Black's access to higher education: what changed between 2000 and 2010

Abstracts

In recent years, the reduced presence of blacks in higher education has occupied an increasingly significant space in discussions of public policy agendas, social movement and academia. This work aims to study whether and how the observed changes in the profile of undergraduate students have changed in recent years. To this end, micro data of the Population Census 2000 and 2010 are used for the construction of the profile of students attending undergraduate school with a complementary analysis of their different training areas. The analysis indicates an improvement in the access rates to higher education for blacks, but with even distant results of their participation in the total Brazilian population. The paper presents the Racial Parity Index - RPI - as a synthetic indicator of the distance between blacks and whites in the different aspects presented.

Blacks; Higher Education; Racial Inequalities; Índice de Paridade Racial - IPR

Nos últimos anos, a presença reduzida de negros no ensino superior tem ocupado um espaço cada vez mais expressivo nas discussões das agendas de políticas públicas, do movimento social e da academia. Este trabalho busca estudar se e de que forma as mudanças observadas no perfil dos estudantes de graduação se alteraram nos últimos anos. Para tanto, utilizam-se os microdados dos Censos Demográficos de 2000 e 2010 para a construção do perfil do alunado que frequenta cursos de graduação com uma análise complementar de suas diferentes áreas de formação. As análises indicam uma melhora nas taxas de acesso à graduação para os negros, porém com resultados ainda distantes de sua participação no total da população brasileira. O trabalho apresenta o Índice de Paridade Racial - IPR - como um indicador sintético da distância entre negros e brancos nos diferentes aspectos apresentados.

Negros; Ensino Superior; Desigualdades Raciais; Índice de Paridade Racial - IPR

Los últimos años, la reducida presencia de negros en la educación superior ha ocupado un espacio cada vez más expresivo en las discusiones de las agendas de políticas públicas, del movimiento social y la academia. Este trabajo intenta estudiar si y de qué forma los cambios observados en el perfil de los estudiantes de graduación se alteraron los últimos años. Para ello se han utilizado los microdatos de los Censos Demográficos de 2000 y 2010 para construir el perfil del alumnado que frecuenta cursos de graduación con un análisis complementario de sus diferentes áreas de formación. Los análisis indican una mejora en los índices de acceso a la graduación para los negros, pero con resultados todavía distantes de su participación en la totalidad de la población brasileña. El trabajo presenta el Índice de Paridad Racial - IPR - como un indicador sintético de la distancia entre negros e blancos en los distintos aspectos presentados.

Negros; Ensenanza Superior; Desigualdades Raciales; Índice de Paridade Racial - IPR

Asymmetries between sex and color/race in Brazilian education have been the subject of reflection by many authors over the last thirty years: Hasenbalg (1979); Hasenbalg and Silva (2000); Henriques (2001); Beltran and Teixeira (2004); Paixão (2010);Rosemberg and Madsen (2011), among others. The literature has shown that inequalities in access, progression and completion of the different stages of basic and higher education are also bounded by the color/race of the students: educational indicators found for blacks are well below those observed for whites. Important differences between educational indicators for blacks and whites are observed from the first stage of the Brazilian educational system and should be studied in interaction with other social markers such as sex, region of residence, income and age.

Since 2004, affirmative action policies, mainly quotas (in the public sector) and scholarships in the private sector (partly financed by the University for All Program - Prouni)¹ have changed, even if slowly, the ethnic/racial profile of the students of Brazilian higher education. This work aims to study whether and how the observed changes in the profile of people who attend an undergraduate course have changed in recent years. To this end, 2000 and 2010 Population Census microdata are used (IBGE)² for the construction of the student body profile of undergraduate courses, with a complementary analysis of their different educational areas. In the analysis by training area, information derived from the Higher Education Census, organized by the Institute of Educational Studies and Research Anísio Teixeira - INEP ³ - 2010 data,⁴ is also presented with information disaggregated by sex and color/race. The variable color/race has a high non-response rate, close to 70%, which requires a careful analysis of their results.

Students in Brazilian higher education

One of the main references on the situation of the black segment in higher education are the Economic Analysis Laboratory Historical, Social and Statistics of Race Relations - LAESER publications, organized by Marcelo Paixão: "Annual Reports of racial inequalities in Brazil". The second report, published in 2010, presents data obtained from editions of the National Survey by Household Sampling - PNAD - 1988-2008, which indicate disparities by race/color and gender distribution in different aspects: health, social assistance, victimization, social security and access to education systems.

In the analysis of net attendance rates to higher education, indicating the effectiveness of an education system considering school attendance at the appropriate age, differences by race/color in access to higher education are striking:

The gross rate of schooling in higher education of white resident population rose from 12.4% in 1988 to 35.8% in 2008, representing an increase of 23.4 percentage points, or a proportional increase of 189.2%. In the case of black or parda population, the gross rate of schooling in higher education increased from 3.6% in 1988 to 16.4%. This development corresponds to an increase of 12.7 percentage points, or in proportional terms, from 350.4% in 2008. (PAIXÃO, 2010, p. 229)

The aforementioned report also notes that although the access of blacks and pardos (mixed-race)⁵ to higher education have tripled between 1995 and 2006, in 2006, in the age group of 18 and 24 years of age, these still represented only 6% of young people attending higher education (PAIXÃO, 2010). The report outlines that, in general, the arrival of blacks and pardos to higher education is a recent reality, probably associated with the increased access of this group to basic education in recent decades and affirmative action programs, created in the 2000s.

At the intersection of information by sex and color/race, the expansion of access to higher education is markedly higher for black women, as described in Table 1.

