research paper on teacher training

A Review of the Literature on Teacher Effectiveness and Student Outcomes

• Open Access
• First Online: 24 May 2019

Cite this chapter

You have full access to this open access chapter

• Nathan Burroughs 25 ,
• Jacqueline Gardner 26 ,
• Youngjun Lee 27 ,
• Siwen Guo 28 ,
• Israel Touitou 29 ,
• Kimberly Jansen 30 &
• William Schmidt 31  

Part of the book series: IEA Research for Education ((IEAR,volume 6))

156k Accesses

47 Citations

74 Altmetric

Researchers agree that teachers are one of the most important school-based resources in determining students’ future academic success and lifetime outcomes, yet have simultaneously had difficulties in defining what teacher characteristics make for an effective teacher. This chapter reviews the large body of literature on measures of teacher effectiveness, underscoring the diversity of methods by which the general construct of “teacher quality” has been explored, including experience, professional knowledge, and opportunity to learn. Each of these concepts comprises a number of different dimensions and methods of operationalizing. Single-country research (and particularly research from the United States) is distinguished from genuinely comparative work. Despite a voluminous research literature on the question of teacher quality, evidence for the impact of teacher characteristics (experience and professional knowledge) on student outcomes remains quite limited. There is a smaller, but more robust set of findings for the effect of teacher support on opportunity to learn. Five measures may be associated with higher student achievement: teacher experience (measured by years of teaching), teacher professional knowledge (measured by education and self-reported preparation to teach mathematics), and teacher provision of opportunity to learn (measured by time on mathematics and content coverage). These factors provide the basis for a comparative cross-country model.

You have full access to this open access chapter,  Download chapter PDF

• Opportunity to learn
• Teacher education
• Teacher experience
• Teacher quality
• Trends in International Mathematics and Science Study (TIMSS)

2.1 Defining Teacher Effectiveness

Researchers agree that teachers are one of the most important school-based resources in determining students’ future academic success and lifetime outcomes (Chetty et al. 2014 ; Rivkin et al. 2005 ; Rockoff 2004 ). As a consequence, there has been a strong emphasis on improving teacher effectiveness as a means to enhancing student learning. Goe ( 2007 ), among others, defined teacher effectiveness in terms of growth in student learning, typically measured by student standardized assessment results. Chetty et al. ( 2014 ) found that students taught by highly effective teachers, as defined by the student growth percentile (SGPs) and value-added measures (VAMs), were more likely to attend college, earn more, live in higher-income neighborhoods, save more money for retirement, and were less likely to have children during their teenage years. This potential of a highly effective teacher to significantly enhance the lives of their students makes it essential that researchers and policymakers properly understand the factors that contribute to a teacher’s effectiveness. However, as we will discuss in more detail later in this report, studies have found mixed results regarding the relationships between specific teacher characteristics and student achievement (Wayne and Youngs 2003 ). In this chapter, we explore these findings, focusing on the three main categories of teacher effectiveness identified and examined in the research literature: namely, teacher experience, teacher knowledge, and teacher behavior. Here we emphasize that much of the existing body of research is based on studies from the United States, and so the applicability of such national research to other contexts remains open to discussion.

2.2 Teacher Experience

Teacher experience refers to the number of years that a teacher has worked as a classroom teacher. Many studies show a positive relationship between teacher experiences and student achievement (Wayne and Youngs 2003 ). For example, using data from 4000 teachers in North Carolina, researchers found that teacher experience was positively related to student achievement in both reading and mathematics (Clotfelter et al. 2006 ). Rice ( 2003 ) found that the relationship between teacher experience and student achievement was most pronounced for students at the secondary level. Additional work in schools in the United States by Wiswall ( 2013 ), Papay and Kraft ( 2015 ), and Ladd and Sorenson ( 2017 ), and a Dutch twin study by Gerritsen et al. ( 2014 ), also indicated that teacher experience had a cumulative effect on student outcomes.

Meanwhile, other studies have failed to identify consistent and statistically significant associations between student achievement and teacher experience (Blomeke et al. 2016 ; Gustaffsson and Nilson 2016 ; Hanushek and Luque 2003 ; Luschei and Chudgar 2011 ; Wilson and Floden 2003 ). Some research from the United States has indicated that experience matters very much early on in a teacher’s career, but that, in later years, there were little to no additional gains (Boyd et al. 2006 ; Rivkin et al. 2005 ; Staiger and Rockoff 2010 ). In the first few years of a teacher’s career, accruing more years of experience seems to be more strongly related to student achievement (Rice 2003 ). Rockoff ( 2004 ) found that, when comparing teacher effectiveness (understood as value-added) to student test scores in reading and mathematics, teacher experience was positively related to student mathematics achievement; however, such positive relationships leveled off after teachers had gained two years of teaching experience. Drawing on data collected from teachers of grades four to eight between 2000 and 2008 within a large urban school district in the United States, Papay and Kraft ( 2015 ) confirmed previous research on the benefits experience can add to a novice teacher’s career. They found that student outcomes increased most rapidly during their teachers’ first few years of employment. They also found some further student gains due to additional years of teaching experience beyond the first five years. The research of Pil and Leana ( 2009 ) adds additional nuance; they found that acquiring teacher experience at the same grade level over a number of years, not just teacher experience in general (i.e. at multiple grades), was positively related to student achievement.

2.3 Teacher Professional Knowledge

A teacher’s professional knowledge refers to their subject-matter knowledge, curricular knowledge, and pedagogical knowledge (Collinson 1999 ). This professional knowledge is influenced by the undergraduate degrees earned by a teacher, the college attended, graduate studies undertaken, and opportunities to engage with on-the job training, commonly referred to as professional development (Collinson 1999 ; Rice 2003 ; Wayne and Youngs 2003 ). After undertaking in-depth quantitative analyses of the United States’ 1993–1994 Schools and Staffing Survey (SASS) and National Assessment of Educational Progress (NAEP) data sets, Darling-Hammond ( 2000 ) argued that measures of teacher preparation and certification were by far the strongest correlates of student achievement in reading and mathematics, after controlling for student poverty levels and language status.

As with experience, research on the impact of teacher advanced degrees, subject specializations, and certification has been inconclusive, with several studies (Aaronson et al. 2007 ; Blomeke et al. 2016 ; Hanushek and Luque 2003 ; Harris and Sass 2011 ; Luschei and Chudgar 2011 ) suggesting weak, inconsistent, or non-significant relationships with student achievement. However, several international studies comparing country means found that teacher degrees (Akiba et al. 2007 ; Gustaffsson and Nilson 2016 ; Montt 2011 ) were related to student outcomes, as did Woessman’s ( 2003 ) student-level study of multiple countries.

2.3.1 Undergraduate Education

In their meta-analysis of teacher effectiveness, Wayne and Youngs ( 2003 ) found three studies that showed some relationship between the quality of the undergraduate institution that a teacher attended and their future students’ success in standardized tests. In a thorough review of the research on teacher effectiveness attributes, Rice ( 2003 ) found that the selectivity of undergraduate institution and the teacher preparation program may be related to student achievement for students at the high school level and for high-poverty students.

In terms of teacher preparation programs, Boyd et al. ( 2009 ) found that overall these programs varied in their effectiveness. In their study of 31 teacher preparation programs designed to prepare teachers for the New York City School District, Boyd et al. ( 2009 ) drew from data based on document analyses, interviews, surveys of teacher preparation instructors, surveys of participants and graduates, and student value-added scores. They found that if a program was effective in preparing teachers to teach one subject, it tended to also have success in preparing teachers to teach other subjects as well. They also found that teacher preparation programs that focused on the practice of teaching and the classroom, and provided opportunities for teachers to study classroom practices, tended to prepare more effective teachers. Finally, they found that programs that included some sort of final project element (such as a personal research paper, or portfolio presentation) tended to prepare more effective teachers.

Beyond the institution a teacher attends, the coursework they choose to take within that program may also be related to their future students’ achievement. These associations vary by subject matter. A study by Rice ( 2003 ) indicated that, for teachers teaching at the secondary level, subject-specific coursework had a greater impact on their future students’ achievement. Similarly Goe ( 2007 ) found that, for mathematics, an increase in the amount of coursework undertaken by a trainee teacher was positively related to their future students’ achievement. By contrast, the meta-analysis completed by Wayne and Youngs ( 2003 ) found that, for history and English teachers, there was no evidence of a relationship between a teacher’s undergraduate coursework and their future students’ achievement in those subjects.

2.3.2 Graduate Education

In a review of 14 studies, Wilson and Floden ( 2003 ) were unable to identify consistent relationships between a teacher’s level of education and their students’ achievement. Similarly, in their review of data from 4000 teachers in North Carolina, Clotfelter et al. ( 2006 ) found that teachers who held a master’s degree were associated with lower student achievement. However, specifically in terms of mathematics instruction, teachers with higher degrees and who undertook more coursework during their education seem to be positively related to their students’ mathematics achievement (Goe 2007 ). Likewise, Harris and Sass ( 2011 ) found that there was a positive relationship between teachers who had obtained an advanced degree during their teaching career and their students’ achievement in middle school mathematics. They did not find any significant relationships between advanced degrees and student achievement in any other subject area. Further, using data from the United States’ Early Childhood Longitudinal Study (ECLS-K), Phillips ( 2010 ) found that subject-specific graduate degrees in elementary or early-childhood education were positively related to students’ reading achievement gains.

2.3.3 Certification Status

Another possible indicator of teacher effectiveness could be whether or not a teacher holds a teaching certificate. Much of this research has focused on the United States, which uses a variety of certification approaches, with lower grades usually having multi-subject general certifications and higher grades requiring certification in specific subjects. Wayne and Youngs ( 2003 ) found no clear relationship between US teachers’ certification status and their students’ achievement, with the exception of the subject area of mathematics, where students tended have higher test scores when their teachers had a standard mathematics certification. Rice ( 2003 ) also found that US teacher certification was related to high school mathematics achievement, and also found that there was some evidence of a relationship between certification status and student achievement in lower grades. Meanwhile, in their study of grade one students, Palardy and Rumberger ( 2008 ) also found evidence that students made greater gains in reading ability when taught by fully certified teachers.

In a longitudinal study using data from teachers teaching grades four and five and their students in the Houston School District in Texas, Darling-Hammond et al. ( 2005 ) found that those teachers who had completed training that resulted in a recognized teaching certificate were more effective that those who had no dedicated teaching qualifications. The study results suggested that teachers without recognized US certification or with non-standard certifications generally had negative effects on student achievement after controlling for student characteristics and prior achievement, as well as the teacher’s experience and degrees. The effects of teacher certification on student achievement were generally much stronger than the effects for teacher experience. Conversely, analyzing data from the ECLS-K, Phillips ( 2010 ) found that grade one students tended to have lower mathematics achievement gains when they had teachers with standard certification. In sum, the literature the influence of teacher certification remains deeply ambiguous.

2.3.4 Professional Development

Although work by Desimone et al. ( 2002 , 2013 ) suggested that professional development may influence the quality of instruction, most researchers found that teachers’ professional development experiences showed only limited associations with their effectiveness, although middle- and high-school mathematics teachers who undertook more content-focused training may be the exception (Blomeke et al. 2016 ; Harris and Sass 2011 ). In their meta-analysis of the effects of professional development on student achievement, Blank and De Las Alas ( 2009 ) found that 16 studies reported significant and positive relationships between professional development and student achievement. For mathematics, the average effect size of studies using a pre-post assessment design was 0.21 standard deviations.

Analyzing the data from six data sets, two from the Beginning Teacher Preparation Survey conducted in Connecticut and Tennessee, and four from the United States National Center for Education Statistics’ National Assessment of Educational Progress (NAEP), Wallace ( 2009 ) used structural equation modeling to find that professional development had a very small, but occasionally statistically significant effect on student achievement. She found, for example, that for NAEP mathematics data from the year 2000, 1.2 additional hours of professional development per year were related to an increase in average student scores of 0.62 points, and for reading, an additional 1.1 h of professional development were related to an average increase in student scores of 0.24 points. Overall, Wallace ( 2009 ) identified professional development had moderate effects on teacher practice and some small effects on student achievement when mediated by teacher practice.

2.3.5 Teacher Content Knowledge

Of course, characteristics like experience and education may be imperfect proxies for teacher content knowledge; unfortunately, content knowledge is difficult to assess directly. However, there is a growing body of work suggesting that teacher content knowledge may associated with student learning. It should be noted that there is an important distinction between general content knowledge about a subject (CK) and pedagogical content knowledge (PCK) specifically related to teaching that subject, each of which may be independently related to student outcomes (Baumert et al. 2010 ).

Studies from the United States (see for example, Chingos and Peterson 2011 ; Clotfelter et al. 2006 ; Constantine et al. 2009 ; Hill et al. 2005 ; Shuls and Trivitt 2015 ) have found some evidence that higher teacher cognitive skills in mathematics are associated with higher student scores. Positive associations between teacher content knowledge and student outcomes were also found in studies based in Germany (Baumert et al. 2010 ) and Peru (Metzler and Woessman 2012 ), and in a comparative study using Programme for the International Assessment of Adult Competencies (PIAAC) data undertaken by Hanushek et al. ( 2018 ). These findings are not universal, however, other studies from the United States (Blazar 2015 ; Garet et al. 2016 ; Rockoff et al. 2011 ) failed to find a statistically significant association between teacher content knowledge and student learning.

The studies we have discussed all used some direct measure of teacher content knowledge. An alternative method of assessing mathematics teacher content knowledge is self-reported teacher preparation to teach mathematics topics. Both TIMSS and IEA’s Teacher Education and Development Study in Mathematics (TEDS-M, conducted in 2007–2008) have included many questions, asking teachers to report on their preparedness to teach particular topics. Although Luschei and Chudgar ( 2011 ) and Gustafsson and Nilson ( 2016 ) found that these items had a weak direct relationship to student achievement across countries, other studies have suggested that readiness is related to instructional quality (Blomeke et al. 2016 ), as well as content knowledge and content preparation (Schmidt et al. 2017 ), suggesting that instructional quality may have an indirect effect on student learning.

2.4 Teacher Behaviors and Opportunity to Learn

Although the impact of teacher characteristics (experience, education, and preparedness to teach) on student outcomes remains an open question, there is much a much more consistent relationship between student achievement and teacher behaviors (instructional time and instructional content), especially behaviors related instructional content. Analyzing TIMSS, Schmidt et al. ( 2001 ) found an association between classroom opportunity to learn (OTL), interpreted narrowly as student exposure to instructional content, and student achievement. In a later study using student-level PISA data, Schmidt et al. ( 2015 ) identified a robust relationship between OTL and mathematics literacy across 62 different educational systems. The importance of instructional content has been recognized by national policymakers, and has helped motivate standards-based reform in an effort to improve student achievement, such as the Common Core in the United States (Common Core Standards Initiative 2018 ). However, we found that there was little research on whether teacher instructional content that aligned with national standards had improved student learning; the only study that we were able to identify found that such alignment had only very weak associations with student mathematics scores (Polikoff and Porter 2014 ). Student-reported data indicates that instructional time (understood as classroom time on a particular subject) does seem to be related to mathematics achievement (Cattaneo et al. 2016 ; Jerrim et al. 2017 ; Lavy 2015 ; Rivkin and Schiman 2015 ; Woessman 2003 ).

2.5 Conclusion

This review of the literature simply brushes the surface of the exceptional body of work on the relationship between student achievement and teacher characteristics and behaviors. Whether analyzing US-based, international, or the (limited) number of comparative studies, the associations between easily measurable teacher characteristics, like experience and education, and student outcomes in mathematics, remains debatable. In contrast, there is more evidence to support the impact of teacher behaviors, such as instructional content and time on task, on student achievement. Our goal was to incorporate all these factors into a comparative model across countries, with the aim of determining what an international cross-national study like TIMSS could reveal about the influence of teachers on student outcomes in mathematics. The analysis that follows draws on the existing body of literature on teacher effectiveness, which identified key teacher factors that may be associated with higher student achievement: teacher experience, teacher professional knowledge (measured by education and self-reported preparation to teach mathematics), and teacher provision of opportunity to learn (time on mathematics and content coverage).

Aaronson, D., Barrow, L., & Sander, W. (2007). Teachers and student achievement in the Chicago public high schools. Journal of Labor Economics, 25 (1), 95–135.

Article   Google Scholar  

Akiba, M., LeTendre, G., & Scribner, J. (2007). Teacher quality, opportunity gap, and national achievement in 46 countries. Educational Researcher, 36 (7), 369–387.

Baumert, J., Kunter, M., Blum, W., Brunner, M., Voss, T., Jordan, A., et al. (2010). Teachers’ mathematical knowledge, cognitive activation in the classroom, and student progress. American Educational Research Journal, 47 (1), 133–180.

Blank, R. K., & De Las Alas, N. (2009). The effects of teacher professional development on gains in student achievement: How meta analysis provides scientific evidence useful to education leaders . Washington, DC: Council of Chief State School Officers. Retrieved from https://files.eric.ed.gov/fulltext/ED544700.pdf .

Blazar, D. (2015). Effective teaching in elementary mathematics: Identifying classroom practices that support student achievement. Economics of Education Review, 48, 16–29.

