Results from the Teacher Knowledge Survey

Annex A. The development of the Teacher Knowledge Survey

Pedagogical knowledge is a fundamental component of teachers’ professionalism and effectiveness. It forms a specialised body of knowledge whose relevance is unique to the teaching profession, as it underpins teachers’ decision-making about how to help students to learn. One could argue that pedagogical knowledge is essentially what differentiates a teacher from a professional who is simply very competent on a subject.

The body of research about the importance of teacher knowledge has been growing for several decades but international studies that directly assess pedagogical knowledge have been limited (Weyers, Ligtvoet and König, 2024[1]; König et al., 2011[2]). The Teacher Knowledge Survey (TKS) has been developed to improve our understanding of teachers’ pedagogical knowledge base.

TKS focuses on general pedagogical knowledge (GPK), which is general to all teachers, as opposed to pedagogical content knowledge, which is knowledge about teaching and learning a specific subject. TKS stands as the first international large-scale assessment of teachers’ general pedagogical knowledge, using representative samples of teachers across participating countries (Ulferts, 2021[3]).

TKS 2024 was conducted as an optional module within the fourth cycle of TALIS (TALIS 2024) in eight countries: Chile, Croatia, Morocco, Poland, Portugal, Saudi Arabia, South Africa, and the United States. The Survey includes both a contextual questionnaire and a direct assessment of GPK. The data collected in the Survey can help policymakers and other stakeholders answer questions such as:

• How broad is teachers’ general pedagogical knowledge base?

• Which teachers tend to have more general pedagogical knowledge?

• What system- and school- level resources are associated with having greater knowledge of general pedagogy?

• How does general pedagogical knowledge relate to teachers’ practices?

TKS was designed as an optional module that countries participating in the 2024 Teaching and Learning International Survey (TALIS 2024) could decide to administer together with the core TALIS survey. To maximise synergies between the two surveys, teachers participating in TKS were randomly drawn from the pool of teachers teaching at ISCED Level 2 in the same schools that were also sampled for the TALIS core survey.1 There are therefore also opportunities to analyse TKS results alongside data from principals and other teacher colleagues in the same school.

The TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework (OECD, 2025[4]) presents a detailed overview of the theoretical foundations upon which the TKS instruments were constructed. The development of TKS 2024 largely relied on a pilot study conducted in five countries by the OECD’s Centre for Educational Research and Innovation (CERI) as part of its Innovative Teaching for Effective Learning (ITEL) project (Guerriero, 2017[5]; Sonmark et al., 2017[6]). Development of the TKS 2024 assessment and questionnaire was led by ACER (under a contract with the OECD), with the support and guidance of an international group of experts on teacher knowledge (See Annex F List of contributors).

Defining general pedagogical knowledge

Following Shulman’s (1987[7]) seminal work on conceptualising teacher knowledge into seven different types, most scholars have identified three categories of knowledge as key to the work of teachers: content knowledge (CK), pedagogical content knowledge (PCK) and general pedagogical knowledge (GPK) (Leijen et al., 2022[8]; Baumert et al., 2010[9]). Shulman’s original definition of GPK refers to the “broad principles and strategies of classroom management and organisation that appear to transcend subject matter” (Shulman, 1987, p. 8[7]). Other authors continue to highlight the “cross-curricular” or “generic” nature of GPK, which is ”shared by teaching professionals across disciplines and educational levels” (Ulferts, 2021[3]; Guerriero, 2017[5]; Voss, Kunter and Baumert, 2011[10]). Subsequent definitions of GPK have, however, often been extended to include topics beyond classroom management and organisation, such as learning processes, student motivation or teaching methods (OECD, 2025[4]).

The conceptualisation of general pedagogical knowledge (GPK) for TKS 2024 focuses on its relevance for all teachers regardless of their subject. Thus, GPK is identified as “the specialised knowledge of teachers for creating effective teaching and learning environments for all students, independent of subject matter”, as defined in the OECD ITEL pilot study (Guerriero, 2017, p. 80[5]).

