Using nationality as a proxy for skin colour, it was found that students from Portuguese-speaking African countries (PSAC) ‒ i.e. Black students ‒ were three times more likely in lower secondary education (15% vs 4.2%) and twice more likely in high school (76% vs 35.6%) than native Portuguese students to pursue professional studies, a pathway that hinders access to higher education and places them at a disadvantage. It is important to notice that these discrepancies cannot be attributed solely to socioeconomic status (SES), as they persisted specifically among students from low SES families. So, what other factors can explain these differences? 

Explanations for inequalities in educational outcomes often focus on students’ performance, family resources, and structural factors, such as place of residence. It is possible that teachers, and more specifically teachers’ racial biases in expectations regarding their students’ school pathways, also play a significant role in explaining these differences. Through their recommendations, teachers can either facilitate or hinder access to higher education. Not only do their perceptions influence students’ school pathways, but also their recommendations are relevant for both the students and their parents. 

The purpose of this study among a sample of lower secondary teachers in Portugal was to answer the question: “Do teachers make different recommendations depending on whether students are Black or White?”. It also intended to provide an insight on whether teachers’ recommendations and expectations regarding Black and White students’ school pathways may be influenced by their potentially biased perceptions of achievement and behaviour. 

Finally, the study also sought to assess whether, from a qualitative perspective, narratives about the future school pathways of White and Black students are significantly different (which would be a further indication of possible racial biases).
To this end, all teachers in the sample were presented with a survey containing a vignette that showed the below-average performance of a student and his (somewhat) positive behaviour. They were then asked about their expectations, perceptions, and recommendations regarding this student. A third of the teachers were given a vignette that included the picture of a White student; another third, a vignette with the picture of a Black student, and another third a vignette with no picture at all. 

All teachers were presented with a vignette showing the performance (i.e. grades) and behaviour of a student. The content of all the vignettes was the same, but three picture scenarios were used: one containing the picture of a Black student, another of a White student, and another with no picture at all. Teachers were asked whether they considered the student’s behaviour to be positive or negative. Among the teachers who considered the behaviour depicted to be particularly positive, there was a significant difference in terms of their recommendations for the university-oriented pathway, with the White student being recommended more often than the Black one. This seems to indicate that identical information about a student’s behaviour was considered differently by the teachers depending on whether the student was White or Black. In other words, their recommendation was racially biased, as positive behaviour was enough to give a positive recommendation for the White student, but not for the Black one.

As explained before, all teachers were presented with a vignette showing the performance (i.e. grades) and behaviour of a student, and all the vignettes were the same, except for the picture of the student it contained: that of a White student, that of a Black student, or no picture at all. The performance presented included some negative grades and could therefore be considered below average. Teachers were asked what was most important when making recommendations for future school pathways: the student’s performance or the student’s behaviour. In general, teachers indicated that, in the case of a Black student, performance was more important than behaviour. Thus, it appears that teachers considered the same information about performance differently: a negative performance was considered more informative about the future of the Black student than that of the White one.

Though not supported by statistical significance, a consistent trend across several dimensions shows that teachers at the beginning of their careers reveal a subtle bias in favouring White students in terms of expectations, behavioural assessment and recommendations for the university-oriented pathway. In the sample of this study, less experienced teachers (especially those at the very beginning of their careers) had more positive expectations for the White student, considered the student’s behaviour as more positive and tended to recommend a university-oriented pathway for them. Clearly, these biases were not observed among the more experienced teachers, which may be an indication that years of experience enable the correction of potential racial biases.

Two open-ended questions of the survey allowed participants to refer to the students freely, in a manner unconstrained by survey measures. Specifically, teachers were asked to write down what they remembered about the student. A quantitative analysis of this unconstrained qualitative data showed that teachers used different words depending on whether they were referring to a White student or a Black one. The mere fact that there were differences in the teachers’ choice of words depending on the skin colour of the student is a potential indication of racial inequalities. One specific manifestation of these different word choices translates into a more frequent use of the word interested for the White student (28 times vs 14 times for the Black student) and a more frequent use of the word difficulties for the Black student (9 times vs 2 times for the White student).

