Dissertation_Final Draft_LI

University of East Anglia

School of International Development

Race, Inequality & Fairness A MIXED-‐METHODS APPROACH EXAMINING PERCEPTIONS OF BRAZILIAN UNIVERSITY STUDENTS TOWARDS AFFIRMATIVE ACTION

Student Registration Number: 6256082

Supervisor: Professor Peter-‐Lloyd_Sherlock

A dissertation submitted towards the degree of BA at the University of East Anglia, January 2014.

i

Table of Contents

Abstract ............................................................................................................................... i

Acknowledgements ............................................................................................................. i

1.Introduction .................................................................................................................... 1

2.Summary ......................................................................................................................... 2

3.What do Race, Colour and Identity Mean in Brazil? .......................................................... 3

4.What is Inequality in Brazil? ............................................................................................. 4

5.What is Affirmative Action ............................................................................................... 5

6.Affirmative Action in Brazil: Key Features ........................................................................ 6

7.Racial Quotas in Brazilian Universities: The Rationale ...................................................... 9

8.Methodology ................................................................................................................. 11 8.1-‐Fieldwork Approach ............................................................................................................. 11 8.2-‐Approach Rationale ............................................................................................................. 12 8.3-‐Reflexivity, Positionality and Power Relations ..................................................................... 13 8.4-‐Analytical Methods .............................................................................................................. 13

9.Data Analysis and Findings……………………………………………………………………………………………15 9.1.Perceptions-‐1: Race, Colour and Identity: Problematic for AA in Brazil ........................ 15

9.1a-‐Self-‐Classification: A Simple Process? ................................................................................. 16 9.1b-‐Self-‐Classification: Effective Targeting of ‘Non-‐Whites’ ...................................................... 17 9.1c-‐A Preference for Whiteness? .............................................................................................. 19

9.2.Perceptions-‐2: Education’s Contribution to Racial Inequality in Brazil ......................... 21 9.2a-‐A Basic Education Problem ................................................................................................. 23

9.3.Perceptions-‐3: Fairness ............................................................................................... 24

10.Conclusion ................................................................................................................... 28

11.References ................................................................................................................... 31

Appendix-‐A1: Table_1-‐Quantitative Data .......................................................................... 37

Appendix-‐A2:Table_2-‐Quantified Qualitative Data ........................................................... 38

Appendix-‐B1:PRSOL*Race-‐Crosstabualtion ....................................................................... 39

Appendix-‐B2:Fairness*Race-‐Crosstabulation .................................................................... 40

Appendix-‐C1:Fairness*Race (Ordered-‐Probit) ................................................................... 41

Appendix-‐C2:Fairness*Race (Ordered-‐Probit) ................................................................... 41

Appendix-‐C3:Fairness*Race (Standard-‐Probit) .................................................................. 41

Appendix-‐C4:Fairness*Race (Standard-‐Probit) .................................................................. 42

Appendix-‐D:Qualitative Exerts Table-‐2 Categories ................... Error! Bookmark not defined.

ii

Appendix-‐E:Ethical Protocol .............................................................................................. 43

Appendix-‐F:Questionnaire, Information Sheet, Interview Consent Form (English) ............ 53

Appendix-‐G:Questionnaire, Information Sheet, Interview Consent Form (Portuguese) ..... 59

Appendix-‐H:Written Permission Examples ............................... Error! Bookmark not defined.

i

Abstract

This paper examines how students at the State University of Rio de Janeiro (UERJ) and the

Pontifical Catholic University of Rio de Janeiro (PUC-Rio) perceive race/colour, identity,

inequality and fairness in relation to race-based affirmative action (AA) within public

university admissions. Using a mixed-methods approach I demonstrate how respondent’s

perceptions of race/colour and identity are both complex and inconsistent, yet appear

influential in their conceptualisation of inequality and fairness; thus shaping their attitudes

towards the racial quotas policy (RQP). Findings include a negative correlation between

darker levels of self-classified pardo (brown/mixed-race) and perceived relative standard of

living (PRSOL) and a positive correlation between darker levels of self-classified race/colour

and perceived levels of fairness, robust to alternative discrete choice models; indicating that

while racial classification is ambiguous, respondent’s perceptions are identifiable along racial

lines.

