ENGLISH LANGUAGE TEACHER EDUCATION IN BRAZIL: USING CENSUS DATA TO EXPLORE THE CONFIGURATION OF UNDERGRADUATE COURSES

: Research on English language teacher education (ELTE) in Brazil has prioritized qualitative investigations of specific institutional contexts and has overlooked nationwide explorations of ELTE. The present study addresses this research gap through an investigation of cross-institutional ELTE across the country with a focus on its contextual configuration and social inclusion. To this end, the study has undertaken a secondary analysis of the Brazilian federal government’s 2019 higher education census data. The findings show that future English language teachers’ academic socialization takes place primarily in online undergraduate courses with an embedded teaching qualification, offered by private higher education institutions. These online ELTE courses are mainly attended by White students even though the Brazilian population consists predominantly of Blacks. We argue that considerable work needs to be undertaken to change the current configuration of ELTE in Brazil and promote social inclusion. We urge teacher educators to engage in the analysis of census data to contribute to nationwide teacher education policymaking.


INTRODUCTION
The use of English for communication purposes around the world has prompted local and national governments to introduce policies aimed at its acquisition in educational systems from early childhood to higher education.This situation is not different in Latin American countries (Cronquist, Fiszbein, 2017).In Brazil, the country where the present study took place, the most recent change in its main educational law has made the teaching of English compulsory in lower and upper-secondary education1 (Brasil, 2017), thus replacing the former directive which allowed schools to decide which additional language should be part of the curriculum.This change in the Brazilian legislation can be related to both international and national matters.Globally, there is an interconnectedness of language policies (Rizvi;Lingard, 2010) which are pushing for the teaching of English as part of national development plans (Coleman, 2011).Nationally, the compulsory inclusion of English language education in the curriculum may be seen as an attempt to address Brazil's low proficiency in this additional language (e.g., Education First, 2023).Such change may additionally be an attempt to improve the teaching/learning of English in Brazilian regular schoolsespecially government-sponsored onessince they are perceived to achieve very little in terms of language proficiency (e.g., Dias;Assis-Peterson, 2006;Gasparini, 2005;Gimenez, 2014;Jandre;Viana, 2017).
The disparities between those who can only attend regular schools and those who can afford additional private language courses are one of the main concerns of educational researchers and practitioners who have been investigating classrooms and teacher education programs in order to support government policies (British Council, 2015;2019).Several factors are believed to contribute to these disparities, from material onese.g., overcrowded classrooms, lack of teaching resources, the reduced number of teaching hours dedicated to English -to human resources as in the case of teachers' skills (Gasparini, 2005;Paiva, 2003).
Considering that teachers play an important role in creating successful learning environments and in supporting learners' acquisition of the necessary language skills, teacher education is one of the main aspects to be addressed in language education policies, which need to draw not only on qualitative studies but also on quantitative ones.
With the ultimate goal of contributing to the formulation of relevant data-driven education policies in Brazil, this paper examines English language teacher education (ELTE) based on the 2019 Brazilian Census of Higher Education.To this end, the paper is divided into six sections.Following this introduction, we review the research literature on ELTE in Brazil and we present our research questions.In the next section, we discuss the ways in which census data may be used in research, and we explain our choice for secondary data analysis.We then present and discuss the main results of our empirical analysis, respectively.In the final section, we offer some concluding remarks and urge teacher educators to engage in census data analyses to inform policy development.