Blomeke, S., Olsen, R., & Suhl, U. (2016). Relation of student achievement to the quality of their teachers and instructional quality. In T. Nilson & J. Gustafsson (Eds.), Teacher quality, instructional quality and student outcomes . IEA Research for Education (Vol. 2, pp. 21–50). Cham, Switzerland: Springer. Retrieved from https://link.springer.com/chapter/10.1007/978-3-319-41252-8_2 .

Boyd, D., Grossman, P., Lankford, H., & Loeb, S. (2006). How changes in entry requirements alter the teacher workforce and affect student achievement. Education Finance and Policy, 1 (2), 176–216.

Boyd, D. J., Grossman, P. L., Lankford, H., Loeb, S., & Wyckoff, J. (2009). Teacher preparation and student achievement. Educational Evaluation and Policy Analysis, 31 (4), 416–440.

Cattaneo, M., Oggenfuss, C., & Wolter, S. (2016). The more, the better? The impact of instructional time on student performance. Institute for the Study of Labor (IZA) Discussion Paper No. 9797. Bonn, Germany: Forschungsinstitut zur Zukunft der Arbeit/Institute for the Study of Labor. Retrieved from ftp.iza.org/dp9797.pdf .

Chetty, R., Friedman, J. N., & Rockoff, J. E. (2014). Measuring the impacts of teachers II: Teacher value-added and student outcomes in adulthood. American Economic Review, 104 (9), 2633–2679.

Chingos, M., & Peterson, P. (2011). It’s easier to pick a good teacher than to train one: Familiar and new results on the correlates of teacher effectiveness. Economics of Education Review, 30 (3), 449–465.

Clotfelter, C. T., Ladd, H. F., & Vigdor, J. L. (2006). Teacher-student matching and the assessment of teacher effectiveness. Journal of Human Resources, 41 (4), 778–820.

Collinson, V. (1999). Redefining teacher excellence. Theory Into Practice , 38 (1), 4–11. Retrieved from https://doi.org/10.1080/00405849909543824 .

Common Core Standards Initiative. (2018). Preparing America’s students for success [www document]. Retrieved from http://www.corestandards.org/ .

Constantine, J., Player, D., Silva, T., Hallgren, K., Grider, M., & Deke, J. (2009). An evaluation of teachers trained through different routes to certification. National Center for Education Evaluation and Regional Assistance paper 2009-4043. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, US Department of Education. Retrieved from https://ies.ed.gov/ncee/pubs/20094043/ .

Darling-Hammond, L. (2000). Teacher quality and student achievement. Education Policy Analysis Archives , 8, 1. Retrieved from https://epaa.asu.edu/ojs/article/viewFile/392/515 .

Darling-Hammond, L., Holtzman, D. J., Gatlin, S. J., & Vasquez Heilig, J. (2005). Does teacher preparation matter? Evidence about teacher certification, Teach for America, and teacher effectiveness. Education Policy Analysis Archives, 13, 42. Retrieved from https://epaa.asu.edu/ojs/article/view/147 .

Desimone, L., Porter, A., Garet, M., Yoon, K., & Birman, B. (2002). Effects of professional development on teachers’ instruction: Results from a three-year longitudinal study. Education Evaluation and Policy Analysis, 24 (2), 81–112.

Desimone, L., Smith, T., & Phillips, K. (2013). Teacher and administrator responses to standards-based reform. Teachers College Record, 115 (8), 1–53.

Google Scholar  

Garet, M. S., Heppen, J. B., Walters, K., Parkinson, J., Smith, T. M., Song, M., et al. (2016). Focusing on mathematical knowledge: The impact of content-intensive teacher professional development . National Center for Education Evaluation and Regional Assistance paper 2016-4010. Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, US Department of Education. Retrieved from https://ies.ed.gov/ncee/pubs/20094010/ .

Gerritsen, S., Plug, E., & Webbink, D. (2014). Teacher quality and student achievement: Evidence from a Dutch sample of twins . CPB discussion paper 294. The Hague, The Netherlands: Central Plan Bureau/Netherlands Bureau for Economic Policy Analysis. Retrieved from https://ideas.repec.org/p/cpb/discus/294.html .

Goe, L. (2007). The link between teacher quality and student outcomes: A research synthesis . NCCTQ Report. Washington, DC: National Comprehensive Center for Teacher Quality. Retrieved from http://www.gtlcenter.org/sites/default/files/docs/LinkBetweenTQandStudentOutcomes.pdf .

Gustafsson, J., & Nilson, T. (2016). The impact of school climate and teacher quality on mathematics achievement: A difference-in-differences approach. In T. Nilson & J. Gustafsson (Eds.), Teacher quality, instructional quality and student outcomes , IEA Research for Education (Vol. 2, pp. 81–95). Cham, Switzerland: Springer. Retrieved from https://link.springer.com/chapter/10.1007/978-3-319-41252-8_4 .

Hanushek, E., & Luque, J. (2003). Efficiency and equity in schools around the world. Economics of Education Review, 22 (5), 481–502.

Hanushek, E., Piopiunik, M., & Wiederhold, S. (2018). The value of smarter teachers: International evidence on teacher cognitive skills and student performance . Journal of Human Resources (in press). https://doi.org/10.3368/jhr.55.1.0317.8619r1 .

Harris, D. N., & Sass, T. R. (2011). Teacher training, teacher quality and student achievement. Journal of Public Economics, 95 (7–8), 798–812.

Hill, H., Rowan, B., & Ball, D. (2005). Effects of teachers’ mathematical knowledge for teaching on student achievement. American Educational Research Journal, 42 (2), 371–406.

Jerrim, J., Lopez-Agudo, L., Marcenaro-Gutierrez, O., & Shure, N. (2017). What happens when econometrics and psychometrics collide? An example using the PISA data. Economics of Education Review, 61, 51–58.

Ladd, H. F., & Sorenson, L. C. (2017). Returns to teacher experience: Student achievement and motivation in middle school. Education Finance and Policy, 12 (2), 241–279. Retrieved from https://www.mitpressjournals.org/doi/10.1162/EDFP_a_00194 .

Lavy, V. (2015). Do differences in schools’ instruction time explain international achievement gaps? Evidence from developed and developing countries. The Economic Journal, 125 (11), 397–424.

Luschei, T., & Chudgar, A. (2011). Teachers, student achievement, and national income: A cross-national examination of relationships and interactions. Prospects, 41, 507–533.

Metzler, J., & Woessman, L. (2012). The impact of teacher subject knowledge on student achievement: Evidence from within-teacher within-student variation. Journal of Development Economics, 99 (2), 486–496.

Montt, G. (2011). Cross-national differences in educational achievement inequality. Sociology of Education, 84 (1), 49–68.

Palardy, G. J., & Rumberger, R. W. (2008). Teacher effectiveness in first grade: The importance of background qualifications, attitudes, and instructional practices for student learning. Educational Evaluation and Policy Analysis, 30 (2), 111–140.

Papay, J., & Kraft, M. (2015). Productivity returns to experience in the teacher labor market: Methodological challenges and new evidence on long-term career improvement. Journal of Public Economics, 130, 105–119.

Phillips, K. J. (2010). What does ‘highly qualified’ mean for student achievement? Evaluating the relationships between teacher quality indicators and at-risk students’ mathematics and reading achievement gains in first grade. The Elementary School Journal, 110 (4), 464–493.

Pil, F. K., & Leana, C. (2009). Applying organizational research to public school reform: The effects of teacher human and social capital on student performance. Academy of Management Journal , 52 (6), 1101–1124.

Polikoff, M., & Porter, A. (2014). Instructional alignment as a measure of teaching quality. Educational Evaluation and Policy Analysis, 36 (4), 399–416.

Rice, J. K. (2003). Teacher quality: Understanding the effectiveness of teacher attributes . Washington DC: Economic Policy Institute.

Rivkin, S., Hanushek, E., & Kain, J. (2005). Teachers, schools, and academic achievement. Econometrica, 73 (2), 417–458.

Rivkin, S., & Schiman, J. (2015). Instruction time, classroom quality, and academic achievement. The Economic Journal, 125 (11), 425–448.

Rockoff, J. (2004). The impact of individual teachers on student achievement: Evidence from panel data. The American Economic Review, 94 (2), 247–252.

Rockoff, J. E., Jacob, B. A., Kane, T. J., & Staiger, D. O. (2011). Can you recognize an effective teacher when you recruit one? Education Finance and Policy, 6 (1), 43–74.

Schmidt, W., Burroughs, N., Cogan, L., & Houang, R. (2017). The role of subject-matter content in teacher preparation: An international perspective for mathematics. Journal of Curriculum Studies, 49 (2), 111–131.

Schmidt, W., Burroughs, N., Zoido, P., & Houang, R. (2015). The role of schooling in perpetuating educational inequality: An international perspective. Education Researcher, 44 (4), 371–386.

Schmidt, W., McKnight, C., Houang, R., Wang, H., Wiley, D., Cogan, L., et al. (2001). Why schools matter: A cross-national comparison of curriculum and learning . San Francisco, CA: Jossey-Bass.

Shuls, J., & Trivitt, J. (2015). Teacher effectiveness: An analysis of licensure screens. Educational Policy, 29 (4), 645–675.

Staiger, D., & Rockoff, J. (2010). Searching for effective teachers with imperfect information. Journal of Economic Perspectives, 24 (3), 97–118.

Wallace, M. R. (2009). Making sense of the links: Professional development, teacher practices, and student achievement. Teachers College Record, 111 (2), 573–596.

Wayne, A. J., & Youngs, P. (2003). Teacher characteristics and student achievement gains: A review. Review of Educational Research, 73 (1), 89–122.

Wilson, S. M., & Floden, R. E. (2003). Creating effective teachers: Concise answers for hard questions. An addendum to the report “Teacher preparation research: Current knowledge, gaps, and recommendations . Washington, DC: AACTE Publications.

Wiswall, M. (2013). The dynamics of teacher quality. Journal of Public Economics, 100, 61–78.

Woessman, L. (2003). Schooling resources, educational institutions, and student performance: The international evidence. Oxford Bulletin of Economics and Statistics, 65 (2), 117–170.

Download references

Author information

Authors and affiliations.

CREATE for STEM, Michigan State University, East Lansing, MI, USA

Nathan Burroughs

Office of K-12 Outreach, Michigan State University, East Lansing, MI, USA

Jacqueline Gardner

Michigan State University, East Lansing, MI, USA

Youngjun Lee

College of Education, Michigan State University, East Lansing, MI, USA

Israel Touitou

Kimberly Jansen

Center for the Study of Curriculum Policy, Michigan State University, East Lansing, MI, USA

William Schmidt

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Nathan Burroughs .

Rights and permissions

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Reprints and permissions

Copyright information

© 2019 International Association for the Evaluation of Educational Achievement (IEA)

About this chapter

Burroughs, N. et al. (2019). A Review of the Literature on Teacher Effectiveness and Student Outcomes. In: Teaching for Excellence and Equity. IEA Research for Education, vol 6. Springer, Cham. https://doi.org/10.1007/978-3-030-16151-4_2

Download citation

DOI : https://doi.org/10.1007/978-3-030-16151-4_2

Published : 24 May 2019

Publisher Name : Springer, Cham

Print ISBN : 978-3-030-16150-7

Online ISBN : 978-3-030-16151-4

eBook Packages : Education Education (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Int J Environ Res Public Health

Teacher Training Can Make a Difference: Tools to Overcome the Impact of COVID-19 on Primary Schools. An Experimental Study

Overcoming the impact of the coronavirus pandemic (COVID-19) on primary schools is an emerging need and priority in the current social welfare system. Accordingly, this study presents an empirical learning package to support teachers, who perform frontline work in schools, in coping with stress, preventing burnout, improving their information and communications technology (ICT) competency, and introducing the principles of emotional intelligence (EI) in the classroom. The participants included 141 primary school teachers (M = 38.4 years, SD = 6.84; 54.6% women). They were randomly assigned to an experimental or control group. The experimental group participated in the 14-week teacher training program, whereas the control group did not participate in the program or receive any other training during the intervention. Repeated-measures analysis of variance (time x group) was performed to identify the effects of the teacher training program. Teachers who participated in the training program evaluated it positively and showed significant differences compared to the control group in their abilities to cope with stress and avoid burnout, their ICT competency, and their introduction of EI in the classroom. Implications for supporting teachers are discussed.

1. Introduction

This research aimed to demonstrate the efficacy of a teacher training program intended to support primary school teachers in four key areas: (1) coping with stress, (2) preventing burnout, (3) improving their information and communications technology (ICT) competency, and (4) introducing emotional intelligence (EI) in the classroom. Coronavirus disease 2019 (COVID-19) is having a clear impact on the educational setting of primary school; practical teaching strategies and methodologies are required to face and overcome this new threat to health, social development, and education. The following subsections describe the current state of research in this field and the teacher training program

1.1. Theoretical Framework

1.1.1. work-related stress among teachers.

Several studies have demonstrated a positive relationship between teachers’ well-being and their efficacy in teaching [ 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. However, practical and applied research on training programs to improve teacher well-being and efficacy is still needed, particularly in light of the extreme challenges posed by the COVID-19 crisis. Furthermore, even if teachers do receive training in educational methodologies to manage stress, classroom-level implementation can remain low [ 16 , 17 ].

It is important to consider in order to support and design efficient training programs like this and according to previous researches, that the resources that can mitigate or reduce teacher stress and burnout are internal (managing classroom self-efficacy and instructional self-efficacy) and external (school support resources). Both (internal and external resources) have shown a negative effect on job stressors [ 18 ]. For example, as Doménech-Betoret and Gomez-Artiga [ 19 ] pointed out, there is a strong and significant association between self-efficacy and the coping strategies used by teachers, and coping strategies act as moderator on burnout dimensions.

Appropriate training can mitigate work-related stress among teachers by furnishing them with the skills, strategies, and resources needed to manage stress, improve efficacy, and increase workplace satisfaction [ 20 , 21 , 22 , 23 ]. The key to the effectiveness of such training is to provide teachers with useful strategies and resources and facilitate effective transfer of these skills to the classroom to buffer against teacher stress in everyday school life, as the program developed in the present study does.

On this way, teacher burnout is a serious problem in schools, with links to impoverishment of the teaching occupation, frustration and dissatisfaction with teaching, and job absenteeism [ 24 , 25 , 26 , 27 , 28 , 29 , 30 ]. It is critical to train teachers to face this difficulty, safeguard themselves, and achieve optimal psychological development at work [ 31 , 32 , 33 , 34 , 35 ].

Training to prevent teacher burnout plays a protective role and improves the psychosocial environment and health of teachers’ work life [ 36 , 37 , 38 , 39 , 40 ]. It also increases teachers’ engagement, resilience, perception of their teaching value, self-efficacy, and ability to thrive within their [ 41 , 42 , 43 , 44 , 45 , 46 ]. Thus, a positive relationship exists between teachers’ well-being and their efficacy in teaching [ 47 , 48 , 49 ].

Similarly, preventing teacher burnout has been linked to decreases in disruptive behavior among students and greater general stability of the classroom [ 50 , 51 , 52 ], as well as student motivation and academic commitment [ 53 , 54 , 55 , 56 ]. Teachers with low stress levels and no burnout symptoms and classes with high coping skills have been associated with enriched student outcomes [ 57 , 58 , 59 , 60 , 61 , 62 ].

1.1.2. ICT Educational Approaches

Providing a teacher training program in the ICT area to reduce the impact of COVID-19 is considered important because ICT has become a key resource in 21st-century education, particularly during the COVID-19 outbreak, during which face-to-face classes have not been possible [ 63 , 64 , 65 , 66 , 67 , 68 ]. Several studies have shown that introduction of ICT in teaching and learning improves the quality of education [ 69 , 70 , 71 , 72 ]. However, incorporating ICT into education also places significant professional demand on teachers [ 73 , 74 , 75 , 76 ]. Thus, adequate teacher training on practical ICT competency geared toward the teaching/learning process is necessary [ 77 , 78 , 79 , 80 ]. Without such training, forcing the introduction of ICT educational approaches into the curriculum and assessment process may cause considerable frustrations to both teachers and the learning community [ 81 , 82 ].

1.1.3. EI for Educational Development

Several studies have shown that teachers require EI competency to support student learning, provide opportunities for social development, and promote academic achievement and success among students [ 83 , 84 , 85 , 86 ]. Similarly, EI competency can increase teachers’ own work environment, psychosocial health, and well-being [ 87 , 88 , 89 , 90 , 91 , 92 , 93 , 94 , 95 , 96 , 97 ]. Both applications have health benefits and positive impacts on the learning community.

Developing teachers’ abilities to cope with negative affect and emotional exhaustion is key in an academic context, as it has a meaningful effect on learning processes in the classroom setting and plays a significant role in fostering students’ learning engagement [ 77 , 98 , 99 , 100 , 101 , 102 ]. It also plays a protective role among teachers, increasing the quality and health of teachers’ psychosocial environment and working life [ 103 , 104 , 105 , 106 ].

Finally, emotional exhaustion plays a key role in teacher burnout [ 18 ]. For this reason, providing a teacher training in the EI area is considered important in order to prevent emotional exhaustion and, consequently, to reduce the negative impact of COVID-19.

2. Materials and Methods

2.1. participants.