Table 1
Gross enrollment rate in higher education by gender and color/race - 1988 and 2008

In the combination of color/race and gender, white women are the ones with better indicators, followed by white men, black women and black men. In the period analyzed, the number of blacks enrolled showed the greatest growth, but with rates of attendance still significantly lower than whites, regardless of gender.

A similar result is presented by Rosemberg and Madsen (2011). Analyzing data from 2003 and 2009 PNAD editions, the authors state:

White men and white women representing 73.8% of the undergraduate students in 2003, now account for 42.3% in 2009; black men and black women representing 25.1% in 2003, now account for 35.1% in 2009. The percentage of variation (growth) in the period indicates higher rates among black men (95.6% variation), followed by black women (94.9% variation). On the other hand, white men, but mostly white women had the lowest growth rates: 22.8% and 19.5% respectively. (ROSEMBERG; MADSEN, 2011, p. 32)

It is as though there were a repressed demand for the access of blacks to higher education. However, despite the widening participation, the inequality between blacks and whites is still representative, if the participation in the whole general population is considered.

One of the referenced factors to increase the participation of blacks in higher education are affirmative action policies adopted by governments at the federal and state levels and in private educational institutions in Brazil over the past 10 years aimed at reducing social differences among whites, blacks and indigenous (ROSEMBERG; ANDRADE, 2008; FERES JR, 2011;DAFLON; FERES JR.; CAMPOS, 2013).

According to João Feres Jr. (2011), by the end of the 2010s, affirmative action programs already existed in most public institutions of higher education. In Brazil, 71% of public universities had some form of quota for public school students (87%) or for race (57%).⁶ However, when relating the adoption of affirmative action policies for quality education, as measured by the General Index of Courses - IGC - 2008⁷ the author states: "This calculation is revealing [...] we realize that the percentage of reserved seats for affirmative action programs decreases as the quality of the university rises" (FERES JR., 2011, p. 16).

The greater presence of blacks in higher education should also be analyzed considering the training areas or courses attended. Studies indicate that women and blacks are concentrated in less prestigious programs, although "prestige" is still a notion little explained in these studies. The most common reference in researches dealing with higher education student body composition are the analysis by dream careers (Law, Medicine and Engineering). Other studies consider the candidate-vacancy relation, which, however, limits the analysis to a single higher education institution - HEI (GUIMARÃES, 2002).

Researches in the field of Sociology showed that gender is an important indicator in relation to the prestige of a professional career. Predominantly feminine or masculine careers are the extremes of the scale and show, in the first case, poorly paid occupations and sometimes unstable careers, with shorter and/or marked by insecurity working hours and in the second, professions with higher salaries and better structured from the perspective of the professional career. The concept of the sexual division of labor explains these differences through two principles: the reproductive work is worth less than productive work; productive work of women is worth less than the productive work carried out by men (KERGOAT; HIRATA, 2003, p. 113). This is reflected in wage differentiation, always disadvantageous for women. Bourdieu (2003) also addresses the issue when he recognizes three principles of sexual division: the functions that are more suitable for women are the extension of their domestic duties (teaching, care, etc.); a woman cannot have authority over men; the male monopoly in the maintenance of technical objects and machines (BOURDIEU, 2003, p. 112-113). Studies such as Ribeiro and Klein's (1982) and Ferreti's (1976) were pioneers in addressing aspects of this issue in Brazil.

There are also studies on educational stratification, which analyze the influence of social background on academic performance: the lower the correlation between social background and student performance, more open and democratic is the school system. However, as pointed by Boudon⁸ (apud MONTALVÃO, 2011, p. 393), the expansion of the school does not necessarily lead to a decrease or a cooling of professional stratification process. The educational stratification reflects the social stratification and even in socialist countries, where social inequalities were smoothed, the distinctive effects of belonging to state bureaucratic class families are present (MONTALVÃO, 2011). In this case, one must take into account the development of educational systems and the ways in which prestige is distributed.

In Brazil, it is known that public universities hold advantage, and only a few private institutions can be considered prestigious, beyond occupation. Prates, Silva and Paula (2012), for example, considered the institutional nature of the HEI, as well as and the type of degree (bachelor's or teaching), using Treiman's Standard International Occupational Prestige Scale and the Brazilian Code of Occupations. The results showed that the elite public system provides most prestigious positions.

The racial question operates similarly to genre according to Bourdieu (2003, p. 111) when he declares that:

Whatever their position in the social space, women have in common the fact that they are separated from men by a negative symbolic coefficient that, just like skin color to blacks or any other sign of belonging to a stigmatized social group, affects negatively all they are and do.

One cannot fail to consider, however, the strong economic component associated with the racial question and the overlaps between gender and race and their results. What seems to be interesting to analyze in relation to higher education, is that women are the majority among students, even among black students, the female segment, white and black, experiences extraordinary expansion in the period analyzed, in absolute numbers.

The racial dimension of the population of graduates was treated by Kaizo Beltrão and Moema Teixeira on a text published by the Institute of Applied Economic Research - IPEA - in 2004, based on data from 1960-2000 Population Census, related to Brazilian students' racial belonging and gender and their career choices. The authors indicate:

What we can learn from the Brazilian census data regarding the insertion of different color/race groups in university careers is that this insertion is somehow mirroring the sex scale: in general, more masculine careers have a lower share of blacks and pardos and more feminine careers have greater participation of these groups. (BELTRÃO; TEIXEIRA, 2004, p. 35)

The analysis proposed in this article works with data from Population Census (IBGE) and the Higher Education Census (INEP). These are quantitative researches with distinguished features and structures. Thus, the comparison of its results should be made with caution. The Population Censuses inform which graduation training course only to people who have already had their certificate. It is a sum measure because the information does not depend on the study period. Since the presence of blacks in higher education is recent, probably the results fall short of the rates observed for students attending higher education in the information collection period. As a comparison, data from 2010 Higher Education Census are presented by training areas. It should be noted that high non-response rate for the question about color/race, central for the study proposed here, in the INEP researches limit the analyses.