Acknowledgements

I would like to thank Professor Peter-Lloyd Sherlock, my supervisor, for his wisdom and time

throughout the dissertation process. Thanks also to Dr Lucio Esposito and Dr Robert Grant

for their patient assistance regarding the data analysis conducted in this paper. Moreover,

gratitude goes to Brazilian friends who provided translation assistance and invaluable help in

the field. Finally, I thank staff and students at UERJ and PUC-Rio who have made this

research possible.

1

1. Introduction

More than a decade has passed since AA was enacted in Brazilian public university

admissions, yet public debates surrounding AA -predominantly racial quotas (RQs)- remain

strong and polarised. Conflicting perceptions towards racial classification, the dynamics by

which inequality is/should be measured and issues of fairness ostensibly drive disputes.

However, maintaining the political and social status quo are often cited as the genuine

motivations of RQ opponents (Bailey, 2004; Santos, 2006; Bailey & Peria, 2010). Conversely,

the RQP is viewed as a driver of perceptions; considering university students, Schwartzman

and Silva (2012) highlight that perceptions toward RQs are not only shaped by public

discourse but also everyday experiences of the policy “…generating different understandings

of race relations and socioeconomic inequalities…” (ibid:33). Cicalo (2012) notes these

everyday experiences may have paradoxical outcomes, heightening prejudice and racism yet

promote “…social encounters, solidarity and reciprocal knowledge between groups of

different colours and social backgrounds.” (ibid:259).

The distinctive division of opinion the RQP has created in Brazil, its direct impact upon

society’s perception of race/colour, identity and inequality and the indirect consequences that

have ensued, highlights the importance for further research upon the matter. This study

draws on new data, examining perceptions of 489 university students concerning different

aspects of the RQP. It investigates how students perceive race/colour and identity alongside

what factors influence these perceptions. Moreover, how students perceive Brazil’s racial

inequality, the contribution of education to these inequalities and whether RQs are an

effective/fair intervention in tackling inequalities of access to university education.

The way RQs are perceived is important, not only for their future trajectory in the Brazilian

education system but, ultimately, the country’s overall socioeconomic development. The

RQP directly affects current students and if their views do not align to policy formulators’,

support for, and the effectiveness of the policy, may remain abated. Therefore, one objective

of this primary research is to make a small contribution to some of the wider debates

surrounding RQs in university admissions, primarily in Brazil but also in other countries that

employ such policies.

2

2. Summary

Chapters 3-7 review the literature pertaining to Brazil’s experience of AA -chiefly RQs in

university admissions- providing background knowledge regarding the complexities and

ambiguities in the assessment of race/colour, identity and inequality in Brazil; followed by an

overview of what AA encompasses generally and finally the key features of, and rationale

behind the RQP. Chapter 8 describes my research approach, rationale and reflections upon

the research process alongside my data collection/analysis strategy. Chapter 9 presents my

data analysis and findings in three distinct sections of perceptions concerning topics

examined in the literature review, while in Chapter 10 I summarise and offer some

concluding remarks.

3

3. What do Race, Colour and Identity Mean in Brazil?

“I'm white. I believe that there is no brown race; we are either white or black. Also there is no

mixed-race. Actually, all are mixed races.” (Questionnaire Respondent (QR)-E98)

The composition of race, colour and identity in Brazil is complex; originating from the

miscegenation of three previously distinct races: Europeans, Africans and Indians, which

contributed to the construction of a uniquely coloured Brazilian population, culturally and

biologically merged, strengthened through its hybridism (Bailey, 2004). Thus, race became

equivocal with Brazilians ambivalently using endless terms to classify one another, primarily

according to skin tone and physical appearance as opposed to ancestry, which commonly

determines race in other countries (Reichman, 1999:7).1 This classification procedure also

features in Brazilian census data with longstanding race displaced by colour (Piza &

Rosemberg, 1999:37). The ambiguity regarding racial identity is dualistic; with one’s self-

perceived race/colour often differing from the way others perceive that same individual (ibid;

Schwartzman, 2008). Even self-classification can alter depending on the setting, context and

racial categories available (Sansone, 2003; Bailey, 2008). Furthermore, multiple social

factors can influence perceptions of race/colour including “…one’s education, wealth and

gender as well as the context of social interaction.” (Bailey & Telles, 2006:76) with the

concept that class, education and money whitens frequently debated (Telles, 2004).

