The effects of head teacher training on pupil achievement

An NBER Working Paper examines the impact of implementing management training for head teachers on pupil achievement. The management training focused on lesson planning, data-driven teaching and teacher observation and coaching (approximately 300 hours over two years). Using a school-level randomised experiment, 58 schools in Houston, Texas, were randomised to receive either the training intervention or to serve as a business-as-usual control group.

The study found that offering management training to head teachers led to increased test scores across low-stakes tests in a range of subjects in year one (effect size = +0.19). For high-stakes test scores in maths and reading, the effect size was lower (+0.10). However, the training intervention had no impact on high-stakes tests in year two.

The training was most beneficial for head teachers who were less experienced, had better maths skills, had more internal locus of control, had higher levels of “grit” and remained in the school for both years of the study.

The intervention showed most impact on teachers in the schools who were more experienced and more educated. The intervention showed most impact for pupils who were new to the school, white or Hispanic and economically well-off.

Source: Management and student achievement: Evidence from a randomized field experiment (May 2017), NBER Working Paper No. 23437, National Bureau of Economic Research

 

Unhealthy school lunches nibble away academic performance

A National Bureau of Economic Research (NBER) Working Paper looks at the effect of offering healthier school lunches on end-of-year test scores for pupils in California.

Michael L Anderson and colleagues analysed data collected over a five-year period (academic years 2008/2009 to 2012/2013) from around 9,700 schools that reported test scores on California’s Standardized Testing and Reporting exam, a state-wide test given to all pupils in grades 2 to 11 (Years 3 to 13). In order to determine the link between food quality and pupil achievement, they also collected data from the California Department of Education on school districts’ meal vendors for the same time span. Over that five-year period, about 12% of California schools contracted an outside company to provide lunch for at least one school year. The nutritional value of the school lunch menus was analysed by nutritionists at the Nutrition Policy Institute using the Healthy Eating Index.

The results of the study found that in years when schools offered a healthy lunch menu, pupil test scores were on average higher (effect size +0.03 to +0.04). In addition, test scores for pupils who qualified for reduced-price or free school lunches, (and therefore more likely to eat the healthy lunches), increased by about 40% in comparison to pupils who didn’t receive free school lunches.  The positive effect of healthy lunches on academic achievement persisted for the duration of a long-term contract.

Source: School lunch quality and academic performance (March 2017), NBER Working Paper No. 23218, National Bureau of Economic Research (NBER)

Do computers have a negative effect on children’s social development?

A working paper from the National Bureau of Economic Research reports the findings from a large-scale randomised controlled trial that explores whether owning a home computer has a negative effect on children’s social development.

The study included 1,123 students in grades 6-10 (Years 7-11) in 15 different schools across California. Students were eligible to take part in the trial only if they did not already have a computer at home. Half were then randomly selected to receive free computers, while the other half served as the control group. Surveys were conducted with the students and schools at the start of the school year to collect data on child and household characteristics and school participation. Follow-up surveys were then administered at the end of the school year, and the data compared to establish any causal evidence.

As predicted, Robert W Fairlie and Ariel Kalil found that having computers at home did increase the amount of time that children spent on social networking sites and email as well as for games and other entertainment. However, rather than being socially isolating, children in the treatment group communicated with 1.57 more friends per week than children in the control group, and spent 0.72 more hours with their friends in person. They also found no evidence that the children who received a computer were less likely to participate in sports teams or after-school clubs, or spend any less time in these activities.

Source: The effects of computers on children’s social development and school participation: evidence from a randomized control experiment (December 2016), NBER Working Paper No. 22907, The National Bureau of Economic Research

Specialist teachers in primary schools

Would primary schools be more successful if, like secondary schools, they used specialist teachers for particular classes?

A working paper from the National Bureau of Economic Research reports on an experiment that tried to establish just that. In Houston, Texas, 50 elementary schools were randomised into treatment or control groups. Treatment schools altered their timetables to have teachers specialise in subjects such as maths, science, social studies, and reading based on each teacher’s strengths (assessed by the school principal). A class might be taught by one teacher for maths and science, and another for reading and social studies. Other classes had one teacher for maths, another for reading, and a third for science and social studies.

The results were negative. In the first year, schools with specialist teachers saw an effect size of -0.07 on maths and reading achievement. Over the first two years, the effect size was -0.05 for maths and -0.04 for reading, with the maths result statistically significant. For children in special education, the results were even worse, with an impact of -0.15 for reading and -0.20 for maths.

A teacher survey measured views on lesson planning, teacher relationships with students, enjoyment of teaching, and teaching strategies. Teachers in treatment schools were significantly less likely to report providing tailored instruction for their students. All other survey outcomes on teaching strategy were statistically identical between treatment and control.

Source: The ‘Pupil’ Factory: Specialization and the Production of Human Capital in Schools, National Bureau of Economic Research (2016)

Is professional development better than being dismissed?

The last issue of Best Evidence in Brief reported on a study in which low-performing teachers were dismissed. A new working paper from the National Bureau of Economic Research reports on an experiment where low-performing teachers were provided with coaching from higher-performing peers.

The experiment took place in Tennessee in 14 elementary and middle schools. Tennessee teachers are observed in the classroom many times each year, and scored on 19 specific skills (eg, questioning, lesson structure and pacing, and managing student behaviour). Schools were randomly assigned to a treatment condition or business-as-usual control group. In the treatment schools, low-performing “target” teachers were matched with high-performing teachers, based on the outcomes of their classroom observations. The high-performing teachers were chosen based on their high scores in skills for which the low-performing teachers had received a low score. The pairs were encouraged to work together on these skills, as well as more generally on observing each other’s teaching, discussing strategies for improvement, and following up on each other’s commitments throughout the year.

After a year, students in treatment schools (whether taught by target or non-target teachers) showed a small improvement (effect size +0.06) on maths and English tests, when compared with students in control schools. Gains by students taught by target teachers were higher (+0.12). These improvements persisted and grew. In the following year, the effect for target teachers was a marginally significant +0.25.

Source: Learning Job Skills from Colleagues at Work: Evidence from a Field Experiment Using Teacher Performance Data (2016), The National Bureau of Economic Research.

Can teacher turnover improve student achievement?

Removing low-performing teachers can improve outcomes for students, according to a working paper from the National Bureau of Economic Research.

The study looked at the effect of IMPACT, a performance assessment and incentive system introduced in schools in Washington, DC. IMPACT evaluates all teachers every year based on multiple measures of effectiveness (eg, clearly described standards, several teacher observations by different observers, and student outcomes). Teachers receive one of four ratings – highly effective, effective, minimally effective, or ineffective. Those rated ineffective (or minimally effective two years running) were dismissed. Total teacher turnover rates in the schools were higher (18% per year) than in similar non-participating schools (8-17%). Turnover rates among low-performing teachers (minimally effective or ineffective) were 46%, and for high-performing teachers (effective or highly effective) 13%.

The study found that there was a small increase in student achievement in mathematics (effect size of +0.079) that was statistically significant, and a smaller increase for reading (+0.046) which was not statistically significant. Turnover of high-performing teachers had a negative but not statistically significant effect on student achievement in mathematics (-0.055). Turnover of low-performing teachers improved student performance in mathematics (+0.21) and reading (+0.14).

The authors conclude that under a robust system of performance assessment, turnover of low-performing teachers can generate meaningful gains in student outcomes, particularly for the most disadvantaged students.

Source: Teacher Turnover, Teacher Quality, and Student Achievement in DCPS (2016), The National Bureau of Economic Research.