Based on the research conducted for the ITEL pilot study (König, 2014[11]), GPK is conceived as a multi-faceted construct, which includes the following dimensions:

• instruction: including teaching methods, didactics, structuring a lesson, and classroom management;

• student learning: including their cognitive, motivational, emotional individual dispositions; their learning processes and development; their learning as a group taking therefore into account student heterogeneity and adaptive teaching strategies;

• assessment: including diagnosing principles irrespective of the subject, evaluation procedures.

General pedagogical knowledge and teachers’ professional competence

While general pedagogical knowledge (GPK) can be considered as a core competence for teachers, it interacts with other factors that are also essential for the provision of effective teaching. The CERI framework of teachers’ professional competence illustrates how (general) pedagogical knowledge operates alongside content knowledge and affective-motivational competencies and beliefs in teachers’ selection and implementation of teaching approaches (Figure A A.1). Together, these shape students’ learning experience and outcomes. In turn, research into student learning and teachers’ experiences in the classroom support further opportunities for teachers to learn about pedagogy and thereby the development of teachers’ knowledge base.

Figure A A.1. CERI Conceptual framework of teachers’ professional competence

Source: Guerriero, S. and N. Révai (2017[12]), “Knowledge-based teaching and the evolution of a profession”, in Pedagogical Knowledge and the Changing Nature of the Teaching Profession, http://dx.doi.org/10.1787/9789264270695-13-en, Figure 11.2.

The development of the TKS 2024 instruments

Ensuring that TKS provides a valid and internationally comparable measurement of GPK was a core concern in the development of the survey instruments. Three methods of validation were used in the development of the TKS conceptual and assessment frameworks:

• reviews of teaching models, frameworks and national teaching standards;

• support by the specifically established Teacher Knowledge Expert Group (TKEG);

• expert feedback from participating countries (OECD, 2025[4]).

Review of national teaching standards

To examine the cross-cultural validity of the TKS assessment framework, teacher professional standards from Chile, Croatia, South Africa and the United States were analysed. This was an important step to confirm the relevance of the theorised dimensions and categories of GPK across diverse countries, since previous literature on the topic was based largely on studies of Western, English-speaking or European countries (König, 2014[11]; Toledo-Figueroa, Révai and Guerriero, 2017[13]). The following documents containing teacher professional standards were reviewed, as these were considered likely to reflect the professional knowledge that teachers in each country were expected to have:

• Council for Educators: Professional Teaching Standards (South Africa)

• National Council for Education and Training: Framework of National Qualification Standards for Teachers in Primary and Secondary Schools (Croatia)

• Council of Chief State School Officers: InTASC Model Core Teaching Standards (United States)

• Center for Improvement, Experimentation and Pedagogical Research: Framework for Good Teaching (Chile)

This validation analysis provided support for the use of the TKS GPK assessment framework, as every sub-dimension from the framework was reflected in each set of the teacher professional standards at least once if not multiple times. All sets of professional standards also included reference to topics relating to the emergent priorities for teacher knowledge, as well as core knowledge. In particular, the sub-themes of diversity and technology in education were present in documents from all four countries.

Teacher Knowledge Expert Group (TKEG)

Work on developing the TKS was supported by a Teacher Knowledge Expert Group (TKEG).2

The TKEG provided feedback on the conceptual and assessment frameworks for GPK developed by ACER. Each TKEG member also independently conducted a review of individual GPK items for the TKS assessment, providing ratings and suggestions for consideration. Item ratings from the TKEG supported the selection of items at each stage of survey development, from the pilot and field trial to the main survey.

TKEG members were also consulted in the development of the TKS context questionnaire. Their feedback and suggestions guided the development of new items on teachers’ opportunities to learn and the selection of items shared with the TALIS core survey.