This ambivalence can have important ramifications for Brazilian public policy debates, including those regarding RQs in university admissions whereby students self-classify their race/colour during the application process (see Bailey, 2008). This and other topical race-related elements have made the policy highly controversial, dividing opinion within the political sphere (Santos, 2006) Brazilian academia (Bailey & Peria, 2010) and society generally (Bailey, 2004).2 Chapter 9.1 examines this issue in greater detail.

1 Black in Latin America - Brazil: A Racial Paradise_(2011) Documentary regarding issues of race/identity. 2 See_Aljazeera - Racial Quotas in Brazil http://www.dailymotion.com/video/xr52qg_brazil-s-racial-quotas_news

4

4. What is Inequality in Brazil?

“I believe the biggest inequality is in terms of income, not social.” (QR-B20)

“Unfortunately racial inequality in Brazil is still very high.” (QR-A40)

Brazil is commonly known as one of the world’s most unequal societies. Inequality -as

represented by the Gini-coefficient for the distribution of household income per-capita- has

remained pervasively high since reliable data became available (Barros, et al., 2000:16).

After reaching a near historical and worldwide record of 0.63 in 1989 (World Bank, 2013) this

figure stagnated before steadily declining from 1998 onwards, standing at 54.7 in 2009,3 a

noteworthy improvement. However, globally, Brazil remains among the ten most unequal

countries (World Bank, 2013a) with the highest 10% of the population holding 43% of

national income and the lowest 20% just 3% (World Bank, 2013a). Although universally

employed, the Gini-coefficient and income variable may be viewed as a rudimentary

measurement of inequality as it is calculated at the aggregate level, not considering the

multifaceted dynamics of Brazilian inequality, including: regional (Azzoni, 2001), urban

versus rural (Fiess & Verner, 2003), gender elements (Garcia et al., 2009) the socio-political

sphere (Sherlock, 2009) or -arguably the most prominent combined determinants-

educational attainment and race (Silva & Hasenbalg, 1999; Marteleto, 2012). Commonly

attributed to Brazil’s well-documented history of slavery, irrefutable racial discrimination

(Telles, 2004:139-171) and regional dualism, the “…Brazilian socioeconomic structure (is)

identified as being largely divided along racial lines.” (ibid:137).

These interlinked components of inequality -more specifically unequal access to university

education by race- and how university students in Brazil perceive this issue, are examined in

Chapter 9.2. Education level is both a cause and effect of inequality, accounting for almost a

third of inequality in Brazil, with education distribution - in terms of quality and quantity- being

the most important driving factors (Bourguignon et al., 2008). Higher education has

historically been a contributing factor towards Brazilian social/racial inequality (Lima, 2011:4),

its influence having possibly changed in recent years due to growth in the private university

sector and, thus, enrolment rates.4 Growth that is claimed to have sustained inequity in terms

of educational quality and access by race (McCowan, 2007) with Lima (2011:17) noting racial

differences in educational attainment are greatest “…among those who have completed

higher education: 76.8% are whites.”5 Such developments are likely to have affected the way

debates regarding RQs in university admissions unfold, thus making this a topic of key

interest.

3 Latest available figure-http://data.worldbank.org/indicator/SI.POV.GINI,http://wdi.worldbank.org/table/2.9 4 Net-enrolment rates (ages,18-24) increased from 9% in 2002 (McCowan, 2007:5) to 13.2% in 2008 (Lima, 2011:7). 5 Lima uses data from PNAD (2006-2007-2008). Notably, recent data sets regarding disaggregated examination of educational attainment by race rarely exist, invariably focusing more on gender and socioeconomic indicators.

5

5. What is Affirmative Action

“The quotas system is an attempt to resolve a segregation problem of more than 200 years.”

(QR-D8)

AA can be framed in different ways, either referring to “…a set of practices undertaken by

employers, university admissions offices, and government agencies to go beyond non-

discrimination, with the goal of actively improving the economic status of minorities…”

(Holzer & Neumark, 2006:2).6 Or as “A policy or a program that seeks to redress past

discrimination through active measures to ensure equal opportunity, as in education and

employment.” 7 AA in university admissions originated in the US during the late 1960’s and

early 1970’s (ibid.). Subsequently multiple US states imposed bans upon enacting such

policies (Economist, 2013) and elsewhere the concept also remains highly controversial. In

South Africa policies were ratified to redress the nation’s experience of cultural apartheid and

in Malaysia true natives have been beneficiaries of AA for over 40 years (Economist, 2013a;

NY Times, 2010). Without any socioeconomic factors considered in the admissions process,

AA in South Africa is argued to primarily benefit the black middle class. In Malaysia, AA has

been enforced along ethnic lines, induced by violent clashes between the then economically

disadvantaged ethnic Malays and the country’s wealthier ethnic Chinese and Indian citizens.