LITERATURE REVIEW
The research literature on language teacher education in Brazil and, more specifically, ELTE in the countryis plentiful, especially in the field of Applied Linguistics.Researchers have carried out many qualitative studies to advance our understanding of English language teachers' beliefs and practices (e.g., Andrade et al., 2019;Celani, 2016;Gil, 2005;Mateus et al., 2013;Telles, 2009).Although the contexts where English language teachers are being educatedi.e., undergraduate Letras courseshave also been the object of investigation, there has been a tendency to adopt qualitative approaches, focusing on specific programs (e.g., Calvo: Devico: Novelli, 2022;Ramos et al. 2020).These studies primarily draw on educational contexts where the researchers work and/or where they can have easy access to participants.In this sense, the ELTE research conducted in Brazil follows the international paradigm shift noted by Vélez-Rendón (2002): a focus on "more qualitative-oriented research approaches, that is, inquiry conducted in naturally occurring settings that allows for deeper understanding of phenomena and participants' lived experiences" (p.457).Therefore, there are fewer nationwide studies which portray the profile of future English language teachers and the type of courses they attend, which is the focus of the present paper.
The reduced number of studies which examine Brazilian ELTE contexts from a nationwide perspective is surprising.This lack of attention is not due to unavailability of empirical data: every year the federal government conducts its Census of Higher Education, and the data are made available to the wider community.One notable exception is Silva's (2020) study, which compares the results reported in the 2001 and 2018 census technical summaries to provide a diachronic perspective on ELTE undergraduate courses in Brazil.Silva (2020) indicates that ELTE course provision has expanded in the 17-year time frame considered in his research.This expansion has been primarily noted in relation to on-site courses amongst public higher education institutions (HEIs), especially those funded by the federal and state governments, and online courses amongst private HEIs.The increase in the number of courses and openings has been accompanied by an increase in the number of applicants as well.
However, Silva (2020) points out that, while there is still some competition to secure a place in ELTE undergraduate courses, these coursesespecially online oneshave higher rates of untaken places.Another relevant investigation was carried out by Lebler (2022), who analyzed census data from 2010-2018 regarding Letras courses which included a qualification in the Portuguese language.2She also noted the predominance of online education offered by private HEIs as compared to on-site courses offered by public HEIs, thus demonstrating the trend toward the privatization of language teacher education in Brazil.
A recent study on the English language teacher workforce in public and private schools in Brazil has revealed a serious issue in relation to ELTE.Fewer than 30% of the approximately 985k English language classes offered in Brazilian schools are taught by teachers with adequate degrees (Gimenez;Viana, 2021).This means that most school pupils learn English with professionals who do not hold the minimum qualification set in the Brazilian educational legislation (cf.Brasil, 1996) -namely, a Bachelor's degree in English language with a teaching qualification.
There is therefore the urgent need of researching ELTE in Brazil, especially from a macro perspective, to identify commonalities across the country and understand where the field is heading.In response to the existing research gap in the available literature, the present study is aimed at answering the following two research questions.
(a) In which higher education contexts are future English language teachers educated in Brazil?(b) How socially inclusive are these higher education contexts?
In order to answer these questions, the present study draws on the data collected for the 2019 Brazilian Census of Higher Education, which is explained in the following section.Researchers interested in the Brazilian Census of Higher Education can engage in bibliographic research or in secondary data analysis.The former type of research entails the collection of relevant information from published texts (see Boon, 2018).In the case of the Brazilian census, this would mean a reliance on the information contained in the yearly technical summaries.This is the type of study reported in Silva (2020), which was reviewed in the previous section.Bibliographic research on the census is easier to conduct: it only requires access to the published materials, which are available online.However, without access to the primary data, it is difficult to ask new questions, and it is not possible to subject the data to new analyses.