Expert feedback from participating countries

National project managers (NPMs) in each participating country invited representatives from teacher standards bodies, national experts, and teacher education institutions to review the TKS conceptual and assessment framework. These expert reviewers assessed the cultural sensitivity of materials, their alignment with national teacher standards, and their relevance to teachers and teaching in their country. The Teacher Knowledge Expert Group (TKEG) reviewed this feedback and recommended further revisions before the framework was submitted to the TALIS Governing Board (TGB) for finalisation.

The design of the Teacher Knowledge Survey

TKS 2024 was designed as a 45-minute survey, to be taken within a 24-hour window, including two sections:

• Context questionnaire: A 20-minute set of self-report questions in a selected-response format with Likert scale matrix questions. All teachers received the same questions.

• General pedagogical knowledge assessment: A 25-minute set of selected-response assessment items, including simple and complex multiple-choice items. This section followed a rotated block design, meaning that teachers did not all receive the same questions. Each individual teacher answered 39 assessment items from the total pool of 52 GPK items, which were split between four assessment forms.

Context questionnaire

To help contextualise GPK results, TKS 2024 includes a questionnaire to gather data on characteristics of teachers and features of their teaching practices that are relevant to GPK. For the most part, the wording of items in the TKS context questionnaire follows that of corresponding items administered in the core survey. This was intended to provide opportunities to explore whether analytical inferences can be made at the country level between TKS and the core survey, and whether analyses can be conducted using data combined from the two sources.

A few additional questions were included in the TKS context questionnaire only, to capture constructs expected to be more strongly linked or particularly relevant for teachers’ GPK. By doing so, the TKS contextual questionnaire is more closely aligned with the content of the GPK assessment. The CERI conceptual framework of teachers’ professional competence also guided the choice of such content areas: opportunities to learn, affective-motivational characteristics, and teachers’ work practices. These are summarised in Table A A.1 and explained more fully in the TALIS TKS 2024 Conceptual and Assessment Framework (OECD, 2025[4]).

Table A A.1. TALIS 2024 TKS context areas related to general pedagogical knowledge

TKS context area	Sub-area	Content description	Derived or modified from
Opportunities to learn	Pedagogical content	A detailed set of questions asking the extent to which teachers have learned specific topics of pedagogy in either their initial teacher education or their continuing professional learning. Questions correspond to the dimensions of the GPK assessment framework: instruction, learning, and assessment. A fourth question covers content relating to the emergent priorities for teacher knowledge topics identified in the present TKS framework: teaching diverse students, using technology for teaching, supporting social and emotional learning, and developing 21st century thinking skills. While these correspond to aspects of the teachers’ learning and development content area of TALIS, they are distinctly different questions.	ITEL TKS 2016 Pilot study – Opportunities to learn
Affective-motivational characteristics	Teacher self-efficacy	Records the extent to which teachers feel confident in their ability to achieve certain outcomes in their teaching such as engaging students, effecting student learning, delivering high-quality instruction, managing the classroom and accommodating diversity.	TALIS 2024 ISCED level 2 core survey – Occupational perceptions
	Job satisfaction and teacher’s well-being	Elicits the extent to which teachers feel stress related to workload, student behaviour and other demands. Questions also measure the extent to which teachers enjoy teaching.	TALIS 2024 ISCED level 2 core survey – Occupational Perceptions
	Planned persistence	Indicates the extent to which teachers plan to remain in the teaching profession. Question asks how long teachers wish to remain in teaching.	TALIS 2024 ISCED level 2 core survey – Occupational perceptions
Teachers’ work practices	Teaching practices	Measures the extent to which teachers report engaging in seven general teaching practices: classroom and time management; clarity of teaching; cognitive activation; assessment and feedback; adaptation; supporting students’ social and emotional learning; support for consolidation. Also, use of class time and achievement of lesson aims.	TALIS 2024 ISCED level 2 core survey – Teachers’ work practices
	Professional practices	Indicates the extent to which teachers engage in collaborative activities with colleagues related to aspects of planning, teaching, assessment and professional learning.	TALIS 2024 ISCED level 2 core survey – Teachers’ work practices

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

However, items on teachers’ opportunities to learn (OTL) in the TKS module were developed separately. Unlike the core survey items that are more generally related to teachers’ learning and development, the TKS OTL items ask teachers about the extent to which their OTL prepared them in specific topics related to the key dimensions of GPK: instruction, learning and assessment. In addition, teachers participating in TKS are asked about their preparation in topics related to the emergent priorities for teacher knowledge (classroom diversity, technology in education, social and emotional learning, and 21^st century thinking skills).