The request for participation in the study was sent to 500 teachers from three educational districts in the southeast of Spain. A total of 323 primary school teachers showed interest in participating voluntarily in the study. However, for the present study only those teachers who demonstrated a prior ICT-related training experience (an ICT course of 30 hours achieved in the academic year before) were selected ( n = 141). Consequently, these 141 primary school teachers were randomly assigned to one of two experimental conditions. The first (experimental) group ( n = 70) participated in the teacher training program intended to improve stress management and burnout in the teaching profession, competency and use of ICT to support teaching and learning, and introduction of EI-based pedagogical principles into the classroom. The second (control) group ( n = 71) did not receive any special training in strategies for managing stress or burnout, using ICT to support teaching and learning, or introducing EI into their teaching. Of the 141 teachers, 54.6% were women and 45.4% were men, and their average age was 38.4 years (SD = 6.98 years). Average teaching experience was 13.1 years (SD = 6.84 years). Finally, the procedure was approval for University of Alicante Ethics Committee (UA-2015-07-06).

2.2. Instruments

The following instruments were used in this study to measure teacher stress, burnout, emotional intelligence, and evaluation of the teacher training program.

1. Perceived Stress Questionnaire (PSQ, 20-item) [ 107 ]: The PSQ evaluates subjective experience of perceived stressful situations [ 105 ]. The scale includes 20 items (e.g., “You have many worries”) that are formulated positively and negatively to reduce acquiescence bias, and each item is assessed on a 4-point Likert-type scale from 1 (“almost never”) to 4 (“almost always”). It includes six subscales in addition to the overall score: (1) Harassment–social acceptance, (2) Overload, (3) Irritability–tension–fatigue, (4) Energy–joy, (5) Fear–anxiety, and (6) Self-realization–satisfaction. Higher scores indicate more severe subjective experience of perceived stress. Cronbach´s α values were optimal (between 0.85 and 0.87). The PSQ has been translated and validated in several languages and cultural contexts; the Spanish version was used in the present study [ 108 ].

2. Perceived Stress Scale (PSS, 14-item) [ 109 ] The PSS is a self-reported instrument that assesses level of perceived stress during the last month and consists of 14 items (e.g., “In the last month, how often have you felt that you were unable to control the important things in your life?”) rated on a 5-point scale from 0 (“never”) to 4 (“very often”). Higher scores indicate higher perceived stress. The instrument was originally designed in English [ 109 ] and has been translated and validated in several languages and cultural contexts. The European Spanish version was used in the present study and showed adequate reliability (internal consistency, α = 0.81; test-retest, r = 0.73), validity (concurrent), and sensitivity [ 110 ].

3. RED questionnaire [ 111 ]: The RED questionnaire assesses psychosocial labor risks and workplace stress due to increasingly heavy technological demands or insufficient technological competency. The scale includes 16 items rated on a 7-point scale from 0 (“never”) to 6 (“always/every day”) (e.g., “When I finish working with ICT, I feel exhausted”), and includes four subscales in addition to the overall score: (1) Skepticism, (2) Fatigue, (3) Anxiety, and (4) Inefficiency. Higher scores indicate that participants report feelings of anxiety, fatigue, skepticism and inefficacy beliefs related with the use of ICT at the workplace. The RED questionnaire was originally developed and validated for the Spanish population and demonstrates adequate reliability and validity (Cronbach’s α values were 0.93 for the Skepticism dimension, 0.92 for the Fatigue dimension, 0.83 for the Anxiety dimension, and 0.84 for the Inefficiency dimension).

4. Maslach Burnout Inventory (MBI; 22-item) [ 112 ]: The MBI includes 22 items presented as statements about the feelings and attitudes of the professional in his or her work (e.g., “I feel like I’m at the end of my rope”). It is used to measure both the frequency and intensity of professional burnout. Each item is rated on a 7-point Likert-type scale from 0 (“never”) to 6 “every day”). The instrument includes three subscales: (1) Emotional exhaustion or exhaustion, (2) Depersonalization, and (3) Personal accomplishment. Burnout syndrome is defined by high scores in the first two subscales and low scores in the third. The MBI demonstrates adequate reliability and validity (Cronbach’s α values were 0.89 for the Exhaustion or emotional exhaustion dimension, 0.77 for the Depersonalization dimension, and 0.74 for the Personal accomplishment dimension). This study used a version of the scale that has been adapted and validated for the Spanish population [ 113 ].

5. Emotional Quotient Inventory (EQ-i; 51-item) [ 23 ]: The EQ-i is a self-reported measure of emotionally and socially intelligent behavior. It has been translated into more than 30 languages and demonstrates good reliability and validity. The scale consists of 51 items (e.g., “I believe in my ability to handle most upsetting problems”) assessed on a 5-point Likert-type scale, and evaluates five subfactors of emotional intelligence in addition to the overall EQ-i score: (1) Intrapersonal intelligence, (2) Interpersonal intelligence, (3) Adaptation, (4) Stress management, and (5) Humor. This study used the Spanish version of the EQ-i [ 114 ]. Cronbach’s α values and internal consistency reliability were 0.75 for the Intrapersonal intelligence dimension, 0.77 for the Interpersonal intelligence dimension, 0.84 for the Adaptability dimension, 0.83 for the Stress management, and 0.88 for the Humor dimension.

6. Teacher training evaluation survey: A survey was designed ad-hoc by the authors to determine how satisfied teachers were with the 14-week training program after their participation. The survey consisted of 15 statements rated on a 5-point Likert scale from 1 (“totally disagree”) to 5 (“totally agree”). A comment box was also included to allow teachers to share remarks or suggestions. Please see Appendix A for more information.

2.3. Procedure

Before the training was initiated, all the teachers were fully informed of the study’s details and guaranteed confidentiality of all data obtained. On this way, the study was conducted in accordance with the Declaration of Helsinki and the recommendations and approval of the University of Alicante Ethics Committee (UA-2015-07-06). Participating teachers were randomly assigned to either the experimental or the control group. The experimental group participated in the training program over the next 14 weeks. The control group did not participate in the program or receive any other intervention during this period. In addition, on the one hand, the pretest measures were recollected two weeks before of the beginning of the program in both groups. For this goal, an e-learning Moodle platform was used because all the Spanish population was confined. On the other hand, the posttest measures were recollected two weeks after of the completion of the program in both groups. Also, in this case, the same e-learning Moodle Version 3.9.2 (Moodle.org, Western, Australia) platform was used. In conclusion, in both groups, teachers’ scores in all variables (stress levels, burnout levels, ICT competency, and EI competency) were recorded before and after the 14-week teacher training program. The experimental group also completed the teacher training evaluation survey after the training across the e-learning Moodle platform mentioned after conclusion of the training period.

2.4. Experimental Design and Data Analysis

An experimental study was carried out in this project with two groups (control vs experimental group) and two times of evaluation (pretest phase carried out before the training and the posttest phase carried out after the training). On this way, a general linear model was used to analyze the effects of the 14-week teacher training program. To ensure the equivalence of the groups, comparative analysis using Student’s t -test was conducted before initiating the teacher training program. Subsequently, teachers’ scores on all variables were analyzed using multivariate analysis of variance (MANOVA) and a univariate analysis of variance (ANOVA) of repeated measures, in which the dependent variable measured before and after the training was treated as intra-subjects variable, and the group (control/experimental) was an inter-subject variable.

Finally, the differences between the experimental and control groups were graphically represented for the variables studied. All statistical analyses and graphical representation were completed using SPSS version 22 (IBM Corporation, Armonk, NY, USA).

2.5. Teacher Training Program

First of all, the specific research objectives of the study are the following:

• To analyze if there is a significant decrease in the stress levels of teachers in the experimental group compared to those in the control group after training.
• To analyze if there is a significant decrease in the emotional exhaustion levels and depersonalization levels of teachers in the experimental group compared to those in the control group after training.
• To analyze if there is a significant increase in personal accomplishment levels of teachers in the experimental group compared to those in the control group after training.
• To analyze if there is a significant increase in emotional intelligence levels of teachers in the experimental group compared to those in the control group after training.
• To analyze if there is a significant increase in use of ICT among teachers in the experimental group compared to those in the control group after training.
• To analyze if there is a teacher’ satisfaction with the program considered like a effective training for prevented burnout and promoted psychological well-being in the teaching community.

Consequently, the present study’s 14-week teacher training program was designed to improve the key strategies described in Section 1.1.1 through Section 1.1.3 among primary school teachers. Table 1 describes the details of each lesson in the program.

Teacher Training Program.

3.1. Baseline: Before Teacher Training

The study first analyzed whether the experimental and control groups had any significant differences in (1) stress levels (PSQ and PSS), (2) burnout levels (MBI), (3) ICT competency (RED questionnaire), (4) EI competency (EQ-i), or (5) sociodemographic variables (age, teaching experience, and gender) based on an independent-samples t -test. The results showed no significant differences between the two groups in any of the variables before the teacher training program ( Table 2 ).

Student’s t -test results for difference in mean scores (before training).

3.2. Effects of the Teacher Training Program

Box’s M test revealed no homogeneity of the variance-covariance matrix for all scales. However, violation of this assumption has a minimal effect if the groups are approximately equal in size [ 115 ].

Analyzing the effects of the teacher training program, multivariate analysis indicated that all the variables showed significant variations in interaction test, obtaining the following results for the Wilks’s lambda statistic: Harassment–social acceptance (F = 147.98, η 2 partial = 0.52), Overload (F = 114.57, η 2 partial = 0.49), Irritability–tension–fatigue (F = 166.32; η 2 partial = 0.55), Energy–joy (F = 119.17, η 2 partial = 0.46), Fear–anxiety (F = 115.55, η 2 partial = 0.45), Self-realization–satisfaction (parallelism F = 112.91, η 2 partial = 0.45), and PSQ total (parallelism F = 188.18, η 2 partial = 0.57); PSS total (F = 270.49, η 2 partial = 0.66); Skepticism (parallelism F = 191.88, η 2 partial = 0.58), Fatigue (F = 255.93, η 2 partial = 0.65), Anxiety (F = 353.06, η 2 partial = 0.72), Inefficiency (parallelism F = 350.36, η 2 partial = 0.72), and RED questionnaire total (F = 359.69, η 2 partial = 0.72); Emotional exhaustion (F = 238.85, η 2 partial = 0.63), Depersonalization (F = 444.53, η 2 partial = 0.76), and personal accomplishment (F = 117.82, η 2 partial = 0.46); Intrapersonal intelligence (F = 133.79, η 2 partial = 0.49), Interpersonal intelligence (F = 134.05, η 2 partial = 0.49), Stress management (F = 264.36, η 2 partial = 0.65), Adaptation (F = 258.80, η 2 partial = 0.65), Humor (F = 258.99, η 2 partial = 0.65), and EQ-i total (F = 223.87, η 2 partial = 0.62). The F ratios in all cases were highly significant, p ≤ 0.001, and the observed power was 1.00 and all cases.

Univariate analysis showed that, as shown in Table 3 , the interaction between the evaluation time (pre-test and post-test) and the training is significant ( p ≤ 0.001) for all the variables examined in the present study.

Results of intra-/intersubject univariate ANOVA.

In order to simplify the graphic representations, Figure 1 shows interaction graphs of the direction of the differences observed between the experimental and control groups in the total score of PSQ, PSS, RED questionnaire and EQ-i. Furthermore, in the case of MBI, three factors are considered: Emotional exhaustion, Depersonalization, Personal accomplishment, because this instrument do not have a total scale score. After the 14-week training period, significant changes were observed in experimental group compared to the control group in terms of stress level (decreased total PSQ and PSS scores also represent decreases in all PSQ subscales), burnout levels (decreases in Emotional exhaustion and Depersonalization scores and increase in Personal accomplishment scores on the MBI), ICT competency (decreased total RED questionnaire scores also represent decreases in all RED subscales), and EI competency (increased total EQ-i scores also represent increases in all EQ-i subscales).

Estimated marginal means for ( a1 ) total PSQ scores and ( a2 ) total PSS scores (stress levels); ( b1 ) Emotional exhaustion scores, ( b2 ) Depersonalization scores, and ( b3 ) Personal accomplishment scores on the MBI (burnout levels); ( c ) total RED questionnaire scores (ICT competency); and ( d ) total EQ-i scores (EI competency).

Finally, the experimental group showed high agreement with nearly all statements on the satisfaction survey, indicating high satisfaction with the teacher training program. An exception was item 6 (“The objectives were appropriate for the planned duration and the established work schedule”), which received a positive (“agree” or “strongly agree”) response from 78.9% of the participants but a negative (“disagree”) response from 21.1%. Responses to the open-ended comment box indicated that this was because some teachers would have liked the training to last longer.

4. Discussion

4.1. interpreting the findings.

First of all, according with the first specific research objective, participants in the training showed a decreased in the stress levels. Second, according with the second and third specific research objective, participants in the training showed a decreased emotional exhaustion and depersonalization and an increased sense of personal accomplishment. Third, according with the fourth specific research objective, the participants in the training showed an increase in the emotional intelligence levels.

Fourthly, according with the fifth specific research objective, the teacher training program included training on the use of ICT to overcome adverse educational and social-sanitary conditions; the association of ICT methodologies with key concepts and educational innovations applied to classroom challenges; ICT-based pedagogical methodologies for lesson planning and evaluation; and ICT-based innovations and game-based e-learning methodologies to foster a high-quality teaching/learning environment in primary education. On this way, the experimental group in this study demonstrated increased levels of perceived competency and perceived ability to deal with psychosocial labor risks and workplace stress as a result of the heavy and increasing technological demands imposed by COVID-19. This indicates that after participating in the training, teachers reported less anxiety, fatigue, skepticism, and inefficacy related to using ICT in their work. Considering these results, ICT in education can facilitate educational innovations to face classroom challenges and offers customizable tools to meet teachers’ instructional needs, particularly in the face of adverse educational and social-sanitary conditions.

On the other hand, according with the sixth specific research objective, in the post-training survey, participants strongly agreed that the training program effectively improved teachers’ ICT competency. Participants also agreed with other statements related to ICT and the set of competencies acquired in the training program, such as “the knowledge that I have gained will positively affect my methods in the classroom” and “this teacher training program was a good impetus for me to improve my performance as a teacher.” Finally, the study’s important goal of ensuring transfer of the knowledge gained in the training program in actual teaching practice was evidenced by participants’ strong agreement with the statement “I consider the transfer of knowledge to my teaching practice understandable.”

4.2. Theoretical and Practical Implications of the Study

The four variables examined in this study and addressed by the teacher training program have long been recognized as important issues in teaching and are even more key given the current difficulties posed by COVID-19 in primary education. Teaching is an occupation with high stress and emotional demands [ 88 , 116 , 117 , 118 ]. This study’s teacher training was found to effectively decrease teachers’ perceived stress levels and increase their EI. The evaluation survey revealed that teachers who participated in the program found it helpful in developing these two competencies and implementing this knowledge in the classroom (e.g., participants reported strong agreement with statements such as “the training program effectively improves teachers’ ability to cope with stress” and “the training program effectively trains teachers to apply EI competencies to educational management”).

This study aimed to incorporate a practical approach. Thus, the teacher training program incorporated practical tools, methodologies, and resources to help teachers manage their stress in the classroom, both in normal educational contexts and in adverse social and health conditions; improve their understanding of emotions (academic self-esteem, self-realization, and emotional awareness), ability to identify emotions (improving intrapersonal and interpersonal relationships, increasing empathy and social responsibility in crisis situations), emotional expression, and ability to leverage positive emotions; implement EI strategies and improve their effectiveness in frontline teaching work; and promote positive emotions in educational plans to face and resolve challenges. In the post-training survey, participants strongly agreed that the teacher training program was helpful for developing innovative and successful teaching methodologies and educational plans as a strategic response to address the pandemic. Furthermore, training and improving these competencies are key for teaching because managing stress and applying EI in the classroom are linked to favorable school policies and practices [ 119 , 120 , 121 , 122 , 123 , 124 , 125 , 126 , 127 , 128 , 129 , 130 ].

Preventing teacher burnout is also essential for optimal teacher effectiveness [ 131 , 132 , 133 , 134 , 135 , 136 ]. Burnout syndrome appears due to chronically high levels of stress, overwhelming negative emotions, and sustained work-related exhaustion. The teacher training program in this study incorporated methodologies aimed to prevent burnout along with practical strategies and resources to safeguard psychological well-being in situations of adversity and health emergency.

Training in ICT competency was included in this study’s teacher training program for several reasons. First, the scientific literature shows that ICT competency is an emerging requirement for innovative and high-quality teaching [ 137 , 138 , 139 , 140 , 141 , 142 , 143 , 144 , 145 ]. Second, methodologies based on ICT are useful educational tools valid path towards the teaching [ 146 , 147 , 148 , 149 , 150 ]. Third, creative educational approaches are needed to mitigate the impact of COVID-19 on the learning community [ 151 , 152 , 153 ]. Finally, workable solutions are required to cope with health situations that prevent or restrict face-to-face teaching [ 154 , 155 , 156 ].