METHODS
On the other hand, secondary data analysis allows for a freer exploration of available datasets (see Heaton, 2012;Vartanian, 2011).The limitations in secondary analyses are primarily related to the actual data that have been made available and/or collected.As mentioned earlier, availability is not an issue in relation to the Brazilian Census of Higher Education.With regard to the data that are collected, the census provides a considerable amount of useful data from all over Brazil.For example, the 2019 census surveyed 2,608 HEIs and 40,427 undergraduate courses, indicating that 3,633,320 students joined higher education in that year (INEP, 2021).The scope of the data collected for the census is unlikely to be matched by any academic research project relying on primary data analysis.
The present quantitative study entails a secondary data analysis of the Brazilian Census of Higher Education.This methodological choice is justified by our research aim: our specific focus lies on ELTE courses at undergraduate level.It would not be possible to answer our research questions with bibliographic research since the currently available technical summaries do not examine the census data at this level of detail.Undertaking secondary analysis allowed for the exploration of the census data in such a way that new questions could be asked about the future of ELTE in Brazil.
It was decided to work with the 2019 Brazilian Census of Higher Education, which was the most recent dataset available at the start of the present research project.The data collection for this census took place in 2020, and its completion was affected by the early stages of the COVID-19 pandemic.However, the data do not reflect the educational impact of the pandemic because they date back to 2019, the year before the outset of the pandemic.
Out of the four census components, we decided to work with the student component.
Our analysis considers only those students in Brazilian HEIs who are registered for courses which have an English language component (i.e., single degree courses in English language or joint degree courses in Portuguese and English).
The present secondary analysis was conducted on Microsoft Power BI, an analytics service that provides interactive data visualizations and allows researchers to easily identify relevant differences in large datasets.The data mining process was conducted by a statistician commissioned by the British Council Brazil for the Observatory for English Language Teaching.Our analysis relied primarily on the use of descriptive statistics, indicating the distribution of students per relevant variable (e.g., type of HEI administration) and considering important crossovers between different variables (e.g., ethnicity and type of HEI administration).The results are presented and discussed in the following two sections.

RESULTS
With a view to answering our research questions on the characterization and social inclusivity of Brazilian ELTE contexts, we focus on four main topics.We start by presenting the results of our analysis regarding the type of HEI administration where students are enrolled.
We go on to explore the type of English language undergraduate course taken by students, relating the most frequent type of course to graduates' professional destination.We then examine the delivery mode through which undergraduates study and highlight the relevant cross-over between delivery modes and the type of HEI administration.We later scrutinize students' diversity and consider the extent to which Brazilian HEIs are socially inclusive.