Direct assessment of GPK

The assessment framework for TALIS TKS 2024, which guided the development of GPK assessment items, is based on the ITEL assessment framework and the background research conducted by König (2014[11]) on instruction, learning and assessment (OECD, 2025[4]). This assessment framework elaborates six sub-dimensions of GPK, outlined in Table A A.2.

Table A A.2. TALIS 2024 TKS assessment framework of general pedagogical knowledge

Dimension	Sub-dimension	Description
Instruction	Teaching methods and lesson planning	Knowledge of principles of instruction, how to productively use instructional time, and various teaching methods (e.g. direct instruction, discovery learning); knowing when and how to apply and adapt each teaching method to promote both individual and collective student understanding of learning tasks; how to select and use learning resources for teaching and prepare, structure and evaluate learning objectives, lessons and curricular units.
	Classroom management	Knowledge of classroom management principles, techniques and strategies: awareness of all classroom activity, handling multiple classroom events concurrently, pacing lessons appropriately and using routines to maintain momentum, providing clear directions, maintaining student attention and managing behaviour.
Learning	Learning and development	Knowledge of sources of student heterogeneity and how it impacts learning; knowledge of theories of learning and learner development as well as various cognitive learning processes, including learning strategies, memory and information processing, impact of prior knowledge, causal attributions, effects and quality characteristics of praise, and opportunities for increasing student engagement.
	Affective-motivational dispositions	Knowledge of motivational learning processes (e.g. achievement motivation) and strategies to motivate a single student or whole group.
Assessment	Evaluation and diagnosis procedures	Knowledge of different forms and purposes of formative and summative classroom assessments, various frames of reference (e.g. social, individual, criterion-based), quality of assessment and impacts of assessment feedback on student motivation and learning.
	Data literacy, interpretation and use	Knowledge of how to analyse, interpret, evaluate and use student data or learning data to inform the teaching and learning process.

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

In order to capture the broad range of GPK that is relevant for effective teaching and learning, the TKS GPK assessment framework also includes the additional knowledge categories outlined in Table A A.3

Table A A.3. TALIS 2024 TKS supplementary categories of general pedagogical knowledge

Category	Sub-category	Description
Cognitive demand	Recall	Items of this type require teachers to recall a definition, or identify elements of a phenomenon, term or concept.
	Analyse or apply	Items of this type require teachers to analyse a problem outlined by the item and apply recalled knowledge. Hence, they must understand a phenomenon or concept, and compare, categorise, assign or interpret a phenomenon, situation, or a general term.
Knowledge base	Theoretical/scientific	Items of this type require teachers to use knowledge of educational theories or scientific research that is formal, systematic, ordered and context-independent.
	Practice-based	Items of this type require teachers to apply professional judgement to answer classroom- or situation-based, context-specific items. Teacher judgement is interpreted here as deriving both from theoretical and working knowledge (i.e. a variety of contextually-specific experiences).
Thematic orientation	Core knowledge	Items of this type require teachers to use knowledge of established educational concepts, theories, and pedagogical principles (e.g. Piaget, Vygotsky, Bloom, Bandura, Erikson, etc.).
	Emergent priorities for teacher knowledge	Items of this type require teachers to use knowledge related to recent international priorities in education. For the purpose of TKS 2024, the focus is on: • Teaching diverse learners. Knowledge of how to provide inclusive learning opportunities to meet the diversity of learners and their needs • Using technology for teaching. Knowledge of the affordances of technology resources and how to use them to enhance teaching and learning • Supporting social and emotional learning. Knowledge of social and emotional skills: their impact on learning and how to support their development in students • Developing 21st century thinking skills. Knowledge of transferable, cross-curricular cognitive skills and how to teach them.