Profile of students by race/color based on Population Censuses

This study analyzes 2000 and 2010 Population Censuses microdata. The comparison between the censuses results allows the analysis of the participation of blacks in access to and completion of higher education before and after the increase in public policies of affirmative action. ⁹

The main analysis question is the color/race of the respondent. For the question your color or race is, the response options are: white, black, yellow, pardo and indigenous. The term "black" indicates the sum of the values found for blacks and pardos, as it has been very common in studies of that kind. The use of the term "black" facilitates the analysis, given the social indicators described in the literature approach the values found for blacks and pardos (ROSEMBERG et al., 1986). We have adopted the concept of race according to Guimarães (2002, p. 50), for whom race is a socially constituted category, "not just a political class needed to organize resistance against racism, but it is also an indispensable analytical category: the one that reveals the discrimination and inequality that the notion of 'color' entails are effectively racial and not just 'of class'".

The distribution of the Brazilian population by race/color based on Population Censuses already deserves a first observation. There was a change in the racial configuration between 2000 and 2010, with a greater presence of blacks compared to whites for the total population. For Silva (2013, p. 114), "we may say it is the recovery of the black population's participation, majority in the 19^th century and that, at the beginning of the last century, especially with the great European immigration flow, lost representation in the national population context".

Table 2 presents data from 2000 and 2010 Censuses regarding the distribution of the population by race/color.

Table 2
Distribution of the population by race/color and variation in the period - 2000 and 2010

The decrease of those who declared themselves as white in the general population between the 2000 and 2010 Censuses is compensated by the expansion of those who declared themselves as black or pardo, with a slightly higher rate for blacks (variation of 36.0%) than for pardos (variation of 26.8%). In the population in 2000, blacks and pardos represented 44.6% and in 2010, 50.9% of the universe. That modification has been detected by other authors based on the results of the National Household Survey conducted in the last decade. According to Cunha (2012, p. 3):

This phenomenon can be attributed both to a fertility differential - the global fertility rate of black women is 2.1 and white 1.6 children per female, in the replacement level of the two groups - and/or the systematic increase of the population who declares being black due to a process of awareness of the importance of taking on their own identity.

As a product of self-classification, the mestizo category (designated by the official statistics, pardo) used to swing to the white categorization because of social upward mobility, one of the faces of the whitening process (DOMINGUES, 2002). It should be noted, however, a tendency to reversal of this phenomenon in recent decades, when there seems to be higher self-identification of the population with the black segment (pardo and black), according to Soares (2008). The highest relative growth rate is in the "black" (preto) category, according to the author, probably by the most direct correspondence with the term ""negro" (black).¹⁰

Thus, according to Soares's study (2008), the increase of black and pardo population (black) would not be justified only by the changes in fertility rates, but it would be explained, mainly, by the modification of the self-classification of the population: "It can be said that what is happening is that Brazil is not becoming a nation of blacks, but it is accepting itself as such" (SOARES, 2008, p. 116). In a study with PNAD data from 1982 to 2007, Marteleto (2012) verifies that educational disadvantage among black and pardo groups, higher at beginning of period, becomes lower by the end of the same period, making the group more homogenous in terms of educational attributes.¹¹

In general terms, that is the discussion about the changes of the racial composition of the Brazilian population, a topic that will not be deepened. However, it is important to point that out, since the changes in the distribution of whites and blacks in the general population have a direct impact on the different levels of education set out below and should be considered in the analysis of the results found by color/race. Furthermore, the improvement of educational indicators could be explained, for example, not by increased education, but by changes in the qualifying process of the people. On the other hand, less important than to note the growing identification process with the "black" or "negro" categories among the population, it is important to observe the proximity of black and pardo groups in social and economic terms. Hence the usefulness of grouping them to analyze how this segment of the population is considered in the case of this study, in terms of educational opportunity.

For this study, four types of analysis will be presented: 1) comparison between results of 2000 and 2010 for a selected set of variables; 2) information by sex and color/race, with the formation of subgroups: white women, white men, black women and black men, for the 2010 Census; 3) Gender Parity Index - GPI - and Race Parity Index - RPI - as indicatives of the distances between men and women and blacks and whites in access to higher education; 4) an analysis by training areas considering sex and color/race.

In Chart 1, a comparison between the Censuses for people who attend undergraduate school and general data of the population is made.

Chart 1
Distribution of people attending undergraduate courses and the population, by category of the selected variables, according to the period. Brasil, 2000 and 2010

The results indicate a significant increase in the rates of access to undergraduate school (116.4%) compared to the total population (12.8%). The increase is more pronounced in the variable color/race, with rates that reach 290.7% of black undergraduate students. In the breakdown by blacks and pardos, the rates of change are greater for the first group. In a first reading, the impression can be of a better balance in access to higher education for blacks and whites, but the share of blacks compared to their representation in the Brazilian population still shows that inequalities are a fundamental imprint on the Brazilian higher education. In 2010, blacks accounted for 50.9% of the general population, index decreasing to 35.3% in groups of people attending undergraduate school. Interestingly, if the absolute number of white students (3.9 million) represents 4.3% of the total population of whites (90 million), the same rate for blacks is 2.3% and for pardos, 2.2%.

When the sex and color/race variables are? integrated, it is observed that black women show the highest growth in rates in undergraduate school: 294.9%, followed by black men, with 284.9%.