Student registration per type of HEI administration
HEIs in Brazil are divided into two types depending on their administration: (i) public ones, which are sponsored by city, state or federal governments and which do not charge for their undergraduate courses; and (ii) private ones, which may or may not be for profit and which charge tuition fees.
Overall, there are nearly eight times more private HEIs than public ones: 2,306 vs. 302 HEIs (INEP, 2021).Despite being more numerous, private HEIs are not usually perceived to be on a par with public ones.Research on the results of an annual assessment of undergraduate courses has evidenced the superiority of public HEIs over private ones (Hoffmann et al., 2014), and international evaluations such as the one by the Times Higher Education (see https://www.timeshighereducation.com/student/best-universities/best-universities-brazil)also show that public HEIs are the ones which are ranked highest (with some rare exceptions).
In relation to undergraduate students in English language, our analysis reveals that most students are affiliated with private HEIs (67.19%) and that less than a third of the students have a chance to study at public HEIs (32.81%).This result is a sign of a larger processthe marketization of higher education, which is not restricted to English language undergraduate courses.This process has been observed elsewhere (e.g., McCaig, 2018), but Brazil seems to stand out amongst the members of the Organization for Economic Co-operation and Development (OECD).More than three quarters of Brazilian undergraduates attend private institutions whereas in other OECD countries this proportion is of one third (OECD, 2021).
In Brazil, the marketization of higher education has resulted in the privatization of higher education (Pinto, 2004) and the charging of tuition fees.The distribution of English language undergraduates in Brazilian public and private HEIs seems to suggest that there is an ongoing process in the country where higher education is shifting from being a public good to being a commodity for trade.In discussing these two forces, Tilak (2008) warns that this shift may have dangerous implications, replacing academic values by commercial considerations, social concerns and purposes by individual interests, and long term needs by short term demands.Even if there are some gains to be had from the commoditization of higher education for trade, they may be few and short-lived, while the losses could be immense and may produce very serious, irreversible long-term dangers to the whole society (p.462).
The unbalanced distribution of students across HEIs has clear implications to access to higher education.Having to pay for higher education is not the traditional route in Brazil: students would attempt to be accepted by public HEIs where they can study for free at better ranked institutions.However, what the census data reveal is that public HEIs only offer higher education courses in English language for a minority of the population.Given the socioeconomic reality in Brazil, affording the costs of an undergraduate course in English language at a private HEI is not an option that is available to everyone.The larger provision of undergraduate courses by private HEIs in the country might therefore be construed as a factor that discriminates between those individuals who can access education and those who do not have the (financial) means to do so.
The clearly divergent student numbers in private and public HEIs may be linked to future graduates' professional chances.The minority studying at public HEIs is likely to be in an advantageous position when they transition from academia to the workplace given the social representation that public HEIs offer higher quality education as compared to private HEIs.In other words, graduates from public HEIs are likely to have higher institutionalized cultural capital in Bourdieu's (1986) sense given the way in which these HEIs are perceived in the country.
When the number of undergraduates in English language is examined per student status (cf.Table 1), it becomes clear that private HEIs have much higher percentages of students on leave of absence, withdrawn or transferred to other courses.These quantitative results indicate that the interruption of studies is a more common issue observed in private than in public HEIs be it a temporary or a permanent interruption.The census does not provide information on why larger percentages of undergraduates decide to discontinue their studies at private HEIs.A synthesis of studies by Teixeira, Mentges and Kampff (2019) indicated that the main reasons irrespective of the type of HEI administration (private or public) were financial hardship faced by students, uncertainty about career prospects and low academic achievement.Whatever the reason, the interruption of studies results in a financial loss for students who are self-funded and/or for the government in the case of those students in receipt of scholarships; an educational discontinuity or loss in the student experience; and a delay in graduation and in the transition to the workplace.For these reasons, it is important to design strategies to increase student retention (Araújo; Silva; Pederneiras, 2021).

Student registration per type of English language undergraduate course
In Brazil, the main difference in undergraduate courses in English language lies on whether they are (i) courses which are not focused on a specific career destination; or (ii) courses which contain an education component, preparing students to teach English in fundamental and upper-secondary education in Brazil.The students who graduate from (i) can still decide to teach English at schools if they successfully complete an additional qualification in English language teaching (cf.Brasil, 2017).
The vast majority of English language undergraduates in Brazil are registered for the course that has an educational component and qualifies them for school teaching (97.53%).
This suggests a clear association between the study of English, an additional language in Brazil, and the career destination of graduates.Either these students perceive this academic major as being intrinsically related to teaching or they do not want to be prevented from taking up a post in schools.Since joint degrees in Portuguese and English are much more common than single degrees in English, the appeal of a teaching-oriented courses is even stronger: graduates can teach either language in schools, thus doubling their employability chances.
A pragmatic orientation towards one's choice of university course is not restricted to ELTE undergraduates in Brazil.Copland et al. (2017) have noted the profession is identified as the main driver for students' decision to take an ELTE Master's course in the UK.The student participants in Copland et al. (2017) see their Master's degree as a way into or within the profession.These two professionally driven perspectives may also be shared by undergraduate students in ELTE in Brazil.The degree is the means through which they qualify as teachers, thus meeting the legal requirement to teach at schools; and, for those who are already in the job market, it can be a way of progressing in their careers. 3This pragmatic choice of university courses is aligned with Bourdieu's (1986) discussion on the conversion of capitalsmore specifically, from cultural to economic capital.In other words, students' extrinsic motivation to undertake the course is likely to be related to their perception that they will be able to be paid (or be better paid) for the knowledge that they have.
In the previous section, we noted that most students are based in private rather than in public HEIs (approximately 67% vs. 33%).This distribution does not hold true for both types of English language undergraduate course as shown in Table 2.The concentration of students in private HEIs is seen in relation to those who are pursuing a Bachelor's degree with a teaching qualification, which is the most popular type of ELTE undergraduate course in Brazil overall.This result seems to reinforce the pragmatic choice made by students in private HEIs.As they are paying for their higher education, these students favor a degree that will allow them to transition more seamlessly to the workplace.
However, student distribution between public and private HEIs is the opposite in relation to those who are reading for a Bachelor in English language: they are primarily based at public HEIs.The results also exemplify Tilak's (2008) point about the public good of higher education: Brazilian public HEIs play a key role in the education of students who wish to study English but do not necessarily aspire or plan to become English language schoolteachers.These institutions are vital in the academic socialization of these students into professions involving the English language other than teaching.