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

The composition of the final GPK assessment according to different knowledge dimensions and categories is outlined in Table A A.4 to Table A A.7. The target proportion of each knowledge dimension or category is indicated in parentheses, while the “achieved” percentages indicate the share of items actually included in the main survey instrument following the attrition of items after the pilot and field trial. Targets for each knowledge dimension or category were set up based on the assessment framework, as well as feedback from the Teacher Knowledge Expert Group (TKEG) and teachers and national experts from participating countries.

In terms of GPK dimensions and sub-dimensions, the aim was to have roughly equal coverage of the main GPK dimensions of instruction, learning and assessment. For each dimension, one sub-dimension was identified as containing fundamental content while another included additionally important knowledge, which was reflected in the target weights.

Table A A.4. GPK item proportions by dimension and sub-dimension

Dimension	Instruction (31-35%) Achieved: 38%	Learning (31-35%) Achieved: 32%	Assessment (31-35%) Achieved: 30%
Sub-dimension 1 : 59%	Teaching methods and lesson planning (18-22%) Achieved: 23%	Learning and development (18-22%) Achieved: 18%	Evaluation and diagnosis procedures (18-22%) Achieved: 18%
Sub-dimension 2 (33-45%) Achieved: 41%	Classroom management (11-15%) Achieved: 15%	Motivational-affective dispositions (11-15%) Achieved: 14%	Data literacy, interpretation and use (11-15%) Achieved: 12%

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

Regarding theoretical and practice-based knowledge, feedback from the TKEG, national experts and teacher focus groups suggested that practice-based items are more directly relevant to high-quality teaching. The aim was, therefore, to have a larger proportion of practice-based items in the GPK assessment overall Table A A.5. However, more emphasis was to be placed on theory-based items in the dimension of student learning, as relevant content tends to come from theoretical knowledge (e.g. psychological theory, cognitive science) rather than from classroom practice. Targets for practice-based items were difficult to achieve though, due to the weaker performance of these items in the pilot and field trial.

Table A A.5. GPK item proportions by knowledge base

Knowledge base	Instruction (31-35%) Achieved: 38%	Learning (31-35%) Achieved: 32%	Assessment (31-35%) Achieved: 30%
Theoretical/scientific (30-35%) Achieved: 48%	(6-8%) Achieved: 12%	(18-20%) Achieved: 24%	(6-8%) Achieved: 12%
Practice-based (65-70%) Achieved: 52%	(25-27%) Achieved: 26%	(13-15%) Achieved: 8%	(25-27%) Achieved: 18%

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

Feedback on the cognitive demand of GPK items suggested that the assessment should include a higher share of items requiring teachers to analyse or apply knowledge. Similarly to practice-based items, items requiring teachers to apply knowledge to specific situations were identified as more relevant to effective teaching, while the stand-alone value of recalled knowledge was questioned. Table A A.6 shows the targets set for recall and applied knowledge items, which were achieved in the design of the main survey instrument.

Table A A.6. GPK item proportions by cognitive demand

Knowledge base	Instruction (31-35%) Achieved: 38%	Learning (31-35%) Achieved: 32%	Assessment (31-35%) Achieved: 30%
Recall (33-45%) Achieved: 33%	(11-15%) Achieved: 11%	(11-15%) Achieved: 11%	(11-15%) Achieved: 11%
Analysing or applying (55-67%) Achieved: 67%	(18-22%) Achieved: 27%	(18-22%) Achieved: 20%	(18-22%) Achieved: 20%

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

Target ratios for the thematic orientation of GPK items placed more emphasis on core knowledge. By definition, core knowledge constitutes essential content that is likely to constitute the main part of initial teacher education programmes. As shown in Table A A.7, targets for the split of items were achieved in this domain.

