

At this point, each country has either three or four document-specific scores (ranging from 0 to 10) for each of the three themes (i.e., 9 or 12 total scores, since countries are included if they have at least 3 of the 4 main document types (NCF, grade 3 subject curricula, grade 6 subject curricula, and grade 9 subject curricula).

Within each of the Environment/Sustainability, Climate Change, and Biodiversity themes, the three grade level scores and the NCF score are averaged together (i.e., each contributes ¼ of the total score per theme in a country). For countries with only three document types, the same procedure is done but each document score contributes ⅓ of the total theme-focused score.
A single overall GCI score is now calculated based on a weighted mean, with the Environment/Sustainability core theme weighted 50% and the Climate Change and Biodiversity themes each weighted at 25%.

²

GCED and ESD are mainstreamed if they or their themes and sub-themes are mentioned explicitly in relevant documents and are expected to be implemented by the relevant authorities (e.g. Ministries, regional or local education authorities), educational institutions (e.g. schools, colleges and universities), and/or education professionals (e.g. teachers and lecturers), as appropriate. ↑

³

GCED and ESD are mainstreamed if they or their themes and sub-themes are mentioned explicitly in relevant documents and are expected to be implemented by the relevant authorities (e.g. Ministries, regional or local education authorities), educational institutions (e.g. schools, colleges and universities), and/or education professionals (e.g. teachers and lecturers), as appropriate. ↑

⁴

GCED and ESD are mainstreamed if they or their themes and sub-themes are mentioned explicitly in relevant documents and are expected to be implemented by the relevant authorities (e.g. Ministries, regional or local education authorities), educational institutions (e.g. schools, colleges and universities) and/or education professionals (e.g. teachers and lecturers), as appropriate. ↑

⁵

GCED and ESD are mainstreamed if they or their themes and sub-themes are mentioned explicitly in relevant documents and are expected to be implemented by the relevant authorities (e.g. Ministries, regional or local education authorities), educational institutions (e.g. schools, colleges and universities) and/or education professionals (e.g. teachers and lecturers) as appropriate. ↑

⁶

Different forms of the word are included only due to genders, definite articles, etc. but not when they change the meaning or part of speech. ↑

⁷

For example, in China the phrase ‘ecological civilization’ is now being used much more frequently than ‘sustainable development’ or ‘environmental.’ In Japan, the term ‘sustainable societies’ is becoming more prevalent than the term ‘sustainable development.’ At this point in time, no culture- or language-specific keywords are included in the GCI. ↑

⁸

To determine the accuracy of the counts generated by the Python program, a validation exercise was carried out in October 2024 by sampling 30 documents in English, Spanish, Arabic and French, the four most prevalent languages. A three-way comparison of results from NVivo (the software used for all related UNESCO consultancies), Python, and manual counts identified several minor issues (e.g., keywords split across lines or the lack of a definite article in the Arabic keyword list), which were immediately corrected in the Python program and the keyword list. Since then, the Python program has been reviewed by several experts and undergone further refinements to ensure its counting accuracy is comparable to NVivo and manual counting. ↑

⁹

While there are many documents lacking any keywords related to Environment/Sustainability, Climate Change and Biodiversity, there are no documents with a negative number of keywords. Such a situation represents a zero-bounded distribution and creates a lopsided and non-normal distribution. ↑

¹⁰

While more than half the document types have less than 120 standardized keywords in a theme, they range to over 9,000 (75+ times as much as the median). Log transformations are conceptually useful when dealing with such data. For example, going from 0 to 50 standardized keywords is more significant than going from 1000 to 1050 standardized keywords. ↑

¹¹

This means that for each time the standardized counts are halved, the score goes down by 1. So, for example, if 10,000 standardized references is a score of 10, 5,000 is a score of 9, 2,500 is a score of 8, and so on. ↑

Hesin internet-kagin er ikki longur stuðlaður

Rætta uppseting: 13.3.1