Student registration per delivery mode
English language undergraduate courses are delivered in Brazil through two modes: on campus or online.When the number of students is broken down per delivery mode, there is a slightly higher number of students registered for on-campus education (52.53%) as compared to online education (47.47%), and these figures relate to the 2019 pre-pandemic environment. 4he number of undergraduate courses per delivery mode is not as balanced as the distribution of students: on-campus courses (88.08%) are much more numerous than online ones (11.92%).The reduced number of on-line courses means that there is a considerable larger ratio of student per course in this delivery mode.One unanswered question that this larger ratio raises is how these courses succeed in providing students with individual support and attention, especially considering students' English language learning and their practicum experiences.
This is an aspect that needs to be investigated in the future because it has a clear impact on the quality of ELTE in Brazil.
The provision of online English language undergraduate courses in Brazil is almost exclusively done by the private sector, thus confirming the trend noted by Silva (2020).Out of the 47,161 online students, 94.47% study at private HEIs, especially for-profit ones (i.e., 84.78%).The provision of online education by public HEIs is extremely limited (only 5.53% of the online students in the country are affiliated with such HEIs), and these courses are only offered in 10 out of the 27 federative units in Brazil.The emergence of remote teaching in public HEIs as one of the consequences of the COVID-19 pandemic will probably change the proportion of undergraduate students in online courses.This aspect should be explored in future, post-pandemic studies.
Such an approach is understandable in geopolitical contexts which are a hub for international students like the US and UK.In Brazil, the student cohort in ELTE courses is more homogenous: there is a majority of Brazilian nationals who speak Portuguese as a first language.We have therefore decided to examine student diversity through an analysis of ethnic background, a socio-historically constructed identity that has been underexplored in the ELTE literature to date (see, for instance, Curtis;Romney, 2006;dos Santos;Windle, 2021;Kubota;Lin, 2009).
The ethnicity of undergraduates in Brazilian HEIs is surveyed by means of a closed question in the Census of Higher Education.The seven answers provided in the census questionnaire were combined into five categories in Table 3.The two changes made in our analysis consisted of grouping (i) Black and Mixed (i.e., Black and White) under a single category, which is allowed by the Brazilian law on racial equality (Brasil, 2010); and (ii) the answers 'no available information', 'the student did not want to declare ethnicity' and blank responses under the label 'no answer'.The results show that, while White students are the largest single ethnic group of English language undergraduate students in Brazilian HEIs (43.07%),Black students are the majority in public HEIs (cf.Table 2).It seems that Brazilian public HEIs are more socially inclusive spaces than private ones when it comes to undergraduate courses in ELTE.This result is probably a direct outcome of the affirmative action initiatives implemented by some public HEIs, especially those at federal or state levels. 5These institutions reserve a number of places in their courses for students from disadvantaged backgroundsbe they ethnic groups like 5 Public HEIs sponsored by the federal government have the highest percentage of Blacks (58.82%) when compared to the percentage observed within HEIs sponsored by state governments (42.42%) and city governments (20.15%).In the latter case, there is a clear deviation from the pattern observed for public HEIs, but these institutions have only 0.54% of the English language undergraduates in four (out of 27) federative units in Brazil.
Black, graduates from public high schools or those whose families do not meet a certain income threshold.In total, 27.81% of the students in public HEIs have been admitted to higher education through one of these affirmative action initiatives.This is in line with what has been observed in other studies on higher education in Brazil.For example, between 2014 and 2018, the number of Black graduates in universities have exceeded the number of White students (cf. Observatório do Fórum Nacional de Pró-Reitores de Assuntos Estudantis, 2019; see also Godoi;Santos, 2021).
One surprising result in Table 3 is the number of respondents for whom there is no information on ethnicity.This lack of information can be seen for approximately one-sixth of the population surveyed, thus making it difficult to have a full picture of how social inclusion/exclusion takes place in Brazilian HEIs.Recording one's ethnicity is a delicate matter, but it is extremely important to support policymaking initiatives and to ensure fairness and representativeness in ELTE undergraduate courses.An awareness campaign seems to be needed so that census respondents are alerted to the relevance of declaring ethnicity.