4.3. Limitations and Future Research Directions

This study presents a practical training program to improve teacher competencies and help them mitigate the impact of the COVID-19 crisis on primary education. However, several limitations must be considered. First, the sample size was small. A future goal is to develop an improved version of this training program and implement it with a larger sample of teachers. Additionally, the initiative is small in scope. Unfortunately, there is currently no global or standard educational response to the COVID-19 pandemic; each country is addressing the crisis based on their own opportunities and educational expertise. In future, the scope of the training should be improved to adapt the goals and methodologies to international educational contexts.

Second, in the light of the study’s results, future research must confirm the long-term sustainability of the methodologies trained in the program and their impact on students’ academic achievement. Although this study intended to analyze academic achievement reports, educational authorities have strongly promoted positive academic reports during the pandemic, and thus academic reports published during the COVID-19 outbreak cannot be considered accurate.

Finally, transfer of the methodologies acquired in the teacher training program requires time, and logistical and pedagogical considerations must be guaranteed. The study found that the program’s strategies effectively improved teachers’ ability to cope with stress and avoid burnout, but teacher overload should be realistically considered, taking into account the time and resources that are available. Similarly, knowledge from the teacher training program could be transferred to implement meaningful EI learning into the classroom and promote a free and constructive environment, but a school culture of respect between the members of the learning community is necessary for this to succeed. Finally, the ICT training for teachers in the program is accurate, but the digital gap must still be fought. The entire educational community must have the technical resources and skills to access classes designed in an e-learning environment. For the proposed methodologies to be viable, these educational conditions must be met before teacher training is implemented. Further research on these aspects is necessary.

5. Conclusions

The challenges of the COVID-19 health situation in education have led to the emergence of a teacher overexertion and a depth adaptation to the new environment demands on classroom. Numerous questions have been raised about the short- and long-term impacts of the COVID-19 pandemic on the Primary Education. New demands on current education systems are extensive and have developed suddenly and unpredictably. However, there is consensus that training teacher competencies is key in responding to the pandemic situation. The purpose of this study is to develop a teacher training for support teachers, who perform frontline work in schools, in coping with stress, preventing burnout, improving their information and communications technology and introducing the principles of emotional intelligence in the classroom. Accordingly, the results have shown that fostering teachers’ well-being (through stress management and prevention of burnout) is possible, as well as equipping teachers with the ICT competency necessary to implement new educational practices and incorporating EI into teaching. And consequently, raising teachers’ ability to strategically create a positive emotional atmosphere within the learning community.

Acknowledgments

We are thankful to the editors and the three anonymous reviewers for their helpful comments. They helped to improve the presentation and discussion of our topic a lot.

Appendix A. Description about Teacher Training Evaluation Survey

A survey was designed ad-hoc by the authors to determine how satisfied teachers were with the 14-week training program after their participation. The survey consisted of 15 statements rated on a 5-point Likert scale from 1 (“totally disagree”) to 5 (“totally agree”). A comment box was also included to allow teachers to share remarks or suggestions. The 15 items were as follows:

• The teacher training program effectively improves teachers’ ability to cope with stress.
• The teacher training program effectively prevents burnout among teachers.
• The teacher training program effectively improves teachers’ ICT competency.
• The teacher training program effectively trains teachers to apply EI competencies to educational management.
• The teacher training program is helpful for developing innovative and successful teaching methodologies and educational plans as a strategic response to address the pandemic.
• The objectives were appropriate for the planned duration and the established work schedule.
• The methods used were effective.
• The transfer of knowledge to my teaching practice was understandable.
• The knowledge I have gained will positively affect my methods in the classroom.
• The 14-week teacher training program met my expectations.
• My motivation and interest during the 14-week teacher training program were high.
• The teacher training program fosters good EI in teachers to promote psychological well-being in the teaching community.
• The teacher training program effectively expands teachers’ commitment to quality education and addresses future gaps.
• The training was engaging and beneficial for improving academic achievement based on positive emotions in primary school.
• This teacher training program was a good impetus for me to improve my performance as a teacher.

Author Contributions

Conceptualization, T.P.-R. and R.G.-C.; methodology, R.G.-C. and J.-L.C.; investigation, T.P.-R.; resources, A.I.; data curation, T.P.-R.; writing—original draft preparation, T.P.-R., R.G.-C., A.I. and J.-L.C.; funding acquisition, R.G.-C. All authors have read and agreed to the published version of the manuscript.

This research was supported by the Spanish Ministry of Economy and Competitiveness (EDU2015-64562-R).

Conflicts of Interest

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Search Menu
• Sign in through your institution
• Browse content in A - General Economics and Teaching
• Browse content in A1 - General Economics
• A12 - Relation of Economics to Other Disciplines
• A14 - Sociology of Economics
• Browse content in B - History of Economic Thought, Methodology, and Heterodox Approaches
• Browse content in B4 - Economic Methodology
• B41 - Economic Methodology
• Browse content in C - Mathematical and Quantitative Methods
• Browse content in C1 - Econometric and Statistical Methods and Methodology: General
• C18 - Methodological Issues: General
• Browse content in C2 - Single Equation Models; Single Variables
• C21 - Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions
• Browse content in C3 - Multiple or Simultaneous Equation Models; Multiple Variables
• C38 - Classification Methods; Cluster Analysis; Principal Components; Factor Models
• Browse content in C5 - Econometric Modeling
• C59 - Other
• Browse content in C8 - Data Collection and Data Estimation Methodology; Computer Programs
• C80 - General
• C81 - Methodology for Collecting, Estimating, and Organizing Microeconomic Data; Data Access
• C83 - Survey Methods; Sampling Methods
• Browse content in C9 - Design of Experiments
• C93 - Field Experiments
• Browse content in D - Microeconomics
• Browse content in D0 - General
• D02 - Institutions: Design, Formation, Operations, and Impact
• D03 - Behavioral Microeconomics: Underlying Principles
• D04 - Microeconomic Policy: Formulation; Implementation, and Evaluation
• Browse content in D1 - Household Behavior and Family Economics
• D10 - General
• D12 - Consumer Economics: Empirical Analysis
• D14 - Household Saving; Personal Finance
• Browse content in D2 - Production and Organizations
• D22 - Firm Behavior: Empirical Analysis
• D24 - Production; Cost; Capital; Capital, Total Factor, and Multifactor Productivity; Capacity
• Browse content in D3 - Distribution
• D31 - Personal Income, Wealth, and Their Distributions
• Browse content in D6 - Welfare Economics
• D61 - Allocative Efficiency; Cost-Benefit Analysis
• D62 - Externalities
• D63 - Equity, Justice, Inequality, and Other Normative Criteria and Measurement
• Browse content in D7 - Analysis of Collective Decision-Making
• D72 - Political Processes: Rent-seeking, Lobbying, Elections, Legislatures, and Voting Behavior
• D73 - Bureaucracy; Administrative Processes in Public Organizations; Corruption
• D74 - Conflict; Conflict Resolution; Alliances; Revolutions
• Browse content in D8 - Information, Knowledge, and Uncertainty
• D83 - Search; Learning; Information and Knowledge; Communication; Belief; Unawareness
• D85 - Network Formation and Analysis: Theory
• D86 - Economics of Contract: Theory
• Browse content in D9 - Micro-Based Behavioral Economics
• D91 - Role and Effects of Psychological, Emotional, Social, and Cognitive Factors on Decision Making
• D92 - Intertemporal Firm Choice, Investment, Capacity, and Financing
• Browse content in E - Macroeconomics and Monetary Economics
• Browse content in E2 - Consumption, Saving, Production, Investment, Labor Markets, and Informal Economy
• E23 - Production
• E24 - Employment; Unemployment; Wages; Intergenerational Income Distribution; Aggregate Human Capital; Aggregate Labor Productivity
• Browse content in E4 - Money and Interest Rates
• E42 - Monetary Systems; Standards; Regimes; Government and the Monetary System; Payment Systems
• Browse content in E5 - Monetary Policy, Central Banking, and the Supply of Money and Credit
• E52 - Monetary Policy
• E58 - Central Banks and Their Policies
• Browse content in E6 - Macroeconomic Policy, Macroeconomic Aspects of Public Finance, and General Outlook
• E60 - General
• E61 - Policy Objectives; Policy Designs and Consistency; Policy Coordination
• E62 - Fiscal Policy
• E65 - Studies of Particular Policy Episodes
• Browse content in F - International Economics
• Browse content in F0 - General
• F01 - Global Outlook
• Browse content in F1 - Trade
• F10 - General
• F11 - Neoclassical Models of Trade
• F13 - Trade Policy; International Trade Organizations
• F14 - Empirical Studies of Trade
• F15 - Economic Integration
• F19 - Other
• Browse content in F2 - International Factor Movements and International Business
• F21 - International Investment; Long-Term Capital Movements
• F22 - International Migration
• F23 - Multinational Firms; International Business
• Browse content in F3 - International Finance
• F32 - Current Account Adjustment; Short-Term Capital Movements
• F34 - International Lending and Debt Problems
• F35 - Foreign Aid
• F36 - Financial Aspects of Economic Integration
• Browse content in F4 - Macroeconomic Aspects of International Trade and Finance
• F41 - Open Economy Macroeconomics
• F42 - International Policy Coordination and Transmission
• F43 - Economic Growth of Open Economies
• Browse content in F5 - International Relations, National Security, and International Political Economy
• F50 - General
• F52 - National Security; Economic Nationalism
• F53 - International Agreements and Observance; International Organizations
• F55 - International Institutional Arrangements
• Browse content in F6 - Economic Impacts of Globalization
• F61 - Microeconomic Impacts
• F63 - Economic Development
• F66 - Labor
• Browse content in G - Financial Economics
• Browse content in G0 - General
• G01 - Financial Crises
• Browse content in G1 - General Financial Markets
• G10 - General
• G15 - International Financial Markets
• G18 - Government Policy and Regulation
• Browse content in G2 - Financial Institutions and Services
• G20 - General
• G21 - Banks; Depository Institutions; Micro Finance Institutions; Mortgages
• G22 - Insurance; Insurance Companies; Actuarial Studies
• G23 - Non-bank Financial Institutions; Financial Instruments; Institutional Investors
• G28 - Government Policy and Regulation
• Browse content in G3 - Corporate Finance and Governance
• G32 - Financing Policy; Financial Risk and Risk Management; Capital and Ownership Structure; Value of Firms; Goodwill
• G33 - Bankruptcy; Liquidation
• G38 - Government Policy and Regulation
• Browse content in H - Public Economics
• Browse content in H1 - Structure and Scope of Government
• H11 - Structure, Scope, and Performance of Government
• Browse content in H2 - Taxation, Subsidies, and Revenue
• H20 - General
• H23 - Externalities; Redistributive Effects; Environmental Taxes and Subsidies
• H25 - Business Taxes and Subsidies
• H26 - Tax Evasion and Avoidance
• H27 - Other Sources of Revenue
• Browse content in H3 - Fiscal Policies and Behavior of Economic Agents
• H31 - Household
• Browse content in H4 - Publicly Provided Goods
• H41 - Public Goods
• H43 - Project Evaluation; Social Discount Rate
• Browse content in H5 - National Government Expenditures and Related Policies
• H52 - Government Expenditures and Education
• H53 - Government Expenditures and Welfare Programs
• H54 - Infrastructures; Other Public Investment and Capital Stock
• H55 - Social Security and Public Pensions
• H56 - National Security and War
• H57 - Procurement
• Browse content in H6 - National Budget, Deficit, and Debt
• H60 - General
• H61 - Budget; Budget Systems
• Browse content in H7 - State and Local Government; Intergovernmental Relations
• H71 - State and Local Taxation, Subsidies, and Revenue
• H75 - State and Local Government: Health; Education; Welfare; Public Pensions
• H77 - Intergovernmental Relations; Federalism; Secession
• Browse content in H8 - Miscellaneous Issues
• H83 - Public Administration; Public Sector Accounting and Audits
• H84 - Disaster Aid
• Browse content in I - Health, Education, and Welfare
• Browse content in I0 - General
• I00 - General
• Browse content in I1 - Health
• I10 - General
• I12 - Health Behavior
• I15 - Health and Economic Development
• I18 - Government Policy; Regulation; Public Health
• Browse content in I2 - Education and Research Institutions
• I20 - General
• I21 - Analysis of Education
• I22 - Educational Finance; Financial Aid
• I24 - Education and Inequality
• I25 - Education and Economic Development
• I28 - Government Policy
• Browse content in I3 - Welfare, Well-Being, and Poverty
• I30 - General
• I31 - General Welfare
• I32 - Measurement and Analysis of Poverty
• I38 - Government Policy; Provision and Effects of Welfare Programs
• Browse content in J - Labor and Demographic Economics
• Browse content in J0 - General
• J01 - Labor Economics: General
• J08 - Labor Economics Policies
• Browse content in J1 - Demographic Economics
• J10 - General
• J11 - Demographic Trends, Macroeconomic Effects, and Forecasts
• J12 - Marriage; Marital Dissolution; Family Structure; Domestic Abuse
• J13 - Fertility; Family Planning; Child Care; Children; Youth
• J15 - Economics of Minorities, Races, Indigenous Peoples, and Immigrants; Non-labor Discrimination
• J16 - Economics of Gender; Non-labor Discrimination
• J17 - Value of Life; Forgone Income
• J18 - Public Policy
• Browse content in J2 - Demand and Supply of Labor
• J21 - Labor Force and Employment, Size, and Structure
• J22 - Time Allocation and Labor Supply
• J23 - Labor Demand
• J24 - Human Capital; Skills; Occupational Choice; Labor Productivity
• J26 - Retirement; Retirement Policies
• J28 - Safety; Job Satisfaction; Related Public Policy
• Browse content in J3 - Wages, Compensation, and Labor Costs
• J38 - Public Policy
• Browse content in J4 - Particular Labor Markets
• J48 - Public Policy
• Browse content in J5 - Labor-Management Relations, Trade Unions, and Collective Bargaining
• J58 - Public Policy
• Browse content in J6 - Mobility, Unemployment, Vacancies, and Immigrant Workers
• J61 - Geographic Labor Mobility; Immigrant Workers
• J62 - Job, Occupational, and Intergenerational Mobility
• J63 - Turnover; Vacancies; Layoffs
• J68 - Public Policy
• Browse content in J8 - Labor Standards: National and International
• J88 - Public Policy
• Browse content in K - Law and Economics
• Browse content in K2 - Regulation and Business Law
• K23 - Regulated Industries and Administrative Law
• Browse content in K3 - Other Substantive Areas of Law
• K32 - Environmental, Health, and Safety Law
• K34 - Tax Law
• Browse content in K4 - Legal Procedure, the Legal System, and Illegal Behavior
• K40 - General
• K41 - Litigation Process
• K42 - Illegal Behavior and the Enforcement of Law
• Browse content in L - Industrial Organization
• Browse content in L1 - Market Structure, Firm Strategy, and Market Performance
• L11 - Production, Pricing, and Market Structure; Size Distribution of Firms
• L14 - Transactional Relationships; Contracts and Reputation; Networks
• L16 - Industrial Organization and Macroeconomics: Industrial Structure and Structural Change; Industrial Price Indices
• Browse content in L2 - Firm Objectives, Organization, and Behavior
• L20 - General
• L23 - Organization of Production
• L25 - Firm Performance: Size, Diversification, and Scope
• L26 - Entrepreneurship
• Browse content in L3 - Nonprofit Organizations and Public Enterprise
• L33 - Comparison of Public and Private Enterprises and Nonprofit Institutions; Privatization; Contracting Out
• Browse content in L5 - Regulation and Industrial Policy
• L51 - Economics of Regulation
• L52 - Industrial Policy; Sectoral Planning Methods
• Browse content in L9 - Industry Studies: Transportation and Utilities
• L94 - Electric Utilities
• L97 - Utilities: General
• L98 - Government Policy
• Browse content in M - Business Administration and Business Economics; Marketing; Accounting; Personnel Economics
• Browse content in M5 - Personnel Economics
• M53 - Training
• Browse content in N - Economic History
• Browse content in N1 - Macroeconomics and Monetary Economics; Industrial Structure; Growth; Fluctuations
• N10 - General, International, or Comparative
• Browse content in N3 - Labor and Consumers, Demography, Education, Health, Welfare, Income, Wealth, Religion, and Philanthropy
• N35 - Asia including Middle East
• Browse content in N5 - Agriculture, Natural Resources, Environment, and Extractive Industries
• N55 - Asia including Middle East
• N57 - Africa; Oceania
• Browse content in N7 - Transport, Trade, Energy, Technology, and Other Services
• N77 - Africa; Oceania
• Browse content in O - Economic Development, Innovation, Technological Change, and Growth
• Browse content in O1 - Economic Development
• O10 - General
• O11 - Macroeconomic Analyses of Economic Development
• O12 - Microeconomic Analyses of Economic Development
• O13 - Agriculture; Natural Resources; Energy; Environment; Other Primary Products
• O14 - Industrialization; Manufacturing and Service Industries; Choice of Technology
• O15 - Human Resources; Human Development; Income Distribution; Migration
• O16 - Financial Markets; Saving and Capital Investment; Corporate Finance and Governance
• O17 - Formal and Informal Sectors; Shadow Economy; Institutional Arrangements
• O18 - Urban, Rural, Regional, and Transportation Analysis; Housing; Infrastructure
• O19 - International Linkages to Development; Role of International Organizations
• Browse content in O2 - Development Planning and Policy
• O20 - General
• O22 - Project Analysis
• O23 - Fiscal and Monetary Policy in Development
• O24 - Trade Policy; Factor Movement Policy; Foreign Exchange Policy
• O25 - Industrial Policy
• Browse content in O3 - Innovation; Research and Development; Technological Change; Intellectual Property Rights
• O31 - Innovation and Invention: Processes and Incentives
• O32 - Management of Technological Innovation and R&D
• O33 - Technological Change: Choices and Consequences; Diffusion Processes
• O34 - Intellectual Property and Intellectual Capital
• O38 - Government Policy
• Browse content in O4 - Economic Growth and Aggregate Productivity
• O40 - General
• O41 - One, Two, and Multisector Growth Models
• O43 - Institutions and Growth
• O44 - Environment and Growth
• O47 - Empirical Studies of Economic Growth; Aggregate Productivity; Cross-Country Output Convergence
• Browse content in O5 - Economywide Country Studies
• O55 - Africa
• O57 - Comparative Studies of Countries
• Browse content in P - Economic Systems
• Browse content in P1 - Capitalist Systems
• P14 - Property Rights
• Browse content in P2 - Socialist Systems and Transitional Economies
• P26 - Political Economy; Property Rights
• Browse content in P3 - Socialist Institutions and Their Transitions
• P30 - General
• Browse content in P4 - Other Economic Systems
• P43 - Public Economics; Financial Economics
• P46 - Consumer Economics; Health; Education and Training; Welfare, Income, Wealth, and Poverty
• P48 - Political Economy; Legal Institutions; Property Rights; Natural Resources; Energy; Environment; Regional Studies
• Browse content in Q - Agricultural and Natural Resource Economics; Environmental and Ecological Economics
• Browse content in Q0 - General
• Q01 - Sustainable Development
• Browse content in Q1 - Agriculture
• Q10 - General
• Q12 - Micro Analysis of Farm Firms, Farm Households, and Farm Input Markets
• Q13 - Agricultural Markets and Marketing; Cooperatives; Agribusiness
• Q14 - Agricultural Finance
• Q15 - Land Ownership and Tenure; Land Reform; Land Use; Irrigation; Agriculture and Environment
• Q16 - R&D; Agricultural Technology; Biofuels; Agricultural Extension Services
• Q17 - Agriculture in International Trade
• Q18 - Agricultural Policy; Food Policy
• Browse content in Q2 - Renewable Resources and Conservation
• Q25 - Water
• Browse content in Q3 - Nonrenewable Resources and Conservation
• Q30 - General
• Q33 - Resource Booms
• Browse content in Q4 - Energy
• Q41 - Demand and Supply; Prices
• Q43 - Energy and the Macroeconomy
• Browse content in Q5 - Environmental Economics
• Q51 - Valuation of Environmental Effects
• Q52 - Pollution Control Adoption Costs; Distributional Effects; Employment Effects
• Q54 - Climate; Natural Disasters; Global Warming
• Q56 - Environment and Development; Environment and Trade; Sustainability; Environmental Accounts and Accounting; Environmental Equity; Population Growth
• Q57 - Ecological Economics: Ecosystem Services; Biodiversity Conservation; Bioeconomics; Industrial Ecology
• Q58 - Government Policy
• Browse content in R - Urban, Rural, Regional, Real Estate, and Transportation Economics
• Browse content in R1 - General Regional Economics
• R11 - Regional Economic Activity: Growth, Development, Environmental Issues, and Changes
• R12 - Size and Spatial Distributions of Regional Economic Activity
• R13 - General Equilibrium and Welfare Economic Analysis of Regional Economies
• R14 - Land Use Patterns
• Browse content in R2 - Household Analysis
• R20 - General
• R23 - Regional Migration; Regional Labor Markets; Population; Neighborhood Characteristics
• R28 - Government Policy
• Browse content in R3 - Real Estate Markets, Spatial Production Analysis, and Firm Location
• R38 - Government Policy
• Browse content in R4 - Transportation Economics
• R40 - General
• R41 - Transportation: Demand, Supply, and Congestion; Travel Time; Safety and Accidents; Transportation Noise
• R48 - Government Pricing and Policy
• Browse content in R5 - Regional Government Analysis
• R52 - Land Use and Other Regulations
• Browse content in Y - Miscellaneous Categories
• Y8 - Related Disciplines
• Browse content in Z - Other Special Topics
• Browse content in Z1 - Cultural Economics; Economic Sociology; Economic Anthropology
• Z13 - Economic Sociology; Economic Anthropology; Social and Economic Stratification
• Advance articles
• Author Guidelines
• Open Access
• About The World Bank Research Observer
• About the World Bank
• Editorial Board
• Advertising and Corporate Services
• Self-Archiving Policy
• Dispatch Dates
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

