Mindsets: The Influence of Intelligence Theories on Educational Outcomes

The intervention was a structured workshop of 5 weeks to elementary education teachers with the objective of promoting pedagogical practices with "growth mindset". Our hypothesis is that, in promoting changes in the beliefs about the intelligence, the implementation of mindset workshops helps teachers in their pedagogical practices to improve student learning. The workshops included discussions about brain functioning and their relation to the concept of mindset, the re-signification of error as fundamental to the learning process, and the elaboration of strategies to deal with error and frustration in the learning process. Finally, the workshops dealt with the role of effort and the concept of “stereotype threat” (Steele and Aronson 1995) in vulnerable contexts.

2019-03-13

2019-04-11

We will evaluate the impact of the intervention on student performance.

We will use the database of the national student performance standardized exam (SAEB). The national exam will be applied in all Brazilian public 5th grade schools by the end of October 2019.

We are also interested in analyzing (i) teachers’ mindsets, (ii) teaching practices, and (iii) students’ mindsets.

By the end of 2019 (around November), we will apply questionnaires to teachers and students from all of the candidate schools to measure differences in behavior between the control and treatment groups, thus assessing the extent of change in attitudes toward intelligence generated by the participation of the teachers in the five week workshop. These questionnaires measure, through scales, the extent of the participant’s growth mindset according to a multidimensional definition of this concept. We have prepared our questionnaires by adapting sections of different published questionnaires. It was necessary, however, to revise many of the items to our specific context, adjusting their language and references to the realities of Brazilian students and teachers. In addition, it was necessary to translate the questions to Portuguese, a process that involved the translation of the same items simultaneously by two different translators, who then worked together to write a consensual version. Since individual’s mindsets are also manifested in their attitudes, the questionnaires ask additional questions about teachers’ and students’ opinions and behaviors. Finally, each questionnaire included questions aimed at measuring how sincere participants were, in order to deal with potential bias of "social desirability" of the questions. A first version of the questionnaires was applied to the pilot group, in 2018. The analysis of the response data served to generate final versions of the teachers and students’ questionnaires, which will be used in the 2019 data collection.

Furthermore, what happens inside the classroom is central for the students’ learning process. In order to measure changes in teachers’ pedagogical practices as a result of the transformations in their conceptions of intelligence resulting from the mindset workshops, we will use the TEACH+ observation instrument in the classroom. The TEACH+ instrument is designed and adapted to the Brazilian context by specialists of the TEACH and Stallings classroom observation methods (Bruns, Costa & Cunha, 2018; Molina et al. al., 2018). TEACH+ seeks to measure the time teachers spend in the learning processes, the extent to which students remain attentive to their tasks, and the quality of teaching techniques that help to develop students’ social and emotional skills.

We ran a two-level clustered randomized experiment with treatment at the second (school) level. We randomly selected 50% of 178 candidate schools from the municipality to the treatment group and the other 50% to the control group. The randomization was done by pairing, using the predicted value of the students’ results inferred from the characteristics of the schools in previous years. It is estimated that schools have the harmonic average of 60 students in the 5th grade, divided on average in two classes, according to data from 2018.

The principals of schools with one or two 5th grade classes from the municipal public network of Rio de Janeiro first decided whether they wanted to enroll their schools in the project. Out of the 178 schools that chose to participate, we randomly selected one group - 50% of the sample - to receive the treatment and 50% to serve as the control. The randomization was done by pairing, using the predicted value of the students' results inferred from the characteristics of the schools in previous years (specifically, we used data from the student performance standardized exam (SAEB) in 2015 and 2017 and the School Census in 2016 and 2018 to predict what scores the schools will likely achieve in the student performance standardized exam in 2019).
Treated schools have taken part in the intervention consisting of a mindset workshop (of 5 meetings in March and April 2019) aiming to change the perception that teachers have about their own capacity and intelligence, and how such perceptions can be passed on to their students. Within the schools, we will evaluate teachers’ mindsets and teaching practices, their students’ mindsets and, finally, student outcomes.

Randomization done in office by a computer.

The randomization is clustered at the school level: Students within classes, within schools.

Yes

178 schools with one or two 5th grade classes from the municipal public network of Rio de Janeiro were evenly and randomly distributed into the treatment and control groups.

A total sample of 323 teachers and approximately 10,852 students (taking the standardized test in 2019).

89 schools were selected to the treatment group and another 89 to the control group.

Our MDES is 0.13 for Mathematics, and 0.12 for Language results in student standardized test.