Temporary AA measures in university admissions still remain today, viewed as a social

injustice by non-beneficiaries and ‘out-dated’ by 71% of surveyed Malaysians (ibid:2013a;

2010).

Theoretically, AA policies may appear progressive and necessary; however in practice their

justification, formulation and implementation can be complicated and provocative. 8 For

instance: who should/should not benefit from such a policy? Who should make this decision?

How is this decision taken? (Bailey, 2008) And, most pertinently, are they effective?

(Schwartzman, 2009) And are they fair? (Frias, 2012). Concerning Brazil, these questions

are of particular interest given the discussed complexities attached to racial classification,

identity and inequality.

6 Holzer & Neumark, (2006:3) took “minority” to refer to groups that are typically under-represented in universities and better jobs in the contemporary economy. 7 Taken from-http://www.thefreedictionary.com/affirmative+action 8 Wide Angle: Brazil in Black & White

6

6. Affirmative Action in Brazil: Key Features

“UERJ’s quotas system takes into account not only the applicant’s race, but also their

financial situation.” (QR-A61)

“The quotas regime isn’t just for black and indigenous people, it can be for sons of killed

police officers, fire-fighters, military servicemen and students from public schools.”

………………………………………………………………………………………….(Interviewee-A)

AA is a relatively modern concept in Brazil; momentum towards race-related policies was

initiated under President Cardoso (1995-2003) who created a political space for Afro-

Brazilian movements and networks of scholars to condemn racism (Htun, 2004). This

represented a new discourse at the state-level, which had historically peddled the notion of

racial democracy9 (ibid.); a term popularized originally by the work of leading academic

Gilberto Freyre (1956). Influenced by his mentor, anti-racialist anthropologist Franz Boas-

who claimed Brazilian society to be comprised of a Brazilian race, distinguishable only by

cultural and social differences- Freyre extended this notion arguing that Brazil’s racially fluid

population were free from racism (Telles, 2004:33). This concept encompassed a stance of

anti-racialism, discrediting the use of race as a parameter to comprehend socioeconomic

factors given the extent of miscegenation in Brazil (Bailey & Peria, 2010). A theory which

holds even today within certain public spheres who dispute the need for race-oriented

policies (ibid:595), despite a body of scholarship evidencing racial disparities in various

domains of Brazilian life (Twine, 1998:4). Whether Boas and Freyre’s notions remain

prevalent within the perceptions of current university students is a topic examined in Chapter

9.2.10

In addition to Cardoso’s efforts, the 2001 United Nations Conference on racism in Durban,

South Africa, is seen as pivotal in guiding Brazil to enact RQs within various governmental

spheres (Telles, 2004); including the public university vestibular entrance exam system,

whereby students choose their career paths upfront, register for the vestibular in that field

with course admittance determined by that individual score. This process involves thorough

preparation and has always been highly competitive with only 10% of applicants being

accepted at some of the top public/federal universities (Bailey & Peria, 2010). 11 The

introduction of quotas intensified competition in the admissions process, bringing into

question the aspects of meritocracy and fairness; topics discussed in Chapter 9.3.

9 Democracy here refers to the Spanish connotation of the term; brotherhood or fluid social relations, not a type of political institution (Bastos, 2001) as cited by Telles (2004:33) 10 See Bailey (2004) for discussion regarding racial attitudes in Brazil. 11 Cicalo (2008:71) notes that only the top scoring 168 non-quota applicants to UERJ’s law degree were admitted out of 3000 with successful quota student’s scores inbetween the 1000th-1800th position.