• < Previous

Teacher Professional Development around the World: The Gap between Evidence and Practice

• Article contents
• Figures & tables
• Supplementary Data

Anna Popova, David K Evans, Mary E Breeding, Violeta Arancibia, Teacher Professional Development around the World: The Gap between Evidence and Practice, The World Bank Research Observer , Volume 37, Issue 1, February 2022, Pages 107–136, https://doi.org/10.1093/wbro/lkab006

• Permissions Icon Permissions

Many teachers in low- and middle-income countries lack the skills to teach effectively, and professional development (PD) programs are the principal tool that governments use to upgrade those skills. At the same time, few PD programs are evaluated, and those that are evaluated show highly varying results. This paper proposes a set of indicators—the In-Service Teacher Training Survey Instrument—to standardize reporting on teacher PD programs. An application of the instrument to 33 rigorously evaluated PD programs shows that programs that link participation to career incentives, have a specific subject focus, incorporate lesson enactment in the training, and include initial face-to-face training tend to show higher student learning gains. In qualitative interviews, program implementers also report follow-up visits as among the most effective characteristics of their professional development programs. This paper then uses the instrument to present novel data on a sample of 139 government-funded, at-scale professional development programs across 14 countries. The attributes of most at-scale teacher professional development programs differ sharply from those of programs that evidence suggests are effective, with fewer incentives to participate in PD, fewer opportunities to practice new skills, and less follow-up once teachers return to their classrooms.

Good teachers have a major impact on student performance, both over the course of the school year ( Araujo et al. 2016 ) and into adulthood ( Chetty, Friedman, and Rockoff 2014 ). However, teachers in low- and middle-income countries often lack the skills they need to teach students effectively. Across seven African countries, only seven percent of fourth-grade teachers had the minimum knowledge necessary to teach language; in four countries, the statistic was zero percent. For math teaching, 68 percent had the minimum knowledge needed to teach math—higher than the seven percent for language, but still leaving one in three teachers with insufficient knowledge. Teachers also scored woefully low in terms of pedagogical knowledge—their ability to prepare a lesson, formulate questions that would elicit student knowledge effectively, and their performance in the classroom ( Bold et al. 2017 ).

The principal tool that countries across the income spectrum use to improve the knowledge and skills of their practicing teachers is professional development (PD), which refers to on-the-job training activities ranging from formal, lecture-style training to mentoring and coaching. However, few PD programs are rigorously evaluated, and among those that are, the evidence of their effectiveness is wildly mixed. Some programs are effective: training teachers to provide literacy instruction using students’ mother tongue in Uganda and training teachers to evaluate student performance more regularly and adjust teaching based on those evaluations in Liberia both had sizeable impacts on student reading ability ( Piper and Korda 2011 ; Kerwin and Thornton 2021 ). Others demonstrate opposite results: a large-scale, government-implemented PD program in China had zero impact on teacher knowledge, teaching practices, or student learning outcomes ( Loyalka et al. 2019 ), and a program that trained teachers to engage their middle school math students more actively in learning in Costa Rica resulted in worse learning outcomes for students ( Berlinski and Busso 2017 ). Indeed, there is much more variation in effectiveness across teacher training programs than across education programs more broadly ( McEwan 2015 ; Evans and Popova 2016a ). With this limited and highly variable evidence, policymakers and practitioners may be left puzzled as to how to structure teacher PD programs effectively.

In this paper, we propose a set of indicators—the In-service Teacher Training Survey Instrument, or ITTSI—to allow comparisons across teacher PD programs with varying impacts. On average, existing studies of PD programs only report on about half of these indicators. We supplement that information through interviews with implementors of evaluated PD programs. We compare the characteristics of 33 rigorously evaluated PD programs to identify which characteristics are associated with larger student learning gains. We then gather data from 139 government-funded, at-scale PD programs across 14 countries. Like most at-scale government programs, none of these programs have been evaluated rigorously. We compare the two samples to examine whether the PD programs that most teachers actually experience exhibit similar characteristics to those of PD programs that have been evaluated and shown to produce sizeable student learning gains.

When we apply our instrument to evaluated PD programs, results suggest that programs deliver high student learning gains when they link participation in PD to incentives such as promotion or salary implications, when they have a specific subject focus, when teachers practice enacting lessons during the training, and when training has at least an initial face-to-face aspect. Meanwhile, program implementers highlight two characteristics of effective training in interviews: mentoring follow-up visits after the PD training, and complementary materials such as structured lesson plans to help teachers apply what they have learned during PD.

When we subsequently use the ITTSI to characterize a sample of at-scale, government-funded PD programs around the world, we find a divergence in the characteristics common to these programs and those that typify evaluated programs that were found to be effective. Relative to top-performing PD programs—defined as those found to be the most effective at increasing student learning—very few at-scale PD programs are linked to any sort of career opportunities, such as promotion or salary implications. Similarly, in-school follow-up support and including time to practice with other teachers is less common among at-scale PD programs. This highlights a substantial gap between the kind of teacher PD supported by research and that currently being provided by many government-funded, at-scale programs.

These results have implications for both researchers and policymakers. For researchers, future evaluations will contribute much more to an understanding of how to improve teachers’ skills if they report more details of the characteristics of the PD programs. Our proposed set of indicators can serve as a guide. For policymakers, at-scale PD programs should incorporate more aspects of successful, evaluated PD programs, such as incentives, practice, and follow-up in-school support. For both, more programs can be evaluated at scale, using government delivery systems, in order to improve the skills of teachers in the future.

Conceptual Framework

The defining attributes of teacher professional development programs fall principally into three categories. The first is the content of the PD program: What is taught? The second is the delivery of the PD program: Who is teaching, when, and for how long? The third is the organization of the program beyond content and delivery: What are the scale and resources of the program? Are there incentives for participation? Was it designed based on a diagnostic of teachers? In this section, we discuss the theory behind each of these three categories.

On the content, PD programs focused on subject-specific pedagogy are likely to be most effective. General pedagogical knowledge—i.e., broad strategies of classroom management and organization—may contribute to student learning, driving the recent development of a range of classroom observation instruments ( La Paro and Pianta 2003 ; Molina et al. 2018 ). However, different subjects require radically different pedagogies ( Shulman 1986 ; Villegas-Reimers 2003 ). A highly scripted approach may work to teach early grade reading, whereas teaching science or civics in later grades—for example—may require more flexible approaches. PD programs that focus on arming teachers with subject-specific pedagogy are thus likely to make the largest contribution to student learning.

With respect to the delivery, the method, trainers, duration, and location of instruction all play a role. First, because working, professional teachers are the students in PD, principles of adult education are relevant to the method of instruction. Adult education tends to work best with clear applications rather than a theoretical focus ( Cardemil 2001 ; Knowles, Holton, and Swanson 2005 ). The method of instruction should include concrete, realistic goals ( Baker and Smith 1999 ) and the teaching of formative evaluation so that teachers can effectively evaluate their own progress towards their teaching goals ( Bourgeois and Nizet 1997 ). Second, the quality of trainers—i.e., those providing the PD—is crucial to learning ( Knowles, Holton, and Swanson 2005 ). In terms of the delivery of PD, this calls into question the common cascade model of PD in low-income environments, in which both information and pedagogical ability may be diluted as a master trainer trains another individual as a trainer, who may go on to train another trainer below her, and so forth.

Third, on the duration of instruction, there is no theoretical consensus on exactly how long training should last, although there is suggestive empirical evidence in the literature in favor of sustained contact over a significant period of time and against brief, one-time workshops ( Desimone 2009 ). Fourth, on the location of instruction, teacher PD in the school (“embedded”) is likely to be most effective so that participating teachers can raise concrete problems that they face in the local environment, and they can also receive feedback on actual teaching ( Wood and McQuarrie 1999 ). However, this will depend on the environment. In very difficult teaching environments, some degree of training outside the school may facilitate focus on the part of the trainees ( Kraft and Papay 2014 ).

Finally, the organization of the PD—which includes overarching aspects such as who is organizing it, for whom, and how—provides an important backdrop when we consider any PD program. This includes aspects such as the scale, cost, and targeting of the program. In general, it is predictably easier to provide high-quality PD through smaller scale, higher cost programs that provide more tailored attention to a given teacher. In terms of targeting, teacher PD will work best if it adjusts at different points in the teachers’ careers: One would not effectively teach a brand-new teacher in the same way as one would train a teacher with 20 years of experience ( Huberman 1989 ). Teachers see their greatest natural improvements in the first five years of teaching, which may be an indicator of greater skill plasticity, so there may be benefits to leveraging that time ( TNTP 2015 ).

What Works in High-Income Countries?

A full review of the literature in high-income countries is beyond the scope of this study. However, it may be useful to highlight recent work on in-service teacher PD from the United States—which spends almost $18,000 per teacher and 19 days of teacher time on training each year ( TNTP 2015 )—and other high-income countries, in order to ensure that low- and middle-income countries are not ignoring well-established evidence. Several promising themes that emerge from this work are the importance of making PD specific and practical, providing sustained follow-up support for teachers, and embedding it in the curriculum.

Specific and practical teacher PD finds support from multiple reviews of teacher PD studies in high-income countries, which conclude that concrete, classroom-based programs make the most difference to teachers ( Darling-Hammond et al. 2009 ; Walter and Briggs 2012 ). More recently, a meta-analysis of 196 randomized evaluations of education interventions—not just PD—in the United States that measure student test scores as an outcome examined the impact of both “general” and “managed” professional development, relative to other interventions ( Fryer 2017 ). General PD may focus on classroom management or increasing the rigor of teachers’ knowledge, whereas managed professional development prescribes a specific method, with detailed instructions on implementation and follow-up support. On average, managed PD increased student test scores by 2.5 times (0.052 standard deviations) as much as general PD and was at least as effective as the combined average of all school-based interventions. A recent review of nearly 2,000 impact estimates from 747 randomized controlled trials of education interventions in the United States proposes that an effect size of 0.05 be considered a “medium” effect size, higher than the average effect size, weighted by study sample size ( Kraft 2020 ), which suggests that these are not trivial impacts.

The importance of sustained follow-up support is echoed by another U.S.-focused review, which found that PD programs with significant contact hours (between 30 and 100 in total) over the course of six to twelve months were more effective at raising student test scores ( Yoon et al. 2007 ). Likewise, a narrative review of U.S. studies concluded that the most effective programs are not “one-shot workshops”: they are sustained, intense, and embedded in the curriculum ( Darling-Hammond et al. 2009 ).

Despite these conclusions, the experimental or quasi-experimental evidence is thin, even in high-income countries. The meta-analysis of 196 evaluations of education interventions included just nine PD studies ( Fryer 2017 ), and another review of 1,300 PD studies identified just nine that had pre- and post-test data and some sort of control group ( Yoon et al. 2007 ). Similarly, a review of PD in mathematics found more than 600 studies of math PD interventions, but only 32 used any research design to measure effectiveness, and only five of those were high-quality randomized trials ( Gersten et al. 2014 ). The question of what drives effective teacher PD remains understudied, even in high-income environments.

We expect teachers in lower and middle-income countries to learn in fundamentally similar ways to their high-income counterparts. However, lower resource contexts are typically characterized by more binding cost constraints and lower teacher and coach pedagogical capacity. These challenges may make certain elements of PD programs more and less relevant in lower-income contexts. Teachers and coaches in low- and middle-income countries may benefit from more prescriptive instructions on implementation and, while they too require ongoing follow-up as part of PD, this may need to be provided in lower-cost forms, whether in group sessions, using technology for remote coaching, or training school principals and experienced peer teachers as coaches.

To understand which characteristics of PD programs are associated with student test score gains, and to analyze the degree to which these effective characteristics are incorporated into at-scale PD programs in practice, we first developed a standardized instrument to characterize in-service teacher training. Second, we applied this instrument to already evaluated PD programs to understand which PD characteristics are associated with student learning gains. Third, we applied the survey instrument to a sample of at-scale PD programs to see how these programs line up with what the evidence suggests works in teacher training. The information we present thus comes from two different samples of PD programs: One sample of evaluated PD programs, those with impact evaluations that include student assessment results; and one sample of at-scale , government-funded PD programs. 1 The remainder of this section introduces the instrument briefly before describing its application to each of the two samples.

The In-Service Teacher Training Survey Instrument (ITTSI)

The ITTSI was designed based on the conceptual framework and empirical literature characterized in the previous sections, as well as on the authors’ prior experience studying in-service teacher PD. We drafted an initial list of 51 key indicators to capture details about a range of program characteristics falling into three main categories: Organization, Content, and Delivery, paralleling the three elements of our conceptual framework ( fig. 1 ). We supplement those categories with a fourth category, Perceptions, which we added to collect qualitative data from program implementors.