In the analysis by regions, it is observed that there is a rise in the North and Northeast regions, which have received a greater allocation of resources from the federal government and funding agencies aiming to reduce regional disparities described in the literature (CAPES, 2010).

Most students attend undergraduate school at the expected age (18-24 years of age). However, the expansion of higher education in the decade analyzed is reflected in rates for those over the age of 30, a population that probably resumed his studies with the expansion of supply and reduction of fees, phenomena observed in most private educational institutions.

Table 3 shows disaggregated information for undergraduate school by educational institution administrative category, for the years 2000 and 2010 in the four groups of sex and color/race (white men, black men, white women and black women) and disaggregated information for blacks and pardos.

Table 3
Distribution of people attending undergraduate courses by gender and color / race, according to the administrative category of the higher education institution (HEI) and the period. Brazil, in 2000 and 2010

The largest share of undergraduate students is concentrated on private educational institutions, and it is close to 70%. White women predominate in both public and private institutions. In the analysis by sex and color/race, there are more black women than black men, indicating that the bigger female presence is maintained in the interface by color/race, as Rosemberg already stated (2001). However, comparing the presence of each group in the general population and the access to undergraduate school, inequality by race/color is again evident. This way, the rates of change from 2000 to 2010, according to the absolute numbers of students which pass the 300% for blacks, in comparison to the 100% increase on the total student body, must be examined with caution. The presentation of information to blacks and pardos reveals a major gap between the two groups, but the greatest inequality will be between the white and the black stratum. Hence, it can be observed that the separate rates of change of blacks and pardos show a similar expanding trend. Greater access to higher education, apparently, is for black and pardo, and the black category may be considered appropriate to display the distances observed between whites and blacks in access to higher education, despite the advances in the decade described.

Another important variable to consider is the distribution of students by region. Blacks are more concentrated in regions where the number of higher education institutions is lower, which may affect the access rates by race/color. Table 4 shows a comparison by region and sex and color/race and the general population.

Table 4
Attendance of undergraduate students and population, by gender and color / race, according to the region. Brazil, 2010

The proportion of each racial segment in higher education institutions is presented by the distribution of these segments in the population of each region of Brazil. It is observed that racial inequalities in access remain. In the North and Northeast, for example, the proportion of blacks in the population approaches 70%. The access rates to higher education, although closer, remains clearly below: in the North, blacks represent 64.4% of students, and in the Northeast, 56.3%. In the Southeast, the disparity appears to be more consistent: the black population corresponds to 43.8% of the total of this region, but the ones attending an HEI account for only 28.1%. In the comparison between black women and black men, the indicators are favorable to black women in all regions. The same situation can be described for the comparison between white women and white men. Yet, it is possible to observe that in all regions, the white population is overrepresented among students in relation to their proportion in the general population, while the black population is always proportionally underrepresented.

A separate analysis for the black and pardo groups by sex is described in Graphs 1 and 2. The choice was to display information only to the Northeast and Southeast regions, which polarize the participation of blacks in the population and in access to higher education. The analysis confirms and the graphics ratify, from another angle, the data presented previously.

Graph 1
People attending undergraduate school compared to the participation in the general population by white, black and pardos men, Northeast and Southeast regions (2010) (%)

Graph 2
People attending undergraduate school compared to the participation in the general population by white, black and pardo women, Northeast and Southeast (2010) (%)

Separating the information about black and pardo, the analysis of the charts indicates a proximity in behavior regardless the declared color or the region. It is worth noting once again, that white men and women have a higher participation in the group of people attending undergraduate school comparatively to their participation in the general population.

Another strategy for the presentation of differences by race/color observed in the groups of students is the use of a measurement of distance between the groups. In the case of gender differences, it is possible to use the GPI¹² developed by UNESCO - United Nations Educational, Scientific and Cultural Organization - in its global monitoring reports "Education for All". An adaptation to the differences by color/race, the RPI is presented by Rosemberg (2006) under the name of Blackness Rate.¹³ This indicator mirrors itself on the GPI, replacing the variable sex by the variable color/race: a numerical indicator for blacks in proportion to the same indicator for whites. The advantage of the application of these indicators is that they emphasize the differences found between men and women and between blacks and whites. The closer the value of the index to 1, the smaller the distance between women and men, in the case of GPI, and between blacks and whites in the case of RPI. Supported by UNESCO's analyses, values between 0.95 and 1.05 as the reference range are accepted, any indicator lower than 0.95 represents an inequality for women or blacks and any indicator higher than 1.05, an advantage (UNESCO, 2004).On the whole of the population, the GPI values are 1.03 for 2000 and 1.04 for 2010. For RPI, the value is 0.83 for 2000, reaching 1.07 in 2010.¹⁴ This way, it is observed that there is gender parity in the analyzed period and an inversion on racial belonging: in 2000, the prevalence in the population was of the whites and 2010, of the blacks, as previously reported.

A pertinent criticism of the use of RPI is that the GPI is built from the comparison of the presence of men and women, a bipolarity which ends in itself.¹⁵ In the case of the color/race category, just considering the possibilities for responses set by IBGE, there are five options to choose from (white, black, yellow, pardo and indigenous). Although the black categories (black and pardo) and white do not constitute dichotomous categories, just as sex, currently they make up more than 98% of the population (cf. Graph 2), each group making up about one half.