7

The first public university forced to employ RQs via a state legislature approved bill was

UERJ in 2001, establishing a quota of 40% for blacks, complimenting an earlier quota of 50%

for students coming from public schools (Schwartzman, 2009:228). Despite the presence of

racial prejudice/discrimination and inequality being acknowledged by the state, scholars and

Brazilian society generally (Bailey, 2002; Telles, 2004:75), the RQP ratification induced

conflicting perceptions. Datafolha national surveys (1995, 2006 2008) suggest there has

been growing “widespread support for affirmative action” (Schwartzman & Silva 2012:32,45);

doubling with regards to university educated Brazilians (Guimarães, 2007:10).12 In RJ state,

Bailey (2004:736-739) shows a majority of Brazilians (54.9%, n=1011) favour RQs; however,

considering those with a high education level, this figure drops to 30%.13 Within media,

academic and public spheres the policy was met with hostility, culminating in Rio state

legislature (2003) altering its original composition of RQs by reducing vacancies to just 20%

of the 45% available (Telles, 2004:74); all of which included socioeconomic criteria

(Schwartzman, 2009:229). In 2004 the University of Brasilia (UnB) became the country’s first

federal university to adopt RQs with 20% of vacancies reserved for black or negro14 students

(Francis & Pianto, 2012). Moreover, AA was integrated into the private sector with the

University for All Programme (ProUni), 15 funding higher education fees and providing

scholarships to socially disadvantaged undergraduate students at private universities

(McCowan, 2007:11); bolstered in 2005 by the federal government supplying tax incentives

to private institutions embracing the program (Bailey & Peria, 2010), including prestigious

PUC-Rio.16

Introducing the umbrella term ‘negro’ in 2003 has been noted as a political move by black

activists to stimulate a sense of black consciousness in Brazil whilst also creating solidarity

among the subjugated non-white population. Moreover, it aimed to heighten the significance

of ethnicity over colour among non-whites, attempting to eliminate the pardo category

(Reichmann, 1999:9-10; Schwartzman & Silva, 2012:40) as well as “…making more visible

the reality of racial inequality in Brazil and better enable its redress.” (Bailey & Telles,

2006:75). However, instead of enhancing racial clarity, Bailey (2008); Loveman et al. (2011)

and Francis and Pianto (2012) show that the negro term may have amplified confusion over

racial classification and induced unintended consequences with regards to AA targeting, thus

reducing the RQP’s effectiveness.

12 From 19.6% (Datafolha-national survey, 1995) to 42% (Datafolha-national survey, 2006) n=455 13 Bailey uses original Data from 1999. Respondents are asked “Do you believe the government should/should not set aside openings for blacks in public universities? 14 Negro, essentially translating as black, collapses the pardo (mixed-race/brown) and preto (black) categories into one. 15 See-http://prouniportal.mec.gov.br 16 See-http://www.puc-rio.br/ensinopesq/ccg/prouni/ for information regarding PUC-Rio’s ProUni programme.

8

In 2010, around half (49) of all federal and state universities employed quotas of some kind

benefitting 45,000 students, although this impressive figure represented just 11% of students

in public higher education (LAESER, 2012).17 In August 2012 the government legislated for

all 59 federal universities and 38 federal technical schools to reserve half of vacancies for

public school students by 2016 (Economist, 2013b), heightening the “…affirmative action

boom…” (Bailey, 2009:11) with familiar arguments opposing RQs regularly emerging,18

including those of meritocracy, bringing high calibre students to university and complications

over racial classification:

"This is bad for the future of Brazil, because the main objective of universities is research, is

to achieve quality." (BBC, 2013, Antonio Freitas, provost of the Getulio Vargas Foundation, a

prestigious private university)

“White skinned candidate is approved for racial quota”

(Headline, O’GLOBO, 2013)

17 As cited by Telles and Paixao (2013). 18 See ‘comments’ sections at-http://www.economist.com/blogs/americasview/2013/04/affirmative-action-brazil#comments_&_http://oglobo.globo.com/educacao/candidato-de-pele-branca-aprovado-por-cotas-raciais-na-1-fase-do-itamaraty-9908199

9

7. Racial Quotas in Brazilian Universities: The Rationale

“It’s what the blacks are owed, it’s the debt, because when the blacks were freed from

slavery, they had to go to the slums, run away, many became robbers, out of necessity, to

eat and until now they are stigmatised for that.” (Interviewee-S)

Basic public education provision in Brazil has improved in recent years (Marteleto, 2012)

although quality is still lacking with most secondary-school students not reaching desired

levels in key subjects such as Portuguese and Maths (Schwartzman, 2011:18). Low

performance is attributed to various factors resulting from underinvestment by the state and