As explained above, teachers were presented with the same vignette of a student, the only difference being the picture it contained. At the end of the questionnaire, participants were asked about the information they had seen and, in particular, whether they remembered seeing the picture of a student. The results of this “information check” indicate that a very significant percentage of the participants did not accurately recall whether there was a picture of a student and, when they did, they did not correctly “guess” the skin colour of the student in the vignette. More importantly, these errors were more frequent when the student was Black. This inattention could be an indication of an “indifferent gaze” towards these students.

The aim of this research study was to answer the question: “Do teachers make different recommendations depending on whether students are Black or White?”. It also intended to provide an insight on whether teachers’ recommendations and expectations regarding Black and White students’ school pathways may be influenced by their potentially biased perceptions of achievement and behaviour. 

The results do not allow the researchers to conclude with certainty that there are significant biases among teachers in Portugal in this respect. At the same time, it is not possible to conclude that these biases do not exist, as there were concrete cases that proved otherwise. Indeed, the results show that information about performance and behaviour is considered differently when making recommendations, so that they favour White students or are unfavourable to Black ones. Furthermore, there was a subtle (though statistically non-significant) indication that these biases are more common among inexperienced teachers. Finally, the simple fact that a student was White or Black was enough to prompt different word choices within open-ended questions. 

This study employed a controlled experimental approach in which the authors presented vignettes varying in terms of the picture of the student it contained (White, Black or no picture at all) in order to observe the impact of skin colour on a number of aspects. In theory, this meant that any differences found between the responses of the teachers when referring to the White or the Black student (regarding perceptions or recommendations) could not be attributed to differences in the student’s performance, age, and behaviour, and would have to be attributed to his skin colour, which proved the existence of a racial bias. However, for this study to be reliable, it was important to make sure that participants paid attention to the manipulated information (i.e. the picture of the student). Unfortunately, the data obtained shows that many teachers did not pay any attention to the picture (or its absence), with a significant percentage of the sample being unable to properly remember the information they had been given. The analyses could not have been carried out without those participants who “failed” this “attention test”, as this would have produced an unviable sample. Moreover, as mentioned before, this inattention was marked by skin colour. And, although this inattention may have compromised some of the potential conclusions of this study, the mere fact that this “limitation” was racially structured may be a revealing indication that racial biases are indeed present in this educational context. Thus, even though it limits the reliability of the conclusions reached, it is a nuanced limitation that provides further potential cues about the existence of bias and the different ways in which teachers think about White and Black students.

Additionally, the vignettes used in this survey referred exclusively to male students. While this was intentional in order to simplify the study design, it does prevent the research from properly taking into account potential gender differences within these phenomena. Future research ought to include vignettes with students of different genders. 

Finally, the direct and explicit surveying of the teachers may also limit the reliability of the study conclusions due to social desirability issues. More implicit indirect surveys and studies should address this in the future.

ABRANTES, P., e ROLDÃO, C. (2019): «The (mis)education of African descendants in Portugal: Towards vocational traps?», Portuguese Journal of Social Sciences, 18(1), 27-55. 

ANYON, Y., et al. (2014): «The persistent effect of race and the promise of alternatives to suspension in school discipline outcomes», Children and Youth Services Review, 44, 379-386.  

MORRIS, E. W., e B. L. PERRY (2016): «The Punishment Gap: School Suspension and Racial Disparities in Achievement», Social Problems, 63, 68-86. 

PASCHALL, K. W., E. T. GERSHOFF, e M. KUHFELD (2018): «A two decade examination of historical race/ethnicity disparities in academic achievement by poverty status», Journal of Youth and Adolescence, 47, 1164-1177.