Summary of the In-Service Teacher Training Survey Instrument (ITTSI)

Source : Authors’ summary of the elements of the In-Service Teacher Training Survey Instrument, as detailed in supplementary online appendices A1 and A2 .

Taking each of these in turn, the Organization section includes items such as the type of organization responsible for the design and implementation of a given teacher training program, to whom the program is targeted, what (if any) complementary materials it provides, the scale of the program, and its cost. The Content section includes indicators capturing the type of knowledge or skills that a given program aims to build among beneficiary teachers, such as whether the program focuses on subject content (and if so, which subject), pedagogy, new technology, classroom management, counseling, assessment, or some combination.

Delivery focuses on indicators capturing program implementation details, such as whether it is delivered through a cascade model, the profile of the trainers who directly train the teachers, the location of the training, the size of the sessions, and the time division between lectures, practice, and other activities. Finally, the Perceptions section includes indicators capturing program implementers’ own perceptions of which elements were responsible for any positive impacts and which were popular or unpopular among teachers. We piloted the draft instrument by using it to collect data on a sample of evaluated programs, and validated its ability to accurately characterize the details of PD programs by sharing our results with a series of expert researchers and practitioners in teacher PD. We updated the indicators in light of this feedback, resulting in a final version of the instrument, which includes 70 indicators plus three pieces of metadata. Further information on the instrument can be found in the supplementary online appendices: Appendix A1 provides a more detailed description of instrument development; appendix A2 presents the final instrument (ITTSI); and appendix A3 presents the Brief In-Service Teacher Training Instrument (BITTSI), a supplementary instrument we developed containing a subset of the 13 most critical questions from the ITTSI based on our reading of the literature.

The ITTSI does not collect extensive data about the broader educational context. Context includes teacher policies (e.g., pre-service training and the structure of the teacher career), other education policies, and the current state of education (e.g., learning and absenteeism rates). Context matters for the impact of teacher PD programs. As a simple example, in a setting where student absenteeism is extremely high, teacher PD programs may have a limited impact on student learning due to few hours of contact between teachers and students. That said, certain principles of teacher PD may translate across cultures, even if the applications vary. Professionals need practice to master skills across contexts, so giving teachers the opportunity to practice lessons during training may be valuable across contexts, even if how they do that may vary. Other survey instruments have been developed and tested broadly to gather a wide range of data on the education system, notably the World Bank's Systems Approach for Better Education Results (SABER) ( Rogers and Demas 2013 ). For a rich view of teacher PD in context, the ITTSI could be complemented with the SABER instrument or other data about the education system.

Applying the ITTSI to Evaluated PD Programs

We searched the existing literature on in-service teacher PD in low- and middle-income countries to identify a sample of PD programs that had been evaluated for their impact on student learning. Our inclusion criteria for the search were impact evaluations of primary and secondary education interventions in low- and middle-income countries that (a) focused primarily on in-service teacher PD or included this as a major component of a broader program, and (b) reported impacts of the program on student test scores in math, language, or science. We included both published and unpublished papers and did not restrict by year of authorship.

In order to identify papers fulfilling the above criteria, we searched a range of databases in 2016 . 2 The search yielded 6,049 results and automatically refined the results by removing exact duplicates from the original results, which reduced the number of results to 4,294. To this we added 20 impact evaluations which mention teacher PD from a recent review ( Evans and Popova 2016a ). We examined the 4,314 results from both sources to exclude articles that—from their title and abstract—were clearly not impact evaluations of teacher training programs. This review process excluded 4,272 results and left 42 full articles to be assessed for eligibility. After going through the full texts, another 18 papers were excluded as the full text revealed that they did not meet the inclusion criteria. This yielded 23 papers, which evaluated 26 different PD programs. In February 2018, we updated this original sample with full articles published between 2016 and 2018 which fit the inclusion criteria. This resulted in seven new papers and teacher PD programs for a total of 30 papers evaluating 33 programs. The search process is detailed in  fig. 2 . The 30 papers are listed in supplementary online appendix A4 .

Search Process and Results for Evaluated Professional Development Programs

Source : Constructed by the authors based on the search described in the text.

Note : The 30 papers documenting the evaluation of the final 33 programs are listed in supplementary online appendix A4 .

Data collection and coding for the sample of 33 evaluated programs comprised two phases. The first of these phases consisted of carefully reviewing the impact evaluation studies and coding the information they provided. The draft version of the instrument for which we collected data included 51 indicators in total, and on average, information on 26 (51 percent) of these indicators was reported in the impact evaluations. Crucially, the amount of program information reported across the impact evaluations varies noticeably by topic ( table 1 ). Sixty-four percent of details concerning the organization of teacher training programs—such as whether the program was designed by a government or by a non-governmental organization (NGO)—can be extracted from the evaluations. In contrast, on average, only 47 percent of information concerning program content and 42 percent of information concerning program delivery is reported.

Data Available on Evaluated Programs from Studies vs. Interviews

Source : Constructed by the authors based on the application of the In-Service Teacher Training Survey Instrument items ( supplementary online appendix A2 ) to the 33 professional development programs identified ( supplementary online appendix A4 ).

Note : Percentage data collected refers to the percentage of indicators for which data were collected across the 33 programs in our evaluated sample. This is calculated by the number of programs for which each indicator has data, summed for every indicator in a given section (or total) and divided by the number of indicators in that section (or total), and finally divided by the 33 programs.

The second phase of data collection sought to fill this gap in reported data by interviewing individuals involved in the actual implementation of each program. To do this, we emailed the authors of each of the impact evaluations in our sample, asking them to connect us with the program implementers. After three attempts to contact the implementers, we received responses from authors for 25 of the 33 programs. We contacted all of the individuals to whom the authors referred us—who in many cases directed us to more relevant counterparts—and were eventually able to hold interviews with program implementers for 18 of the 33 programs. 3 The interviews loosely followed the survey instrument, but included open-ended questions and space for program implementers to provide any additional program information that they perceived as important.

The ITTSI data were gathered retrospectively for this study, which means that in most cases, the evaluation results (and so whether or not the program was effective) were likely to have been known to the interviewee. We propose three reasons that this should not pose a substantive problem for the quality of the data. First, most of the indicators have no normative response. Whether a program is government- or researcher-designed or implemented, whether it has a subject focus or a general pedagogy focus, or whether or not it has a distance learning element have no obvious “right” answers. Second, the survey was administered to program implementers, who usually were not part of the team of researchers who evaluated the program, so they had little stake in confirming research results. Third, the survey had low stakes: interviewees knew that we were independent researchers doing a synthesis review. In some cases, the PD program being discussed no longer existed in the same form. For future PD studies, these data could be collected at the design stage of programs.

For the 18 programs for which we conducted interviews, we were able to collect information for an average of 50 out of the 51 (98 percent) indicators of interest. Consequently, conducting interviews decreased the differences in data availability across categories. The pooled average of indicators for which we had information after conducting interviews (for interviewed and not interviewed programs combined) increased to 79 percent for Organization indicators, 68 percent of Content indicators, and 72 percent of Delivery indicators ( table 1 ).

For our sample of evaluated in-service teacher PD programs, we analyze which characteristics of teacher training programs are associated with the largest improvements in student learning, as measured by test score gains. We conduct both quantitative and qualitative analyses. The analytical strategy for the quantitative analysis essentially consists of comparing means of student learning gains for programs with and without key characteristics, using a bivariate linear regression to derive the magnitude and statistical significance of differences in means. We do not carry out multivariate regression analysis because of the small sample; thus, these results are only suggestive, as multiple characteristics of programs may be correlated. Because we are testing each coefficient separately, we are not able to test the relative value of coefficients, so differences in point estimates are only suggestive.

In preparation for this analysis, we standardize the impact estimates for each of the programs. We convert the program characteristic variables to indicator variables wherever possible to facilitate comparability of coefficients. Although our sample of impact evaluations has a common outcome—impact on student test scores—these are reported on different scales across studies, based on different sample sizes. 4 We standardize these effects and the associated standard errors in order to be able to compare them directly. Supplementary online appendix A5 provides mathematical details of the standardization.

Turning to the independent variables, as originally coded, the 51 indicators for which we collected information capturing various design and implementation characteristics of the PD programs took a number of forms. These consisted of indicator variables (e.g., the intervention provides textbooks alongside training = 0 or 1), categorical variables (e.g., the primary focus of the training was subject content [= 1], pedagogy [= 2], new technology [= 3]), continuous variables (e.g., the proportion of training hours spent practicing with students), and string variables capturing open-ended perceptions (e.g., which program elements do you think were most effective?). To maximize the comparability of output from our regression analysis we convert all categorical and continuous variables into indicator variables. 5

We then conduct our bivariate regressions on this set of complete indicator variables with continuous impact estimates on test scores as the outcome variable for each regression. Because of the limitations associated with running a series of bivariate regressions on a relatively small sample of evaluations, we propose the following robustness check. First, we estimate robust Eicker-Huber-White (EHW) standard errors as our default standard errors (reported in  tables 2 – 4 ) and assess significance according to p -values associated with these. Second, we estimate bootstrapped standard errors and the associated p -values. Third, we run Fisher randomization tests to calculate exact p -values, a common approach in the context of small samples. 6 We report significance under each of these methods separately and report results as robust if they are significant under at least two of the three methods, and if the significant effect is driven by at least two observations—i.e., the results are not explained by a single PD program.

Organization – Bivariate Regressions with Robustness Checks

Source : Constructed by the authors based on data extracted from 33 professional development programs ( supplementary online appendix A4 ) using the In-Service Teacher Training Survey Instrument, and analyzed by regression, as described in the text.

Note : ∗ p  < 0.10, ∗∗ p  < 0.05, ∗∗∗ p  < 0.01 correspond to the significance of p- val ues of robust standard Noteerrors. § corresponds to significance at the 10 percent level or higher for bootstrapped standard errors. † corresponds to significance at the 10 percent level or higher for the Fisher Randomization tests. Numbers specified in parentheses in variable labels are the reported medians for dummy variables in which the variable equals 1 if greater than the median. Total programs refers to the number of programs that report whether or not they have the characteristic. The robust column includes an X if the finding is statistically significant across at least two methods and if the finding is driven by two or more evaluations (i.e., not a single evaluation).

Content – Bivariate Regressions with Robustness Checks

Note : ∗ p  < 0.10, ∗∗ p  < 0.05, ∗∗∗ p  < 0.01 correspond to the significance of p -values of robust standard errors. § corresponds to significance at the 10 percent level or higher for bootstrapped standard errors. † corresponds to significance at the 10 percent level or higher for the Fisher Randomization tests. Total programs refers to the number of programs that report whether or not they have the characteristic. The robust column includes an X if the finding is statistically significant across at least two methods and if the finding is driven by two or more evaluations (i.e., not a single evaluation).

Delivery – Bivariate Regressions with Robustness Checks

Note : ∗ p  < 0.10, ∗∗ p  < 0.05, ∗∗∗ p  < 0.01 correspond to the significance of p -values of robust standard errors. § corresponds to significance at the 10 percent level or higher for bootstrapped standard errors. † corresponds to significance at the 10 percent level or higher for the Fisher Randomization tests. Numbers specified in parentheses in variable labels are the reported medians for dummy variables in which the variable equals 1 if greater than the median. Total programs refers to the number of programs that report whether or not they have the characteristic. The robust column includes an X if the finding is statistically significant across at least two methods and if the finding is driven by two or more evaluations (i.e., not a single evaluation).

We supplement this regression analysis with a qualitative analysis of what works, relying on the self-reported perceptions of program implementers along three dimensions: (a) Which program elements they identified as most responsible for any positive impacts on student learning; (b) which elements, if any, teachers particularly liked; and (c) which elements, if any, teachers particularly disliked.

Applying the ITTSI to At-Scale PD Programs

The sampling process for at-scale programs is detailed in  fig. 3 . To obtain a sample of at-scale, government-funded PD programs across the world, we first identified four to five countries in each region where the World Bank has operations. 7 We worked with regional education managers at the World Bank in each region to select countries in which government counterparts and World Bank country teams had an interest in learning more about in-service teacher PD programs. We made clear that the exercise was appropriate for countries with any level of teacher PD, not specific to countries with recent reforms or innovations. The final set of countries sampled included Burkina Faso, Cambodia, El Salvador, The Gambia, Guinea, India (Bihar state), Jordan, Kazakhstan, the Kyrgyz Republic, Mauritania, Mexico (Guanajato, Oaxaca, and Puebla, and a national PD program for middle school teachers), Moldova, Niger, and the Russian Federation.

Sampling Process for At-Scale Professional Development Programs

Source : Constructed by the authors to reflect the process to identify at-scale professional development programs, as described in the text.

We then obtained permission from the Ministry of Education (MoE) or other relevant government counterparts in each country and worked with them to complete a roster, or listing, of all teacher PD programs conducted between 2012 and 2016. 8 The roster, available in supplementary online appendix A6 , was created along with the ITTSI instrument and collects the following information about each of the teacher PD programs that received government funding: program name; program coordinator's name and contact information; the number of teachers trained; and the types of teachers targeted (e.g., pre-primary, primary, or secondary school teachers). In some countries, such as Mexico and India, where policymaking about teacher PD happens at the state level, we worked with individual states.

After receiving completed roster information about teacher PD programs in a country/state, we used the roster to select a sample of teacher PD programs to interview. In each country/state, we chose the sample by selecting the 10 largest teacher PD programs in terms of teacher coverage, defined as the number of teachers reached by the program during its most recent year of implementation. Of the 10 sampled programs for each country/state, the full ITTSI was administered to the two largest programs targeting primary school teachers and the largest program that targeted secondary school teachers. The brief version of the instrument, the BITTSI, was administered in the remaining seven programs in the country/state. In total, 48 at-scale programs completed the ITTSI and 91 at-scale programs completed the BITTSI across 14 countries.

We applied the ITTSI survey through a combination of phone interviews with and online surveys of PD program coordinators. In a few instances (in The Gambia, El Salvador, and Mexico), depending on the preferences of the program coordinator and their primary language, program coordinators were given the option of completing the ITTSI questionnaire online. For the majority of programs, however, we held phone interviews with program coordinators, in which we asked them the questions included in the ITTSI survey items directly and filled out the instrument ourselves with their responses.

The ITTSI survey applied to the sample of at-scale programs consists of 70 indicators. We were able to collect information for an average of 66 of the 70 (94 percent) indicators of interest for the 48 at-scale teacher PD programs to which the full ITTSI survey was applied, and for 26.5 of the 27 (97 percent) indicators—derived from 13 questions—for the 91 programs to which the BITTSI was applied.

For the sample of at-scale PD programs, we compare the average of observed characteristics of at-scale teacher PD programs with the average for evaluated PD programs that resulted in the largest improvements in student learning (“top performers”), as measured by student test score gains. To determine the characteristics of “top performers,” we ranked all evaluated programs, using their standardized impact on student test scores. We then selected the top half of programs (16 programs, all of which displayed positive impacts), and calculated the average value of program indicators for those “top performers.” We compare them to the means of at-scale PD programs in order to better understand the gap between at-scale PD practices and the best practices of top-performing PD programs.

This section characterizes the specific characteristics of teacher PD programs that successfully improve student learning in low- and middle-income countries and how common these characteristics are across at-scale, government-funded programs. First, we present the results of our quantitative and qualitative analyses examining which PD characteristics are associated with large gains in student learning for the sample of evaluated programs. Second, we present descriptive statistics from the sample of at-scale PD programs and from the top-performing PD programs in the evaluated sample to shed light on how they differ in terms of those PD characteristics found to be associated with positive impacts on student learning.

Which PD Characteristics are Most Associated with Student Learning Among Evaluated Programs?

We discuss, for each of our categories—Organization, Content, and Delivery—those characteristics we observe to be most associated with student learning gains.  Tables 2 – 4 present the results of our bivariate regressions for each of these categories in turn. In each case, we report the results with the three different methods of calculating significance as well as an indicator of robustness.

Among Organization ( table 2 ), two characteristics are robustly associated with significant gains in student learning. These include linking career opportunities (improved status, promotion, or salary) to PD programs and targeting training programs based on teachers’ years of experience. First, in teacher PD programs where participation has no implications for promotion, salary, or status increases, student learning is 0.12 standard deviations lower (significant at 95 percent). In other words, programs that do link participation to career incentives have higher effectiveness. 9 Second, targeting participant teachers by their years of experience is associated with 0.10 standard deviations higher student learning (significant at 90 percent). This is driven by two programs: the Balsakhi program in rural India, which trains women from the local community who have completed secondary school to provide remedial education to students falling behind ( Banerjee et al. 2007 ); and the Science teacher training program in Argentina, which trains teachers in different structured curricula and coaching techniques and finds that coaching is only effective for less experienced teachers ( Albornoz et al. 2018 ). Indeed, these are the only two programs out of the 33 that explicitly targeted teachers based on their experience, both of which resulted in student learning gains. In addition, the provision of complementary materials such as storybooks and other reading materials (e.g., flashcards or word banks) have large coefficients associated with improving student learning (0.11 and 0.13 standard deviations), although these are not statistically significant.