Table A A.7. GPK item proportions by thematic orientation

Thematic orientation
Core knowledge (65-70%) Achieved: 67%
Emergent priorities for teacher knowledge (30-35%) Achieved: 33%

Source: OECD (2025[4]), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework.

The scaling of GPK items and the construction of the GPK scale

ACER was responsible for conducting an exhaustive analysis of the GPK assessment data collected in the Teacher Knowledge Survey. At various points in the process, these analyses were discussed with and vetted by the TALIS Technical Advisory Group and the Teacher Knowledge Expert Group. whose advice and feedback informed further analysis and final scaling procedures.

The following types of analysis were undertaken:

• review of percentages of missing data

• log data analysis, such as time per item and time per whole test

• investigation of the factorial structure of the GPK scale using multi-dimensional item response theory (IRT) models

• review of test reliability (IRT-based) for the overall GPK scale and potential sub-scales

• analyses based on IRT and classical test theory to examine item and test performance, such as:

  • item facility

  • adjusted (rest) and non-adjusted item-total correlation

  • item fit (weighted infit mean square [MNSQ])

  • mean latent construct estimates for respondents (per each question/item)

  • key and distractor analyses

  • item characteristic curves

• differential item functioning (DIF):

  • by gender

  • by country.

The analyses and scaling procedures were conducted using appropriate sampling weights for all TKS teacher data. For analyses and calibrations based on data from the combined international sample, so‑called “senate weights” were used to ensure that each of the eight participating education systems contributed to the results with the same weight.

Established psychometric models grounded in Item Response Theory (IRT), such as the two-parameter logistic model and the generalised partial credit model, were employed for item calibration and scaling.

Chapter 12 of the TALIS Technical Report (OECD, forthcoming[14]) describes in detail the various set of analysis conducted on the GPK assessment items and the main results. Here it is worth discussing the results of differential item functioning analysis, as it has the most direct implications for the interpretation of the results presented in this report.

Comparability of data and results from cross-national large-scale assessments relies on assessment items having comparable properties across participating countries. An assessment item shows item-by-country interaction (country DIF) when respondents with the same levels of knowledge, but from different countries, display a different probability of correctly answering the item. These variations may indicate problems in the translation or national adaptation of the items (for example, by including content that is not appropriate to specific national contexts). Cultural factors may also influence how some assessment items are perceived in different national contexts.

The first step to ensure the universal applicability of the assessment items in all the eight participating countries was the thorough review and piloting of the items at the pilot and field trial stage. Such activities were conducted by the different national teams, under the coordination of the OECD, ACER, and the other members of the international consortium in charge of the implementation of TALIS.

The occurrence of item-by-country interactions was then finally verified by analysing the data collected during the Main Study. Nationally calibrated item parameters were compared with those from a calibration of item parameters based on the combined dataset from all eight participating countries.

National project managers (NPMs) received detailed national reports that included information about the psychometric quality of their GPK assessment items in comparison with results for the international sample. NPMs were also provided with a list of items that were flagged for concerns about their psychometric quality (such as low discrimination, or larger deviations in facility from the international sample results) and asked to review regarding possible context effects or adaptation/translation issues that might explain these anomalies.

The comparisons of psychometric properties (such as facility or discrimination) of items at the national level (considering issues with psychometric quality only occurring in individual national samples), in combination with feedback from countries, led to the identification of five items (two in Chile and three in Poland) whose translations deviated from the original English source version. These items were discarded from the scaling for those countries and were set to “Not administered” in the TALIS database. Preliminary calibrations further showed that one item in Saudi Arabia (KTF4G46E1) had a negative 2PL slope parameter (indicating that respondents correctly answering that item had lower GPK scores, on average, than respondents who provided the wrong answer). This item was discarded at the national level from the international calibration and the national scaling but was preserved in the TALIS database.