DISCUSSION
The present secondary analysis of the 2019 Brazilian Census of Higher Education has been instrumental in addressing our two research questions.Regarding the higher education contexts where future English language teachers in Brazil are educated, the results reveal that most are reading for online courses with an embedded teaching qualification offered by private HEIs.The results have also shown that the social inclusivity of this context is a matter of concern with a majority of White students despite the country's majority of Blacks.
Our data-driven results show that the main ELTE context in Brazil needs to address two main challenges: (a) student completion: private HEIs have higher percentages of students who have interrupted their studieseither temporarily or permanentlythan public ones (see Table 1); and (b) social inclusion: Black students are on the margins of the provision of ELTE undergraduate courses by private HEIs (see Table 3).
The first challenge might be related to other issues such as (i) the lower institutional reputation of private HEIs (e.g., Hoffmann et al., 2014) and (ii) the academic quality of online ELTE course provision, which requires further research.Approaching these issues in an integrated way might be a way forward.If the academic quality of online courses offered almost exclusively by private HEIs in Brazil is improved, this is likely to have a positive impact in the reputation of these institutions in the long run.Both quality and reputation improvement may contribute to reduce the number of students who decide to interrupt their undergraduate courses in English language.However, student completion deserves further investigation so that the crux of this challenge is clearly identified (e.g., whether it is indeed of an academic nature or whether it relates to financial issues).
The social inclusion challenge pointed out in the present study requires urgent attention and action by a range of stakeholders, educational policy makers and the owners of private HEIs.The student demographic of ELTE courses in Brazilian private HEIs follows the historical association between a White elite and ELT (Curtis & Romney, 2006;dos Santos & Windle, 2021;Ferreira, 2007Ferreira, , 2014;;Melo, 2015;Santos & Passos, 2021).An excellent example of social inclusion is found in public HEIs: they have a comparatively higher percentage of Black students than private ones, almost mirroring the percentage of this ethnic group in the Brazilian population (cf.IBGE, 2011).The present student make-up did not happen by chance or overnight.Brazilian public HEIs have implemented affirmative action initiatives, which, after a few years, have led to a structural transformation in that they have become more socially inclusive spaces.This example of good practice should be followed by private HEIs, thus democratizing access to ELTE courses and hopefully, as a result, to ELT.This aligns with Nero's (2006) suggestion that affirmative action should not only take place in the professional arena; it should also target "students of color at an early age who show promise for professional careers" (p.30).
In addition to the lack of inclusion of Black students in private HEIs, our analysis of the 2019 Brazilian Census of Higher Education also reveals a major social concern.English language undergraduate courses are not a public good available to everyone: only a minority can study at free-of-charge public HEIs.This is not limited to undergraduate courses in English language, though.The expansion of higher education in Brazil in the last decades has been largely due to the growth of the private sector, and the census data have confirmed this worrisome trend (cf.INEP, 2021).
The prevalence of private HEIs in ELTE is especially important to reconsider given the profile of students who are willing to opt for teacher education undergraduate courses in Brazil.
A study with final-year high-school students has shown that the interest in pursuing a teaching degree is higher among those from lower socio-economic classes as compared to those from higher classes (Fundação Carlos Chagas, 2009).This indicates that the students who may decide to read for a teaching degree are unlikely to have the means to pay for the tuition fees charged by private HEIs and they may face an issue in their successful applications to public HEIs given the limited places on