Among the Content variables ( table 3 ), programs with a specific subject focus result in higher learning gains than more general programs. Specifically, programs with no subject focus show 0.24 standard deviations lower impact on student learning (significant at 99 percent). A deeper look reveals that within focus areas, programs that are not focused on a given academic subject—such as those focused on counseling—are associated with 0.2 lower standard deviations in student learning (significant at 99 percent). Lastly, when a teacher PD program involves teaching practice through lesson enactment, it is associated with a 0.10 standard deviation increase in student learning (significant at 90 percent).

Turning to Delivery characteristics ( table 4 ), three characteristics of teacher PD programs are robust. First, teacher PD programs that provide consecutive days of face-to-face teacher training are associated with a 0.14 standard deviation increase in student learning (significant at 99 percent). Second, holding face-to-face training at a central location—such as a hotel or government administrative building (as opposed to a university or training center, which was the omitted category)—is associated with a 0.13 lower standard deviation in student learning (significant at 90 percent). Third, teacher PD training programs that are conducted remotely using distance learning are associated with a 0.10 standard deviation decrease in student learning (significant at 90 percent). In alignment with recent literature highlighting the overly theoretical nature of many training programs as an explanation for their limited effects on student learning—as well as the above finding that training programs that involve teaching practice are associated with 0.16 larger gains in student learning—the proportion of training time spent practicing with other teachers is highly correlated with learning impacts (although not consistently statistically significant). Also, the inclusion of follow-up visits to review material taught in the initial training—as opposed to visits for monitoring purposes alone or no follow-up visits—is associated with a 0.14 standard deviation higher program impact on student learning (not significant, but one of the largest coefficients). These findings support the literature that subject-focused teacher PD programs with consecutive days of face-to-face training that include time for teachers to practice with one another, are associated with improved student learning outcomes.

We supplement the quantitative results with an analysis of self-reported perceptions by the implementers of the evaluated programs. These concern the characteristics of their programs which they believe are most responsible for any positive effects on student learning, as well as those elements which were popular and unpopular among the beneficiary teachers. We elicited these perceptions using open-ended questions and then tallied the number of program implementers that mentioned a given program element in their response, albeit not necessarily using the exact same language as other respondents. These responses come from 18 interviewees, so they should be taken as suggestive. That said, the results broadly align with the quantitative results: Five of 18 interviewees—tied for the most common response—mentioned that mentoring follow-up visits were a crucial component in making their training work. Similarly, five of the 18 interviewees discuss the importance of having complementary materials, such as structured lessons or scripted materials that provide useful references in the classroom and help to guide teachers during the training sessions. The next most commonly reported elements were engaging teachers for their opinions and ideas—either through discussion or text messages—and designing the program in response to local context, building on what teachers already do and linking to everyday experiences: both were mentioned by four of 18 interviewees.

We also asked the program implementers about the program characteristics that they believed teachers liked and disliked the most about their training programs and, interestingly, we only found two common responses for what teachers particularly liked and one common response for what they disliked. 10 Seven of the 18 interviewees reported that the part of their program that teachers most enjoyed was that it was fun and engaging (or some variation of that). In other words, teachers appreciated that certain programs were interactive and involved participation and discussion rather than passive learning. In addition to having “fun” teacher PD programs, five of the 18 interviewees suggested that teachers especially liked the program materials provided to them. Similarly, in terms of unpopular program elements, four of the 18 program implementers we interviewed reported that teachers disliked the amount of time taken by participating in the training programs, which they perceived as excessive.

What Do We Learn from At-Scale PD Programs?

Government-funded, at-scale teacher PD programs have a number of characteristics in common ( supplementary online appendix tables A7.1–A7.3 ). The vast majority are designed by government (80 percent) and implemented by government (90 percent). Almost all provide materials to accompany the PD (96 percent), and most include at least some lesson enactment (73 percent) and development of materials (73 percent). Most have a subject focus (92 percent) and include an initial period of face-to-face training for several days (85 percent). Most do not formally target teachers by subject (only 19 percent do), grade (31 percent), or years of experience (13 percent), and few have negative consequences if teachers are poorly evaluated (17 percent). These at-scale programs differ sharply from programs that are evaluated in general, as well as from top-performing evaluated programs specifically. We provide a full list of average characteristics of at-scale programs and all evaluated programs (not just top-performers) in supplementary online appendix tables A7.1–A7.3 .

Our principal focus in this section is how at-scale programs compare to evaluated programs that deliver relatively high gains in student learning. We assess the top half of programs (N = 16) from the sample of evaluated programs by selecting those characteristics that produced the largest standard deviation increases in student assessment scores. In  table 5 , we compare the means of at-scale programs and top-performing, evaluated programs. We focus specifically on the characteristics shown to have a statistically significant relationship with student learning outcomes and those with large coefficients, identified for interest (as identified in  tables 2 – 4 ).

Comparison of Means of At-Scale Programs and Top-Performing, Evaluated Programs

Source : Constructed by authors, comparing summary statistics for the top performing professional development (PD) programs among rigorously evaluated PD programs to at-scale PD programs.

Note : For the full list of statistics, see supplementary online appendix Tables A7.1–A7.3 .

Regarding Organization ( table 5 ), two key characteristics—whether or not the training is linked to career opportunities and whether or not the program targets teachers based on their years of experience—are robustly associated with improved student learning gains. There are notable and substantive differences between top-performing PD programs and the sample of at-scale PD programs when it comes to providing incentives; 88 percent of top-performing PD programs link training to status or to new career opportunities such as promotion or salary, as compared to only 55 percent of at-scale programs. Our results suggest that without incentives, training may not have a meaningful impact. Furthermore, top-performing programs and at-scale PD programs are similar in the degree to which they target teachers based on their years of experience. For instance, 13.3 percent of top-performers and 12.5 percent of at-scale programs target teachers based on their experience. Other notable organizational characteristics include the provision of complementary materials such as storybooks and reading materials. Top-performing PD programs and at-scale PD programs are similar in the amount of materials they provide, but our results suggest that the kinds of complementary materials may differ somewhat. For instance, only 12.5 percent and 21 percent of at-scale programs provide storybooks and reading materials, respectively—materials correlated with student learning gains—as compared to 36 percent and 43 percent of evaluated programs.

Turning next to Content ( table 5 ), top-performing PD programs and at-scale PD programs perform similarly. In both instances, the majority of programs include subject content and subject-specific pedagogy as either a primary or secondary focus. Few programs—none of the top performers—and only eight percent of at-scale programs lack a subject focus. Moreover, no top-performing programs and few at-scale programs (fewer than six percent) focus on general training in areas such as counseling or providing training on how to use a specific tool—types of training that are statistically linked to lower gains in student learning.

Finally, Delivery characteristics ( table 5 ) include whether or not there are consecutive days of face-to-face training, training location, the amount of time teachers spend practicing with one another, and follow-up visits. Specifically, 100 percent of top-performing programs include consecutive days of face-to-face training as compared to 85 percent of evaluated programs. Our research further suggests that the location of PD training programs may influence program effectiveness, and training held at central locations such as hotels or conference rooms (as opposed to universities or training centers) may be less effective. Currently 73 percent of at-scale, government-funded programs are held at central locations as compared to only 38 percent of evaluated programs.

Follow-up visits with teachers and the amount of time teachers spend practicing with other teachers during the training program are shown to be positively correlated with large coefficients (albeit not statistically significant) on student learning. In both instances, top-performing PD programs include more follow-up visits (five versus two median visits among programs with visits) and spend more time allowing teachers to practice with other teachers (40 percent versus 16 percent of training time) than do at-scale programs. 11 Results of our analysis suggest that training may be more effective if there are follow-up visits. This is an imperative finding when comparing top-performing PD programs, in which 85 percent include follow-up visits, with government-funded, at-scale PD programs, in which only half of programs include follow-up visits. Also, in top-performing PD programs, teachers spend more time practicing what they have learned with other teachers (40 percent of overall training time) relative to at-scale programs (only 16 percent). An existing body of research suggests that when teachers have opportunities to practice the new skills they acquire in PD programs, they are more likely to adopt these new skills in their classrooms ( Wiley and Yoon 1995 ; Wenglinsky 2000 ; Angrist and Lavy 2001 ; Borko 2004 ).

Governments spend enormous amounts of time and money on in-service professional development. Many countries have multiple in-service PD programs running simultaneously, as evidenced by our sample of at-scale PD programs. Many go unevaluated and may be ineffective. This paper makes three major contributions: first, it reveals broad weaknesses in reporting on teacher PD interventions. There are almost as many program types as there are programs, with variations in subject and pedagogical focus, hours spent, capacity of the trainers, and a host of other variables. Yet reporting on these often seeks to reduce them to a small handful of variables, and each scholar decides independently which variables are most relevant to report. We propose a standard set of indicators—the ITTSI—that would encourage consistency and thoroughness in reporting. Academic journals may continue to pressure authors to report limited information about the interventions, wishing instead to reserve space for statistical analysis. However, authors could easily include the full set of indicators in an appendix attached to the paper or online.

Second, this paper demonstrates that some characteristics of teacher PD programs—notably, linking participation to incentives such as promotion or salary implications, having a specific subject focus, incorporating lesson enactment in the training, and including initial face-to-face training—are positively associated with student test score gains. Furthermore, qualitative evidence suggests that follow-up visits to reinforce skills learned in training are important to effective training. Further documentation of detailed program characteristics, coupled with rigorous evaluation, will continue to inform effective evaluations.

The impacts of these characteristics are not small: having a specific subject focus and incorporating lesson enactment are associated with 0.24 and 0.10 more standard deviations in learning, respectively, for example. Comparing these effect sizes to those from a sample of 747 education-related randomized controlled trials in the United States puts them both above the 50th percentile in terms of effectiveness ( Kraft 2020 ). Comparing to a set of 130 randomized controlled trials in low- and middle-income countries likewise put them at or above the 50 th percentile of 0.10 standard deviations ( Evans and Yuan 2020 ). In high-income countries, Kennedy (2019) proposes that the impact of teacher PD programs be benchmarked against a much less costly “community of practice” model in which teachers help each other, like Papay et al. (2020) . While we are not aware of a rigorously evaluated, costed model of that class of program in a low- or middle-income country, an alternative would be to compare teacher PD results to a pure monitoring model, such as an increase in inspections. Along these lines, Muralidharan et al. (2017) show—using data from India—that increased frequency of monitoring would be a much more cost-effective way to reduce effective class sizes (through reduced teacher absenteeism) than hiring more teachers. These are useful avenues to pursue for future research as countries consider the cost-effectiveness of alternative investments in teachers.

Third, by comparing the means of at-scale PD programs with top-performing evaluated programs, our findings highlight gaps between what evidence suggests are effective characteristics of teacher PD programs and the contextual realities of most teacher PD programs in their design, content, and delivery. In particular, our findings taken together suggest that at-scale programs often lack key characteristics of top-performing training programs. At-scale programs are much less likely to be linked to career incentives, to provide storybooks or other reading materials, to have a subject content focus, to include time for practicing with other teachers, or to include follow-up visits.

The approach taken by this paper centers on using the ITTSI to collect and compare data on rigorously evaluated and at-scale, government-funded teacher PD programs. This approach has limitations. First, the evidence of what works within rigorously evaluated programs is limited by those programs that have been evaluated. There may be innovative PD programs that are not among the “top performers” simply because they have yet to be evaluated. While this evidence base can push policymakers away from approaches that do not work, it should not deter policymakers from innovating and evaluating those innovations.

A second, related limitation concerns the relatively small sample of evaluated teacher PD programs in low- and middle-income countries, on which our findings about effective PD characteristics are based. Some of the larger coefficients in the regressions are driven by a small number of teacher training programs. These instances have been noted in the text. As more evaluations of PD programs are conducted, the ITTSI can be applied to these and our analyses re-run to shed further light on the specific characteristics associated with PD programs that improve student learning. The ITTSI data were already updated once in this way in 2018, increasing the number of evaluated programs in our sample from 26 to 33.

Third, a conceptual concern with evaluating teacher professional programs is the risk that impacts may be explained by observer effects (also referred to as Hawthorne effects). These effects have been documented in education ( Muralidharan et al. 2017 ) and health in low- and middle-income countries ( Leonard 2008 ; Leonard and Masatu 2010 ). The impact of any education intervention may partly be due to observer effects, since the introduction of an intervention suggests that someone is paying attention to the teacher's efforts. Both randomized controlled trials and more traditional monitoring and evaluation may enhance these effects, as teachers may further respond favorably to the observation associated with measurement. Randomized controlled trials and quasi-experimental studies with a credible comparison group overcome part of this concern, as the observer effect associated with measurement will exist in both the treatment and comparison groups, and measured program impacts should be net of those effects.

That leaves the impact of the intervention itself. In this review, all of the studies we include evaluate interventions and, as such, all may be subject to an observer effect. Our analysis implicitly assumes the magnitude of this observer effect to be constant across different types of PD. By comparing PD characteristics across programs, we observe whether those characteristics are associated with a larger total effect on learning. Part of that total effect may stem from increased teacher skills, and part may be explained by certain PD characteristics inducing greater observer effects (since any observer effects that are uncorrelated with PD characteristics would be absorbed in our regression constant terms). In the short run, the impact for students is observationally equivalent. Even with longer run studies (of which there are very few in education and development), observer effects may fade, but teacher skills may also depreciate ( Cilliers et al. 2020 ). As a result, we consider the total association of PD characteristics with student learning, including through increased teacher human capital and observer effects.

Fourth, there are challenges in comparing evaluated PD programs with at-scale PD programs. As the data demonstrate, at-scale PD programs tend to be larger programs designed by governments, often at the national level, and aimed at providing broad training to teachers. In light of these differences, we highlight the fact that top-performing programs—regardless of their core objectives—share certain common sets of characteristics that most at-scale programs do not share. Awareness of these characteristics may be useful in the conceptualization and implementation of future teacher PD programs in low- and middle-income countries, including large-scale programs funded by governments.

One key reason that at-scale programs may differ from successful, evaluated programs is that the latter group of evaluations may not be designed in a way that is conducive to scaling. Evaluated programs tend to be much smaller than at-scale programs: in our data, evaluated programs reached an average of 96 schools versus at-scale programs that reached more than 6,000 schools on average ( supplementary online appendix table A7.1 ). These smaller programs often have higher per-pupil costs ( Evans and Popova 2016b ), so scaling them nationwide requires cutting elements. Smaller programs are easier to staff and easier to monitor. Evaluated programs were three times as likely to be designed by researchers and less than one-third as likely to be implemented by government ( supplementary online appendix table A7.1 ). One solution, obviously, is more large-scale evaluations, like Loyalka et al. (2019) . However, even smaller evaluations can do more to mimic scalable policies. Gove et al. (2017) , reflecting on programs evaluated both at pilot and at scale in Kenya and Liberia, suggest the value of testing as many elements as possible in the pilot, using government systems in the pilot as much as possible, and to make sure that pilot costs are within what a government budget can handle. Duflo et al. (2020) combine these two approaches in a recent nationwide, five-arm randomized controlled trial in Ghana, to test the scalability of four different models to reach remedial learners, which had previously been tested in small pilot randomized controlled trials elsewhere. When implemented within existing government systems, they find all four interventions to be effective, pointing to the program's inception within the government as key, as opposed to an initial non-government organization initiative subsequently and imperfectly implemented by the government.

Improving in-service teacher professional development may be a clear win for governments. They are already spending resources on these programs, and there is broad support for these programs among teachers and teachers’ unions. Interventions such as the above provide learning opportunities for country governments and stakeholders seeking to design effective teacher PD programs. While no single characteristic of top-performing PD programs may transform an ineffective PD program into an effective one, this paper highlights trends in top-performing programs, such as including incentives, a specific subject focus, and lesson enactment. These are characteristics that, if included and implemented successfully, have the potential to improve the quality of teacher PD programs, and ultimately, the quality of instruction and student learning.

The authors are grateful for comments from Denise Bello, Luis Benveniste, Barbara Bruns, Martin Carnoy, Joost de Laat, Margaret Dubeck, Deon Filmer, Susanna Loeb, Prashant Loyalka, Ezequiel Molina, Andrew Ragatz, and Halsey Rogers. They are also grateful to Fei Yuan for excellent research assistance, to Veronica Michel Gutierrez, Olga A. Rines, Lea Jeanne Marie Lungmann, Fata No, and Elissar Tatum Harati for their support with data collection, and to numerous teacher training implementers for providing information on programs. This paper subsumes an earlier paper, “Training Teachers on the Job: What Works and How to Measure It” (World Bank Policy Research Working Paper Number 7834).

This work was supported by the Bill & Melinda Gates Foundation, the World Bank's Systems Approach for Better Education (SABER) Trust Fund, which was supported by the United Kingdom's Department for International Development (DFID) and Australia's Department of Foreign Affairs and Trade (DFAT), and the Strategic Impact Evaluation Fund at the World Bank.

Both samples focus on teacher training programs at the primary and secondary school level. Pre-primary schools are excluded.