It should be noted, however, that, on the one hand the idea of parity means "feature of what is even" and also means "equality, balance (e.g., of salary, between similar levels of different professions, of exchange, of price etc.)" (HOUAISS, 2009, p. 1436). We think, therefore, that the RPI has potential as a measure of equal participation of the two main ethnic groups of the population. Thus, the possibility of using an indicator that measures inequality between blacks and whites, even disregarding the yellow and indigenous segments, can serve to highlight the distance between the groups (black and white) by different variables. The RPI thus constituted, instead of working with two values (numbers or percentages separately for black and white groups), synthesizes the distance between the groups in a single indicator. It is noteworthy that other studies present results using the expressions "white" and "nonwhite" (ANDRADE; DACHS, 2007; MAGALHÃES, 2009; SILVEIRA, MUNIZ, 2014; OBSERVATORIO DAS METROPOLES, 2013); or blacks and non-blacks (DIEESE, 1999; SANTOS, 2005; SEADE, 2012), which further reinforces the idea that the race issue, for good or ill, has been understood in dichotomous terms not only in studies which have addressed the issue but also in race equality policies of recent years.

Sergio Costa (2002) discusses the polarization of analytical categories "white" and "non-white" or "black" in the grouping of categories about blacks and pardos as opposed to "white" used by IBGE. According to the author, the opposition "gives visibility to racial ascriptions which co-determine social injustice in Brazil, becoming [...] analytically legitimate contribution and from the point of view of an anti-racist policy, precise and indispensable" (COSTA, 2002, p. 49).

Table 5 presents the RPI for the people attending the undergraduate school and the general population for the years 2000 and 2010.

Table 5
Gender Parity Index (GPI) and Racial (RPI) for people who attend undergraduate or post graduation and for the general population, according to the period. Brazil, 2000 and 2010

The distances between blacks and whites are more clearly emphasized with the use of a single indicator. What is observed in the analysis of the results is that, apart from the change in racial self-identification profile of the general population from 0.83 (2000) to 1.07 (2010), access to undergraduate school for blacks has been expanded over the period considered: in 2000 the ratio was four whites to a black in undergraduate school, this rate decreased to two whites to every black in 2010.Compared to the GPI, there is maintenance in distribution, favorable to women, as observed in 2000.

Training areas and prestige

An important aspect in the analysis by sex and color/race is the different distributions by careers or courses. Chart 2 shows the ranking of the 20 training areas with greater demand. It is noteworthy that information derived from the IBGE indicates the area of training of students, a combined indicator, that is, all undergraduate students by that date, captured by the data collection for the 2010 Census. The data from the INEP indicate the students enrolled in the HEI in 2010. So, the comparisons made have a tendency and approximation nature. Again, the RPI for Higher Education Census data should be considered with caution because of the high rate of non-response to the question in INEP basis.

Chart 2
Top twenty training areas which comprise the greatest number of graduated students, with other selected variables. Brasil, 2010

The 20 areas with higher frequency aggregate 76.5% of graduates and 78.7% of students. There has been a position overlapping in the ranking considering the five areas with the highest participation, that gathers 50% of students or graduates (calculation based on the columns of percentage of IBGE and INEP). The presentation of GPI and RPI columns (columns 3 and 4) shows that the differences found are best defined by sex than by the color/race. The GPI rate switches from 0.23, for civil and construction engineering - IBGE, to 11.88 for Education Sciences - INEP, a variation of 5.065%. In the case of RPI, the predominance of whites, as expected, occurs in all courses, with values of 0.16 to dentistry - IBGE, to 0.84, for nursing and primary care - INEP, a variation of 425%. The high non-response rates to the question of color/race in the Higher Education Census - INEP should be considered in the analysis, but in comparison with those submitted by IBGE, the trends remain.

Compared by sex, the largest GPIs are observed in education (science education, IBGE 11.70 and 11.88 INEP) and healthcare (nursing and primary care, IBGE 6.60 and 5.23 INEP), higher female concentration areas as described in literature (BELTRÃO; TEIXEIRA, 2004; MELO; LASTRES; MARQUES, 2004). The areas of engineering concentrate the lowest GPIs, with a ratio of four men to every woman in the IBGE and three men to every woman in INEP. It is good to remember that each research prospects a different universe, IBGE - graduates and INEP - students, these differences can be understood through a more recent presence of women in areas where previously there was a higher male predominance, as pointed out by Bruschini et al. (2011, p. 160-161), among others.

Path taken and current challenges

This initial study aimed to describe the changes in access rates to higher education in Brazil based on the variable color/race and sex from the last two Population Censuses, by taking some auxiliary data from the Higher Education Census as well. The latter, although very detailed in many aspects, not only in the presentation of information on the institutional nature of HEI but also in the form of students admission, still need to move forward in completing the racial aspect. It is known, however, that this problem is not the sole responsibility of INEP, the completion of the data must be done by the HEIs themselves through online registration.

The Higher Education Census is an important database, with unique information, yet still little explored in studies on higher education. Attributes of the HEI, such as the administrative category (universities, colleges, etc.), the administrative organization (public and private, with or without profit) and the different forms of education (classroom or distance), associated with gender and race compositions may give clues to the prestigious positions in higher education and those occupied by women and blacks.

Qualifying the student body of this level of education is important and necessary, taking into consideration its recent transformations. In the last two decades, the Brazilian higher education system has undergone a significant expansion. In the middle of last decade, the introduction of a policy of affirmative action in higher education is another factor to be considered.

In the period analyzed by this study, women remain majority, while the black segment had an extraordinary relative growth. Not surprisingly, the highest relative growth was among black women (black and pardo). Still, the black segment is underrepresented in all regions, not only in the ones where it is still a minority (such as South and Southeast) but also where it is majority (North, Northeast and Midwest). Although most of the students are in the age range considered ideal (18 to 24), there is considerable growth in the age range 25 to 29 and above 30 years of age, which indicates late access to students, many of them laborers. On the other hand, the little difference between the group of black students in private and public higher education institutions (with a slight advantage to the latter), continues to show the their influence to the group, even if the differences in Brazilian regional and ethnic/racial composition are taken into consideration.

