Bias in teacher and student expectations in relation to high-stakes exam results

• In Event: Inequality and attainment in schools

Proposal

Bias of teachers and students’ expectations in relation to final exam results

The phenomenon of “expectations” and “self-fulfilling prophecy” has occupied a special place in sociological and psychological studies of education over the last 50 years with its influence on the academic results and students’ educational trajectories. (Brophy, 1983; Rosenthal and Jacobson, 1968; Jussim & Harber, 2005). In connection with the facts mentioned above, a separate interest raises the question of factors related to biases in the expectations of students and teachers regarding educational outcomes. A large body of research is devoted to biases in terms of gender (Ready and Wright, 2011; Riegle-Crumb and Humphries, 2012), racial and ethnic characteristics (Glock & Klapproth, 2017; Kaiser et al., 2016; McKown & Weinstein, 2008; Rubie -Davies et al., 2006; Tenenbaum and Ruck, 2007). At the same time, the topic of communication between the socio-economic characteristics of children and teachers and their expectations is the least studied in this area (Ready & Wright, 2011). Moreover, previous studies were mainly devoted to one-sided study of biases in expectations - on teacher predictions of students’ results or on students' self-esteem. Their mutual biases are rarely studied in comparison although it is known from the literature that positive students’ self-esteem is connected with higher expectations of teachers (Jussim, 1989).

Given everything mentioned above, the following research questions were posed:

1) How is the accuracy of teacher and students’ predictions of the educational results of children with the economic and cultural capital of their families?
2) How does the accuracy of teacher and student predictions vary depending on the cultural capital and professional experience of the teacher?
3) How do the biases in teachers and students’ expectations relate to each other?

In order to get answers to the research questions, the data from the longitudinal study “Trajectories in education and careers” (trec.hse.ru) was used. This study was conducted in 210 educational institutions in 42 regions of the Russian Federation. Its feature is the presence of a representative sample of students who participated in the international TIMSS study in 2011 and in PISA survey in 2012. These same respondents were subsequently questioned in order to track their educational and career trajectories. The information obtained from the surveys allowed us to get a fairly complete picture of the socio-economic context in which the students are located as well as the professional characteristics and cultural capital of the teachers working with them. The analysis uses the results of the USE (Unified State Exam) students in the Russian language and mathematics. Biases in expectations are defined as the difference between the real exam score obtained by the student at the exam and the predicted by the teacher / student one month before the exam. As a method of analysis, a linear regression (OLS) was used.

The results of the study showed that the most significant factors causing bias in teachers and students’ expectations are the previous student’s academic results. The teachers of both subjects overestimate pupils with better results and underestimate poorly performing ones. At the same time, low-achievers and the most successful pupils tend to underestimate their results in comparison with those who had average academic performance.

As for the other characteristics of children, the cultural capital of the family in the form of mother’s education is a factor in biasing teacher expectations in the Russian language. Moreover, students whose mothers graduated from high school tend to overestimate their scores in the Russian language as well. The family wealth is associated with the students’ overestimation of their results. However, this factor does not cause significant biases in teacher expectations.

Among the characteristics of teachers, indicators of experience (years of teaching and qualification category) and cultural capital (number of books at teacher’s home) are important. Teachers of the Russian language with the highest category overestimate students in comparison with less experienced colleagues. The increase in experience for 1 year, in contrast, is associated with a slight increase in the underestimation of students’ results. In mathematics, teachers with the maximum number of books at home tend to underestimate students but the experience and category did not show a significant relationship with an assessment bias.

As a result of the analysis, it was possible to identify several patterns relating to coincidences or discrepancies in teachers and students’ expectations: mutual bias in the expectations of teachers and students, polar divergence of students' and teachers' expectations and mismatch in teachers and students’ expectations. The last pattern should be noted separately - underestimation and overestimation arose in the area of ​​disproportions of the level of cultural capital (experience, assessments) of teachers and students. Teachers with a high level of cultural capital overestimate students with a low level, and teachers with a low cultural level, on the contrary, overestimate high – cultured ones. For example, math teachers with a high cultural level underestimate children who have low marks or few books at home. Similarly, teachers who don’t have a category overestimate children whose mothers have a higher education.

In general, the obtained results do not contradict the data of previous studies. However, the underestimation of their capabilities by students with a higher level of cultural capital as well as the discrepancy between the expectations of teachers and students in the comparison are unexpected and require further study. On the other hand, the coincidences in biases concerning expectations of teachers and students, despite the fact that they look predictable, pose a question for further study of cause-effect relationships. If teacher expectations can raise students’ self-esteem, this may be an important mechanism for improving the results - “self-fulfilling prophecy.”

Authors

• Anastasia Getman, National Research University Higher School of Economics, Moscow, Russia

• Andrei Zakharov, National Research University Higher School of Economics