Subsequently, item fit was further reviewed to identify items with larger country DIF warranting treatment as (separate) national items in the final scaling with their corresponding national item parameters. To this end, the 49 adjudicated items were included in an international calibration. Item fit was then reviewed at the national level based on the following fit statistics, in combination with visual inspections of expected and observed item characteristic curves for each national sample:

• the mean deviation (MD) statistic

• the root mean square deviation (RMSD) statistic

• the standardised chi square (S-X2) statistic

• the weighted mean square (Infit) statistic.

All four fit statistics were calculated based on comparisons between the expected and observed probabilities of correct responses to each assessment item.

Following the initial calibration with all items, two further calibrations were performed to derive the final item parameters for the scaling of GPK items. After discarding three items (KTF4G33D1, KTF4G45T1 and KTF4G641) from the item set and those national items that had translation deviations or negative discrimination, the remaining 49 items were calibrated, and the residual-based fit statistics were calculated within each participating education system.

After the first calibration, 11 national items were identified as presenting lack of item fit and treated as national items with country-specific parameters. The second calibration showed no further instances of item misfit at the national level and was used as the final calibration for the scaling of GPK items.

Table A A.8 shows a list of items that were either: (i) excluded for all countries from scaling; (ii) excluded from national scaling and the national database due to translation deviations; (iii) excluded from national scaling due to negative discrimination parameters; or (iv) treated as national items for scaling due to lack of measurement invariance.

Table A A.8. List of items that were excluded or treated as national items

Item	Level	Treatment	Rationale
KTF4G45T1	International	Discarded from scaling for all countries	Poor psychometric item quality
KTF4G641
KTF4G33D1
KTF4G36A1	Poland	Discarded from national database and scaling	Translation error reported by national centre
KTF4G721	Chile
KTF4G70E1	Poland
KTF4G51S1
KTF4G69D1	Chile
KTF4G46E1	Saudi Arabia	Discarded from national scaling	Negative slope parameter in preliminary calibration
KTF4G54H1	Chile	Treated as national item for scaling	Country DIF (national item fit issues)
KTF4G591
KTF4G50S1
KTF4G49E1
KTF4G78Q1	Croatia
KTF4G55F1	Morocco
KTF4G591	Poland
KTF4G54H1	Portugal
KTF4G29T1
KTF4G32F1	Saudi Arabia
KTF4G42T1	United States

Source: OECD (forthcoming[14])

For the final scaling of GPK, the Generalized Partial Credit Model was used, and plausible values were computed as final GPK scale scores. Using item parameters anchored at their estimated values from the calibration based on the international TKS data, plausible values were generated as random draws from the posterior distribution that depend on response data (the conditional item response model), other available context data (conditioning variables) and group membership, which in the case of TALIS TKS 2024 was determined by the education system in which an individual was assessed.

In a first step, GPK items were calibrated with a sample consisting of data from all eight countries participating in the TALIS TKS, using senate weights (see information on item calibration procedures above). In a second step, the item parameters obtained from this calibration were anchored for the subsequent generation of plausible values, that was conducted separately for each national datafile. For each teacher, five plausible values were drawn, that provide a robust basis for computing the measurement error as part of the final reported standard error for all statistics based on TKS GPK data.

The conditioning part of the model included some direct regressors (number of not-reached responses, average GPK of all other teachers at the school, explicit stratum, teachers’ gender) as well as factor scores derived from a principal component analysis (PCA) of all other information available in the TKS questionnaire. Including all available information is important to strengthen the process of generating plausible values as well as to allow an unbiased analysis between questionnaire data and GPK, regardless of the size of correlations between conditioning variables and GPK. All variables (both direct regressors and factor scores from the PCA) that are included in the conditioning model were scored so that they have no missing information (by giving cases with missing data the national mean or mode of the respective variable) and (dummy) missing indicators will be added to the model. School differences were accounted for by adding a variable reflecting the average GPK score (adjusted for the contribution of each respondent) as well as explicit stratification variables.