The databases we searched were the Education Resources Information Center (ERIC); Academic Search Complete; Business Source Complete; Econlit with Full Text; Education Full Text (H. W. Wilson); Education Index Retrospective: 1929–1983; Education Source; Educational Administration Abstracts; Social Science Full Text (H. W. Wilson); Teacher Reference Center; and EconLit. We looked for articles containing the terms (“teacher training” OR “teacher education” OR “professional development”) AND (``learning'' OR ``scores'' OR ``attainment'') AND (“impact evaluation” OR ``effects'') AND (“developing country 1” OR “developing country 2” OR “developing country N”), where “developing country” was replaced by country names.

In six cases, program implementers failed to schedule an interview after three attempts at contact, and in the case of one older program, the implementer had passed away. Interviews were held over the phone or in-person, and lasted between 45 and 90 minutes for each program.

A limitation is that some of the impact estimates from school-randomized control trials in our evaluated sample are over-estimates because the authors fail to account for the clustering of children within teachers or schools ( Hedges 2009 ).

For categorical variables, this is straightforward. For example, we convert the original categorical variable for the location of the initial teacher PD—which includes response options of schools, a central location, a training center, or online—into four dummy variables. In order to convert the continuous variables to a comparable scale, we create a dummy for each continuous variable which, for a given program, takes a value of 1 if the continuous variable is greater than the median value of this variable across all programs, and a value of 0 if it is less than or equal to the value of this variable across all programs. We apply this method to the conversion of all continuous variables except three—proportion of teachers that dropped out of the program, number of follow-up visits, and weeks of distance learning—which we convert directly to dummy variables that take a value of 1 if the original variable was greater than 0, and a value of 0 otherwise.

We estimate bootstrapped standard errors by resampling our data with replacement 1,000 times. We run Fisher randomization tests by treating each indicator PD characteristic as a treatment and calculating a randomization distribution of mean differences (the test statistic) across treatment assignments. Specifically, for 1,000 permutations, we randomly reassign values of 0 or 1 to the independent variables in our regressions, while maintaining the overall proportion of 0s and 1s observed in the empirical sample for a given variable. We then calculate Fisher exact p -values by finding the proportion of the randomization distribution that is larger than our observed test statistic ( Fisher 1925 , 1935 ; Imbens and Rubin 2015 ).

These regions include: Africa, Eastern and Central Europe, Latin American and the Caribbean, the Middle East and North Africa, and East and South Asia.

This includes programs ongoing in 2016 and programs that were implemented anytime in the range of 2012 to 2016. Hence, the programs could have been designed prior to 2012. We still include them if they were implemented any time between 2012 and 2016. We were not successful in obtaining roster information in all countries. For instance, in Morocco and the Arab Republic of Egypt, the Ministries of Education were in the process of making changes to the structure and delivery of teacher training programs and indicated that it was not a good time for data collection. In Tanzania there was a change in leadership among government counterparts during efforts to complete the roster and data collection process, and we were not able to properly sample and apply the ITTSI in all teacher-training programs in the country. In India, we had initially identified two states, Bihar and Karnataka, to work with at the subnational level, but ultimately only collected data in one state, Bihar, since the principal government counterpart in Karnataka was not available to complete the roster.

In some cases, we test a negative (e.g., no implications for status in table 2 or no subject focus in table 3 ) because we are testing an exhaustive series of indicators derived from the same question (e.g., subject focus is math, subject focus is literacy, or no subject focus).

Because it is difficult to imagine an effective teacher professional development program that teachers actively dislike (they have to learn for it to work, after all), their preferences are relevant.

When we include programs with no follow-up visits, the median number of follow-up visits to teachers in top programs becomes 3.5 as compared to 0 for at-scale programs.

Albornoz   F. , Anauati   M. V. , Furman   M. , Luzuriaga   M. , Podestá   M. E. , Taylor   I. . 2018. “ Training to Teach Science: Experimental Evidence from Argentina .” Policy Research Working Paper 8594 , World Bank, Washington, DC .

Angrist   J. D. , Lavy   V. . 2001 . “ Does Teacher Training Affect Pupil Learning? Evidence from Matched Comparisons in Jerusalem Public Schools .” Journal of Labor Economics   19 ( 2 ): 343 – 69 .

Google Scholar

Araujo   M. C. , Carneiro   P. , Cruz-Aguayo   Y. , Schady   N. . 2016 . “ Teacher Quality and Learning Outcomes in Kindergarten .” Quarterly Journal of Economics   131 ( 3 ): 1415 – 53 .

Baker   S. , Smith   S. . 1999 . “ Starting off on the Right Foot: The Influence of Four Principles of Professional Development in Improving Literacy Instruction in Two Kindergarten Programs .” Learning Disabilities Research & Practice   14 ( 4 ): 239 – 53 .

Banerjee   A. , Cole   S. , Duflo   E. , Linden   L. L. . 2007 . “ Remedying Education: Evidence from Two Randomized Experiments in India .” Quarterly Journal of Economics   122 ( 3 ): 1235 – 64 .

Berlinski   S. , Busso   M. . 2017 . “ Challenges in Educational Reform: An Experiment on Active Learning in Mathematics .” Economics Letters   156 : 172 – 5 .

Bold   T. , Filmer   D. , Martin   G. , Molina   E. , Rockmore   C. , Stacy   B. , Svensson   J. , Wane   W. . 2017 . “ What Do Teachers Know and Do? Does It Matter? Evidence from Primary Schools in Africa .” Policy Research Working Paper 7956 , World Bank, Washington, DC .

Borenstein   M. , Hedges   L. V. , Higgins   J. P. T. , Rothstein   H. R. . 2009 . Introduction to Meta-Analysis .   Chichester, United Kingdom : John Wiley & Sons .

Google Preview

Borko   H.   2004 . “ Professional Development and Teacher Learning: Mapping the Terrain .” Educational Researcher   33 ( 8 ): 3 – 15 .

Bourgeois   E. , Nizet   J. . 1997 . Aprendizaje y formación de personas adultas .   Paris, France : Presses Universite de France .

Cardemil   C.   2001 . “ Procesos y condiciones en el aprendizaje de adultos .” Jornada Nacional de Supervisores. Supervisión para aprendizajes de calidad y oportunidades para todos. Educación Rural .   Santiago : Ministerio de Educación . https://repositorio.uahurtado.cl/handle/11242/8517 .

Chetty   R. , Friedman   J. N. , Rockoff   J. E. . 2014 . “ Measuring the Impacts of Teachers II: Teacher Value-Added and Student Outcomes in Adulthood .” American Economic Review   104 ( 9 ): 2633 – 79 .

Cilliers   J. , Fleisch   B. , Kotze   J. , Mohohlwane   M. , Taylor.   S.   2020 . “ The Challenge of Sustaining Effective Teaching: Spillovers, Fade-out, and the Cost-effectiveness of Teacher Development Programs .” Unpublished Working Paper . https://www.dropbox.com/s/6xmv7283oxoysj2/The%20Challenge%20of%20Sustaining%20Effective%20Teaching%20with%20appendix.pdf?dl=0 .

Darling-Hammond   L. , Wei   R. C. , Andree   A. , Richardson   N. , Orphanos   S. . 2009 . Professional Learning in the Learning Profession .   Washington, DC : National Staff Development Council . https://edpolicy.stanford.edu/sites/default/files/publications/professional-learning-learning-profession-status-report-teacher-development-us-and-abroad.pdf .

Desimone   L. M.   2009 . “ Improving Impact Studies of Teachers’ Professional Development: Toward Better Conceptualizations and Measures .” Educational Researcher   38 ( 3 ): 181 – 99 .

Duflo   A. , Kiessel   J. , Lucas   A. . 2020 . “ External Validity: Four Models of Improving Student Achievement .” Working Paper No. w27298 , National Bureau of Economic Research , Cambridge, MA .

Evans   D. K. , Popova   A. . 2016a . “ What Really Works to Improve Learning in Developing Countries? An Analysis of Divergent Findings in Systematic Reviews .” World Bank Research Observer   31 ( 3 ): 242 – 70 .

Evans   D. K. , Popova   A. . 2016b . “ Cost-Effectiveness Analysis in Development: Accounting for Local Costs and Noisy Impacts .” World Development   77 : 262 – 76 .

Evans   D. K. , Yuan   F. . 2020 . “ How Big are Effect Sizes in International Education Studies? ” Working Paper 545 , Center for Global Development , Washington, DC .

Fisher   R. A.   1925 . Statistical Methods for Research Workers , first edition. Edinburgh : Oliver and Boyd Ltd .

Fisher   R. A.   1935 . The Design of Experiments ,  sixth edition. Edinburgh : Oliver and Boyd, Ltd , 1951 .

Fryer, Jr   R. G.   2017 . “ The Production of Human Capital in Developed Countries: Evidence from 196 Randomized Field Experiments .” Handbook of Economic Field Experiments   2 : 95 – 322 .

Gersten   R. , Taylor   M. J. , Keys   T. D. , Rolfhus   E. , Newman-Gonchar   R. . 2014 . Summary of Research on the Effectiveness of Math Professional Development Approaches .   Washington, DC : Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, U.S. Department of Education; and Regional Educational Laboratory Southeast at Florida State University . http://files.eric.ed.gov/fulltext/ED544681.pdf .

Gove   A. , Poole   M. K. , Piper   B. . 2017 . “ Designing for Scale: Reflections on Rolling Out Reading Improvement in Kenya and Liberia .” New Directions for Child and Adolescent Development   2017 ( 155 ): 77 – 95 .

Hedges   L.V . 2009 . “ Effect Sizes in Nested Designs .” In The Handbook of Research Synthesis and Meta-analysis , edited by Cooper   H. , Hedges   L. V. , Valentine,   J. C.   337 – 56 . New York, NY : Russell Sage Foundation .

Huberman   M.   1989 . “ The Professional Life Cycle of Teachers .” Teachers College Record   91 ( 1 ): 31 – 57 .

Imbens   G. W. , Rubin   D. B. . 2015 . Causal Inference in Statistics, Social, and Biomedical Sciences .   Cambridge, UK: Cambridge University Press .

Kennedy   M. M . 2019 . “ How We Learn About Teacher Learning .” Review of Research in Education   43 ( 1 ): 138 – 62 .

Kerwin   J. T. , Thornton   R. L. . 2021 . “ Making the Grade: The Sensitivity of Education Program Effectiveness to Input Choices and Outcome Measures .” Review of Economics and Statistics   103 ( 2 ): 251 – 64 .

Knowles   M. S. , Holton   E. F. , Swanson   R. A. . 2005 . The Adult Learner , sixth edition.   Burlington, MA : Elsevier .

Kraft   M. A.   2020 . “ Interpreting Effect Sizes of Education Interventions .” Educational Researcher   49 ( 4 ): 241 – 53 .

Kraft   M. A. , Papay   J. P. . 2014 . “ Can Professional Environments in Schools Promote Teacher Development? Explaining Heterogeneity in Returns to Teaching Experience .” Educational Evaluation and Policy Analysis   36 ( 4 ): 476 – 500 .

La Paro   K. M. , Pianta   R. C. . 2003 . CLASS: Classroom Assessment Scoring System . Charlottesville, VA : University of Virginia .

Leonard   K. L.   2008 . “ Is Patient Satisfaction Sensitive to Changes in the Quality of Care? An Exploitation of the Hawthorne Effect .” Journal of Health Economics   27 ( 2 ): 444 – 59 .

Leonard   K. L. , Masatu   M. C. . 2010 . “ Using the Hawthorne Effect to Examine the Gap Between a Doctor's Best Possible Practice and Actual Performance .” Journal of Development Economics   93 ( 2 ): 226 – 34 .

Loyalka   P. , Popova   A. , Li   G. , Shi   Z. . 2019 . “ Does Teacher Training Actually Work? Evidence from a Large-Scale Randomized Evaluation of a National Teacher Training Program .” American Economic Journal: Applied Economics   11 ( 3 ): 128 – 54 .

McEwan   P.   2015 . “ Improving Learning in Primary Schools of Developing Countries: A Meta-analysis of Randomized Experiments .” Review of Educational Research   85 ( 3 ): 353 – 94 .

Molina   E. , Fatima   S. F. , Ho   A. , Hurtado   C. M. , Wilichowski   T. , Pushparatnam   A. . 2018 . “ Measuring Teaching Practices at Scale: Results from the Development and Validation of the Teach Classroom Observation Tool .” Policy Research Working Paper 8653 , World Bank , Washington, DC .

Muralidharan   K. , Das   J. , Holla   A. , Mohpal   A. . 2017 . “ The Fiscal Cost of Weak Governance: Evidence from Teacher Absence in India .” Journal of Public Economics   145 : 116 – 35 .

Papay   J. P. , Taylor   E. S. , Tyler   J. H. , Laski   M. E. . 2020 . “ Learning Job Skills from Colleagues at Work: Evidence from a Field Experiment Using Teacher Performance Data .” American Economic Journal: Economic Policy   12 ( 1 ): 359 – 88 .

Piper   B. , Korda   M. . 2011 . EGRA Plus: Liberia (Program evaluation report) . Durham, NC : RTI International .

Rogers   H. , Demas   A. . 2013 . The What, Why, and How of the Systems Approach for Better Education Results (SABER) . Washington, DC: World Bank . http://wbgfiles.worldbank.org/documents/hdn/ed/saber/supporting_doc/Background/SABER_Overview_Paper.pdf .

Shulman   L. S.   1986 . “ Those Who Understand: Knowledge Growth in Teaching .” Educational Researcher   15 ( 2 ): 4 – 14 .

TNTP . 2015 . The Mirage: Confronting the Hard Truth about Our Quest for Teacher Development . The New Teacher Project. http://files.eric.ed.gov/fulltext/ED558206.pdf .

Villegas-Reimers   E.   2003 . Teacher Professional Development: An International Review of the Literature . Paris : UNESCO International Institute for Educational Planning . http://www.iiep.unesco.org/en/publication/teacher-professional-development-international-review-literature .

Walter   C. , Briggs   J. . 2012 . What Professional Development Makes the Most Difference to Teachers .   Oxford : University of Oxford Department of Education . https://www.oupjapan.co.jp/sites/default/files/contents/events/od2018/media/od18_Walter_reference.pdf .

Wenglinsky   H.   2000 . “ How Teaching Matters: Bringing the Classroom Back into Discussions of Teacher Quality .” Policy Information Center Report , Educational Testing Service (ETS) .

Wiley   D. , Yoon   B. . 1995 . “ Teacher Reports of Opportunity to Learn: Analyses of the 1993 California Learning Assessment System .” Educational Evaluation and Policy Analysis   17 ( 3 ): 355 – 70 .

Wood   F. H. , McQuarrie   F. Jr.   1999 . “On the Job Learning. New Approaches will Shape Professional Learning in the 21st Century .” Journal of Staff Development   20 : 10 – 13 .

Yoon   K. S. , Duncan   T. , Lee   S. W. Y. , Scarloss   B. , Shapley   K. . 2007 . Reviewing the Evidence on how Teacher Professional Development Affects Student Achievement (Issues & Answers Report No. 033) . Washington, DC : Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, U.S. Department of Education; and Regional Educational Laboratory Southeast at Florida State University . http://files.eric.ed.gov/fulltext/ED498548.pdf .

Supplementary data

Email alerts

Citing articles via.

• Recommend to your Library

Affiliations

• Online ISSN 1564-6971
• Print ISSN 0257-3032
• Copyright © 2024 World Bank
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

• DOI: 10.3126/JTD.V1I0.13084
• Corpus ID: 36147787

Assessing the Effectiveness of Teacher Training Programs to Improve the Quality of School Education in Nepal

• Bhawani Shankar Subedi
• Published 31 July 2015

10 Citations

Education policies and practices for ensuring quality education in nepal, evidence-based training approach for higher education faculty: brief model of inclusion and training of the disabled, redesigning continuing professional development training (cpdt) in higher education, exploring the tensions between policy and practice: a case study of nepali secondary school teachers and leaders interpretations of the national curriculum framework's learner-centred reform policy, understanding the nepali classroom practices: a constructivist perspective, barriers in implementing tqm in secondary schools: a comparison between privately-owned and government-owned schools of pakistan, effectiveness of teachers’ professional development in dalits’ opportunities for capability development in nepal, evaluation of training session applying gagne’s events of instructions.

• Highly Influenced

Disrupting power through reflexivity: new possibilities for researchers and teachers working with Dalits in Nepal

Can teachers learn online – evidence from armenia during the covid-19 pandemic, 5 references, related papers.

Showing 1 through 3 of 0 Related Papers

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

•  We're Hiring!
•  Help Center

Teacher Training

• Most Cited Papers
• Most Downloaded Papers
• Newest Papers
• Last »
• Teacher Education Follow Following
• Teachers' professional development Follow Following
• Learning and Teaching Follow Following
• Education Follow Following
• Educational Research Follow Following
• Teaching and Learning Follow Following
• Educational Technology Follow Following
• Critical Pedagogy Follow Following
• Education Policy Follow Following
• Higher Education Follow Following

Enter the email address you signed up with and we'll email you a reset link.

• Academia.edu Journals
•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

Website unavailable

Due to unforeseen circumstances the University website is unavailable. We are working to reinstate the website as quickly as possible.

If your query is urgent please call the University switchboard on +44 (0)23 8059 5000.

Apologies for any inconvenience.