In this regard, although the discussion about the usefulness of an indicator that measures the distance between blacks and whites needs to be developed, taking into account its advantages and limitations, it was considered that the RPI is an important tool to give visibility to the racial composition of higher education, as well as to underscore, as a synthetic indicator, the inequalities between the two main ethnic groups that make up the Brazilian population.

Finally, from the point of view of inclusion of historically disadvantaged and discriminated segments of the population (blacks and pardas, women), it can be said that important spaces were conquered. However, the distances are still significant. Their measurement and breakdown will help to understand and support the political decisions to tackle them. Measuring how much and in what ways that distance was changed by affirmative action policies and other socio-economic factors in the period studied is a path to be taken which requires new studies for its unveiling.

• ANDRADE, Cibele; DACHS, J. Norberto. Acesso à educação por faixas etárias segundo renda e cor/raça. Cadernos de Pesquisa, São Paulo, v. 37, n. 131, p. 399-422, 2007.
• BELTRÃO, Kaizo; TEIXEIRA, Moema de P. O vermelho e o negro: raça e gênero na universidade brasileira - uma análise da seletividade das carreiras a partir dos censos demográficos de 1960 a 2000. Rio de Janeiro: Ipea, 2004. (Texto para Discussão n. 1052)
• BOURDIEU, Pierre. A dominação masculina. Rio de Janeiro: Bertrand Brasil, 2003.
• BRASIL. Coordenadoria de Aperfeiçoamento de Pessoal em Nível Superior. Plano Nacional de Pós-Graduação - PNPG 2011-2020. Brasília, DF: Capes, 2010.
• BRASIL. Ministério da Educação. Tire suas dúvidas. Brasília, DF: MEC. 2015. Disponível em: <http://siteprouni.mec.gov.br/tire_suas_duvidas.php#conhecendo>. Acesso em: ago. 2015.
  » http://siteprouni.mec.gov.br/tire_suas_duvidas.php#conhecendo
• BRUSCHINI, Cristina et al. Trabalho, renda e políticas sociais: avanços e desafios. In: BARSTED, Leila; PITANGUY, Jacqueline (Org.). O progresso das mulheres no Brasil 2003-2010. Rio de Janeiro: Cepia; Brasília, DF: ONU Mulheres, 2011. p. 142-177.
• COSTA, Sergio. A construção sociológica da raça. Estudos Afro-Asiáticos, ano 24, n. 1, p. 35-61, 2002.
• CUNHA, Estela Maria G. P. Brasil está reduzindo suas disparidades raciais? In: ENCONTRO NACIONAL DE ESTUDOS POPULACIONAIS - ABEP, 18., Águas de Lindoia, 2012. Anais... Águas de Lindoia: Abep, 2012.
• DAFLON, Verônica Toste; FERES JUNIOR, João; CAMPOS, Luiz Augusto. Ações afirmativas raciais no ensino superior público brasileiro: um panorama analítico. Cadernos de Pesquisa, São Paulo, v. 43, n. 148, p. 302-327, abr.2013. Disponível em: <http://dx.doi.org/10.1590/S0100-15742013000100015>. Acesso em: 22 ago. 2014.
  » http://dx.doi.org/10.1590/S0100-15742013000100015
• DEPARTAMENTO INTERSINDICAL DE ESTATÍSTICA E ESTUDOS SOCIOECONÔMICOS. Mapa da população negra no mercado de trabalho. São Paulo: Dieese, 1999.
• DOMINGUES, Petrônio José. Negros de almas brancas? A ideologia do branqueamento no interior da comunidade negra em São Paulo, 1915-1930., Estudos Afro-Asiáticos Rio de Janeiro, v. 24, n. 3, p. 563-600, 2002.
• FERES JR., João (Org.). Ação afirmativa no ensino superior brasileiro hoje: análise de desenho institucional 2011. Rio de Janeiro: Gemaa - Grupo de Estudos Multidisciplinares da Ação Afirmativa IESP/UERJ, 2011.
• FERRETI, Celso. A mulher e a escolha vocacional., Cadernos de Pesquisa São Paulo, n. 16, p. 20-40, 1976.
• GUIMARÃES, Antônio Sergio. Classes, raças e democracia. São Paulo: 34, 2002.
• HASENBALG, Carlos. Discriminação e desigualdades raciais no Brasil. Rio de Janeiro: Graal, 1979.
• HENRIQUES, Ricardo. Desigualdade racial no Brasil: evolução das condições de vida na década de 90. Rio de Janeiro: Ipea, 2001. (Textos para Discussão n. 807).
• HOUAISS, Antonio; VILLAR, Mauro de Salles. Dicionário Houaiss da língua portuguesa. Rio de Janeiro: Objetiva, 2009.
• INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍTICA. Censo Demográfico. Microdados, 2010.
• ______. Censo Demográfico. Microdados, 2000.
• KERGOAT, Danièle; HIRATA, Helena. A divisão sexual do trabalho revisitada. In: MARUANI, Margaret; HIRATA, Helena (Org.). As novas fronteiras da desigualdade: homens e mulheres no mercado de trabalho. São Paulo: Senac, 2003. p. 111-123.