The plausible values were transformed to an international metric, where the average score was set to 250 and the standard deviation to 50 for the international dataset using senate weights.

References

[9] Baumert, J. et al. (2010), “Teachers’ Mathematical Knowledge, Cognitive Activation in the Classroom, and Student Progress”, American Educational Research Journal, Vol. 47/1, pp. 133-180, https://doi.org/10.3102/0002831209345157.

[5] Guerriero, S. (ed.) (2017), Pedagogical Knowledge and the Changing Nature of the Teaching Profession, Educational Research and Innovation, OECD Publishing, Paris, https://doi.org/10.1787/9789264270695-en.

[12] Guerriero, S. and N. Révai (2017), “Knowledge-based teaching and the evolution of a profession”, in Pedagogical Knowledge and the Changing Nature of the Teaching Profession, OECD Publishing, Paris, https://doi.org/10.1787/9789264270695-13-en.

[11] König, J. (2014), Designing an international instrument to assess teachers’ general pedagogical knowledge (GPK): Review of studies, considerations, and recommendations, Centre for Educational Research and Innovation (CERI) Governing Board, OECD Directorate for Education and Skills, https://one.oecd.org/document/EDU/CERI/CD/RD(2014)3/REV1/en/pdf.

[2] König, J. et al. (2011), “General Pedagogical Knowledge of Future Middle School Teachers: On the Complex Ecology of Teacher Education in the United States, Germany, and Taiwan”, Journal of Teacher Education, Vol. 62/2, pp. 188-201, https://doi.org/10.1177/0022487110388664.

[8] Leijen, Ä. et al. (2022), “What constitutes teachers’ general pedagogical knowledge and how it can be assessed: A literature review”, Teachers and Teaching, Vol. 28/2, pp. 206-225, https://doi.org/10.1080/13540602.2022.2062710.

[4] OECD (2025), TALIS Teacher Knowledge Survey 2024 Conceptual and Assessment Framework, OECD Publishing, Paris, https://doi.org/10.1787/65903902-en.

[14] OECD (forthcoming), TALIS 2024 Technical Report, TALIS, OECD Publishing, Paris.

[7] Shulman, L. (1987), “Knowledge and Teaching:Foundations of the New Reform”, Harvard Educational Review, Vol. 57/1, pp. 1-23, https://doi.org/10.17763/haer.57.1.j463w79r56455411.

[6] Sonmark, K. et al. (2017), “Understanding teachers’ pedagogical knowledge: report on an international pilot study”, OECD Education Working Papers, No. 159, OECD Publishing, Paris, https://doi.org/10.1787/43332ebd-en.

[13] Toledo-Figueroa, D., N. Révai and S. Guerriero (2017), “Teacher professionalism and knowledge in qualification frameworks and professional standards”, in Guerriero, S. (ed.), Pedagogical Knowledge and the Changing Nature of the Teaching Profession, OECD Publishing, Paris, https://doi.org/10.1787/9789264270695-en (accessed on 15 January 2026).

[3] Ulferts, H. (2021), Teaching as a Knowledge Profession: Studying Pedagogical Knowledge across Education Systems.

[10] Voss, T., M. Kunter and J. Baumert (2011), “Assessing teacher candidates’ general pedagogical/psychological knowledge: Test construction and validation.”, Journal of Educational Psychology, Vol. 103/4, pp. 952-969, https://doi.org/10.1037/a0025125.

[1] Weyers, J., R. Ligtvoet and J. König (2024), “How does pre-service teachers’ general pedagogical knowledge develop during university teacher education? Examining the impact of learning opportunities and entry characteristics over five time points using longitudinal models”, Journal of Curriculum Studies, Vol. 56/4, pp. 448-467, https://doi.org/10.1080/00220272.2024.2355923.

Notes

← 1.  A teacher could not participate in both TKS and the core TALIS survey, though.

← 2.  The list of TKEG members is contained in Annex F.