CONCLUSION
The present study has offered an original contribution to the field of ELTE, expanding its research boundaries.There is a dearth of studies examining the configuration of ELTE contextsboth in Brazil and internationally.This is especially the case with regard to the diversity of these contexts since the research conducted to date has primarily focused on an investigation of students' nationality and/or language background.In response to Romney's (2006) call that "the [TESOL] profession needs to move […] toward an acknowledgement of the demographics of the English language worldwide" (p.195), the present study has drawn on empirical data to shed light on the Brazilian ELTE contexts and their social inclusivity.
The scope of the present study is unparalleled, thus adding to its rigor.It draws on the Brazilian educational census data, which are collected by the federal government.Any attempt to gather similar data purely for research purposes is unlikely to have the same success rate, especially in a large country as Brazil.In this sense, the secondary analysis conducted for the present study has been instrumental in overcoming the challenges of reaching out to and engaging participants.The census data have been invaluable in the provision of a comprehensive description of how pre-service ELTE is implemented in Brazil, thus advancing our knowledge and understanding of the field at national level.
In relation to its significance, the study has made a strong case for the need to challenge and reconsider ELTE provision in Brazil.Our analysis has pointed out that this provision is characterized by the increased marketization of higher education.Further applied work needs to be undertaken to democratize access to ELTE courses, following a Freirean approach (Freire, 1987).We believe that language teacher policies which are truly committed to the economic and social development of ELTE should strengthen public HEIs since they play a key part in the inclusion of disadvantaged groups.These policies should additionally keep the private sector under control to make sure that quality is not sacrificed in the name of profit.
The applied work outlined here requires joint effort from several relevant stakeholders such as government officials, policy makers, researchers, practitioners, students and parents.
While the need for joint work is undeniable, this should not be used as an excuse for not engaging in it and waiting for government officials to act.As teacher education professionals, we should push the social agenda forward and help to trigger much needed changes in our field.
We would urge teacher educatorsin all disciplinesto engage in analyses of census data as a way of informing nationwide debates on the future directions of our specific fields.
This engagement provides us with most useful empirical results to support our suggestions, and it addresses the usual criticism of non-generalizability, targeted at localized and qualitative studies in specific HEIs.Having a detailed understanding of census data allows us to change from a passive standpoint where we are simply responsible for implementing the latest educational policy to an active standpoint where we can speak the language of policy makers, contribute to the national agenda and suggest key action points.This is an extremely relevant part of our work which must not be backgrounded nor entirely delegated to those from outside our field.
All HEIsboth public and privateare required to answer the yearly Census of Higher Education, conducted by the National Institute for Educational Studies and Research Anísio Teixeira (INEP).The census contains four components focusing on institutions, courses, lecturers and students.The main census findings are released in yearly technical summaries (e.g., INEP, 2021), and the primary census datain anonymized formare made available online.

Table 1 -
Status of undergraduate students in English language in Brazilian HEIs Source: The authors, based on the data from the 2019 Brazilian Census of Higher Education.

Table 2 -
Type of English language undergraduate course among students in Brazilian HEIs Source: The authors, based on the data from the 2019 Brazilian Census of Higher Education.

Table 3 -
Ethnic groups of undergraduate students in English language in Brazilian HEIs Source: The authors, based on the data from the 2019 Brazilian Census of Higher Education.