• MAGALHÃES, Alânia. Trabalho e educação: os efeitos da educação sobre o processo de estratificação social entre brancos e não brancos no Brasil. In: SEMINÁRIO NACIONAL DE SOCIOLOGIA E POLÍTICA, 1., UFPR, 2009. Grupo de trabalho 5
• MARTELETO, Letícia J. Educational inequality by race in Brasil, 1982-2007: structural changes and shifts in racial classification. Demography, n. 49, p. 337-358, 2012.
• MELO, Hildete Pereira; LASTRES, Helena Maria Martins; MARQUES, Teresa Cristina Novaes. Gênero no sistema de ciências, tecnologia e inovação no Brasil. Revista Gênero, v. 4, n. 2, p. 73-94, 2004.
• MONTALVÃO, Arnaldo. Estratificação educacional no Brasil do século XXI. Dados - Revista de Ciências Sociais, Rio de Janeiro, v. 54, n. 2, p. 389-430, 2011.
• OBSERVATÓRIO DAS METRÓPOLES. Distância social entre brancos e não-brancos. 2013. Disponível em: <http://web.observatoriodasmetropoles.net/index.php?option=com_content&view=article&id=45&Itemid=114&lang=pt>. Acesso em: 19 set. 2013.
  » http://web.observatoriodasmetropoles.net/index.php?option=com_content&view=article&id=45&Itemid=114&lang=pt
• ORGANIZAÇÃO DAS NAÇÕES UNIDAS PARA A EDUCAÇÃO, A CIÊNCIA E A CULTURA. Relatório de monitoramento global educação para todos, 2003-2004. São Paulo: Moderna, 2004.
• PAIXÃO, Marcelo (Org.). Relatório anual das desigualdades raciais no Brasil: 2009-2010. Rio de Janeiro: UERJ, 2010.
• PRATES, Antonio A. P.; SILVA, Matheus F.; PAULA, Tulio S. Natureza administrativa das instituições de ensino superior, gestão organizacional e o acesso aos postos de trabalho de maior prestígio no mercado de trabalho. Sociedade e Estado, v. 27, n. 1, p. 25-44, jan./abr. 2012.
• RIBEIRO, Sergio Costa; KLEIN, Ruben. A divisão interna da universidade: posição social das carreiras. Educação e Seleção, n. 5, p. 29-43, jan./jun. 1982.
• ROSEMBERG, Fúlvia. Ação afirmativa no ensino superior brasileiro. Programa de Ações Afirmativas em Debate. São Carlos: UFSCar, 2006.
• ______. Educação formal, mulher e gênero no Brasil contemporâneo. Revista Estudos Feministas, Florianópolis, v. 9, n. 2, p. 515-540, 2001.
• ROSEMBERG, Fúlvia; ANDRADE, Leandro. Ação afirmativa no ensino superior brasileiro: a tensão entre raça/etnia e gênero. Cadernos Pagu, n. 31, p. 419-437, jul./dez. 2008.
• ROSEMBERG, Fúlvia; MADSEN, Nina. Educação formal, mulheres e gênero no Brasil contemporâneo. In: BARSTED, LeilaL.; PITANGUY, Jacqueline (Org.). O progresso das mulheres no Brasil 2003-2010. Rio de Janeiro: Cepia; Brasília, DF: ONU Mulheres, 2011. p. 390-424.
• ROSEMBERG, Fúlvia et al. Diagnóstico sobre a situação educacional de negros (pretos e pardos) no estado de São Paulo. São Paulo: FCC/Secretaria de Educação do Estado de São Paulo, 1986. 2 v.
• SANTOS, Ângela Maria dos. Vozes e silêncio do cotidiano escolar: análise das relações raciais entre alunos negros e não-negros em duas escolas públicas no município de Cáceres-MT. Dissertação (Mestrado em Educação) - Universidade Federal do Mato Grosso, Cuiabá, 2005.
• SÃO PAULO. Fundação Sistema Estadual de Análise de Dados. Os negros no mercado de trabalho na região metropolitana de São Paulo: o mercado de trabalho em 2012. Sistema de pesquisa de emprego e desemprego. São Paulo: Seade, nov. 2013.
• SILVA, Nelson do Valle;. HASENBALG, Carlos Tendências da desigualdade educacional no Brasil. Dados, v. 43, n. 3, p. 423-445, 2000. Disponível em: <http://dx.doi.org/10.1590/S0011-52582000000300001>. Acesso em: 30 abr. 2014.
  » http://dx.doi.org/10.1590/S0011-52582000000300001
• SILVA, Tatiana Dias. Mulheres negras, pobreza e desigualdade de renda. In: MARCONDES, Mariana et al. (Org.). Dossiê Mulheres Negras: retrato das condições de vida das mulheres negras no Brasil. Brasília, DF: Ipea, 2013.
• SILVEIRA, Leonardo Souza; MUNIZ, Jerônimo Oliveira. Diferenciais de rendimento entre brancos e negros: uma comparação intra e inter metropolitana. Cadernos Metrópole, São Paulo, v. 16 n. 31, jun. 2014.
• SOARES, Sergei. A demografia da cor: a composição da população brasileira de 1890-2007. In: THEODORO, Mario (Org.). As políticas públicas e a desigualdade racial no Brasil 120 anos após a abolição. Brasília, DF: Ipea, 2008.

• 1
  O Prouni é um programa do Ministério da Educação, criado pelo Governo Federal em 2004, que concede bolsas de estudo integrais e parciais (50%) em instituições privadas de ensino superior, em cursos de graduação e sequenciais de formação específica, a estudantes brasileiros sem diploma de nível superior (BRASIL, 2015).
• 2
  O Instituto Brasileiro de Geografia e Estatística - IBGE - é o órgão do governo brasileiro que produz estatísticas nacionais.
• 3
  O Inep é uma autarquia federal vinculada ao Ministério da Educação que promove estudos e pesquisas sobre educação.
• 4
  O Censo do Ensino Superior é publicado pelo Inep desde 1995, porém apenas em 2009 foi instituído o módulo alunos, que permite uma caracterização do perfil de matriculados nos diferentes cursos de graduação.
• 5
  Em nenhuma instituição analisada pelo autor, o pertencimento racial era o critério único de aceite, estando, quando presente, associado à menor renda.
• 6
  O IGP (Índice Geral dos Cursos da Instituição) considera o desempenho dos estudantes no Exame Nacional de Desempenho dos Estudantes - Enade -, bem como a avaliação do perfil do corpo docente, a infraestrutura e a organização didático-pedagógica da instituição durante um triênio.
• 7
  BOUDON, Raymond. Education, opportunity and social inequality. New York: Willey, 1974.
• 8
  Os projetos e as propostas de ação afirmativa têm por público-alvo, além dos negros, os autodeclarados indígenas. Em razão do pequeno universo que representam no geral da população (0,7%) e em função da diversidade do perfil e das demandas para seu atendimento, sua caracterização não será descrita neste estudo.
• 9
  A autora, inclusive, evidencia que o grupo "preto" está, no final do período, em ligeira vantagem em relação ao grupo "pardo", o que reforçaria a hipótese de Soares (2008) de maior identificação com esse segmento. O fenômeno de embranquecimento pela ascensão social estaria dando lugar a um "escurecimento com educação" (darkening with education), isto é, o maior nível educacional levando a uma maior autoidentificação racial negra (MARTELETO, 2012, p. 339).
• 10
  Índice de Paridade de Gênero - IPG: razão do valor mulher/homem de determinado indicador. Um IPG com valor 1 indica paridade entre os gêneros; um IPG que varia de 0 a 1 significa uma disparidade em favor de meninos; um IPG maior do que 1 indica disparidade em favor de meninas (UNESCO, 2004, p. 386).
• 11
  A adoção dessa nova denominação será explicada adiante.
• 12
  Adaptando o intervalo de referência para os valores brasileiros, tem-se a paridade para as mulheres nos intervalos de 0,98 a 1,08 (2000) e 0,99 a 1,09 (2010); para a taxa de negritude, a paridade em 2000 é qualquer valor entre 0,78 e 0,88 e para 2010, 1,02 a 1,12.
• 13
  Mesmo com as recentes discussões teóricas que refletem a ideia dicotômica da construção sexual, a categoria sexo ainda é útil e coletada como uma variável dicotômica constituída por homens e mulheres. Não avançaram as discussões sobre como pesquisar as diferentes orientações e performances de gênero, de forma alternativa à categoria sexo.
• 14
  Prouni is a program of the Ministry of Education, created by the Federal Government in 2004, which grants full and partial scholarships (50%) in private institutions of higher education in undergraduate and sequential specific training, to Brazilian students without higher education qualifications (TIRE SUAS..., 2015).
• 15
  The Brazilian Institute of Geography and Statistics - IBGE - is the Brazilian government agency that produces national statistics.
• 16
  INEP is a federal agency under the Ministry of Education that promotes studies and research on education.
• 17
  The Census of Higher Education has been published by INEP since 1995, but only in 2009 the student module was set up, which allows a characterization of the profile of students enrolled in different undergraduate courses.
• 18
  Given the peculiarity of the Brazilian racial classification system, it was decided not to translate term "brown", which corresponds to a designation of the mestizo population, especially between blacks and whites, but also indigenous.
• 19
  In all the institutions analyzed by the author, the racial status was not the only criterion accepted and was, when present, associated with lower income.
• 20
  The IGP - General Index of Courses - considers students' performance in the National Student Performance Exam - Enade - as well as the evaluation of the academic staff, the infrastructure and the didactic-pedagogical organization of the institution during a three-year period.
• 21
  BOUDON, Raymond. Education, opportunity and social inequality. New York: Willey, 1974.
• 22
  The projects and affirmative action proposals aim not only blacks but also the self-declared indigenous. Because of the small universe they represent in the general population (0.7%) and depending on the diversity of profiles and demands for their assistance, their profiling will not be described in this study.
• 23
  The term "negro" (black) (in Portuguese) is preferred by Brazilian black movements to designate the group of blacks and pardos. Among black American movements, that designation was rejected for being associated with the long history of slavery.
• 24
  The author even shows that at the end of period, the "black" group is, in slight advantage over the "pardo" group, which would strengthen Soares's hypothesis (2008) of greater identification with this segment. The whitening phenomenon for social mobility would be giving place to a "darkening with education", ie the higher education level leading to greater black racial self-identification (MARTELETO, 2012, p. 339).
• 25
  Gender Parity Index (GPI): indicator of differences in values obtained in women and men, the GPI of 1 indicates the existence of parity between the sexes, GPI, which is between 1 and 0 means a disparity in favor of boys, a GPI greater than 1 indicates a disparity in favor of girls (UNESCO, 2004, p. 386).
• 26
  The adoption of this new designation will be explained later.
• 27
  Adapting the reference interval for Brazilian values, there is parity for women in the ranges of 0.98 to 1.08 (2000) and 0.99 to 1.09 (2010); for blackness rate, parity is any value between 0.78 and 0.88 in 2000 and from 1.02 to 1.12 for 2010.
• 28
  Even with recent theoretical discussions that reflect the dichotomous view of sexual construction, the category sex is still useful and collected as a dichotomous variable formed by men and women. The discussions on how to research the different orientations and gender performances, alternatively to the sex category did not proceed.

Publication Dates

• Publication in this collection
  Oct-Dec 2015

History

• Received
  Apr 2015
• Accepted
  Aug 2015