“limitations and disadvantages related to their (student’s) origins that have accumulated over

the years.” (ibid:19).19 Public primary/secondary-school students are commonly lower-class

Brazilians whose parents cannot afford private schooling, a majority of which are non-white

(Hernandez, 2005:688). With reduced quality and quantity of schooling, these students are

highly underprepared to pass the vestibular examination to enter the contrastingly prestigious,

well-funded and tuition-free public universities. Moreover, they must compete with upper and

middle-class students -predominantly light-skinned or white- who are educated in excellent

private high schools (Cicalo, 2012:239); thus further diminishing these student’s chances of

entering public universities and with private universities typically unaffordable, tertiary

education is unattainable. These disparities led to Brazil being branded “…the world

champion in social injustice in higher education…” (Castro, 2001);20 a statement supported

by data (1997) showing just 2.2% of non-whites held a university degree compared to 9.6%

of whites (Cicalo, 2012:239.),21 despite, at the time, Brazil’s population of non-white (46%)

and whites (54%) being fairly even (Bailey & Peria, 2010).22 This underrepresentation has

had concomitant effects with elite media, political and economic circles within Brazil being

distinctly white dominated (Htun, 2004:63,74) with this ‘group dominance’ actively preserved

(Bailey, 2004).

The rationale behind RQs is posited as reparation for past/present wrongdoings against the

non-white population. A sense of fairness aligning to Rawls (1971) theory of justice, whereby

inequality is largely attested to luck or extrinsic factors which are beyond the individual’s

control, or variables of ‘circumstance’, as theorised by Roemer (1998), who proposes that

circumstances which produce systematic disparities in individual outcomes are unfair and

stem from scarce opportunities as opposed to effort (World Bank, 2004:6). Proponents of

19 Hernadez (2005) cites many studies regarding this topic. 20 As cited by Telles (2004:125) 21 See IBGE data for 2007 available at-http://noticias.terra.com.br/educacao/interna/0,,OI3205059-EI8266,00-IBGE+n+de+brancos+com+diploma+e+vezes+maior+que+o+de+negros.html 22 Data available at-http://www.ibge.gov.br/home/estatistica/populacao/censo2000/tendencias_demograficas/tendencias.pdf

10

RQs appear to believe the same, arguing that without them, Brazil “…will continue

reproducing the cycle of deep racial inequality that has been the mark of the entire history of

the republic to the present.” (Pro-quota, 2006:8).23

23 As cited by Bailey and Peria (2010:596).Pro-quota (2006) refers to a manifesto presented by academics to Brazil’s national congress. See-http://www.observa.ifcs.ufrj.br/manifesto/index.htm

11

8. Methodology

8.1-‐Fieldwork Approach

This research uses a mixed-methods approach.24 Data collection includes a large sample of

questionnaire respondents (n=489) allowing for robust quantitative statistical data analysis,

complemented and extended by interviews which provide a qualitative richness to the

complexities apparent within the research topic; otherwise remaining concealed solely using

a quantitative approach. A quantitative questionnaire method can sometimes be restrictive,

especially with regards to perceptions; constrained by time, respondents must formulate their

opinions impulsively. Interviews give participants more time to contemplate the topic,

formulate and elaborate upon their opinions. Moreover, questionnaire responses are often

impersonal with quantitative data unable to either convey emotion or provide depth of

reasoning that may indicate signposts as to what influences one’s perception, subtleties

better obtained within qualitative interviews.

The primary data collection was carried out during June-July 2013 via an observational

research design. In supervised classroom settings, anonymous questionnaires were

administered to 489 students (predominantly undergraduates) from different disciplines in

two separate universities located in Rio-de-Janeiro (RJ): PUC-Rio, (265 respondents) and

UERJ (224 respondents). The distribution of disciplines within these two universities along

with a description of all variables used in my analysis and the relevant descriptive statistics

are viewable in Table 1.25

Moreover, ten semi-informal interviews were conducted, totalling around six hours of

recorded audio. As a closing remark during classroom visits, the interview process was

explained, advising any willing participants to contact either my research assistant or me

directly via the contact details left upon the board. This method led to securing four

interviewees while three students immediately volunteered within the classroom setting. The

remaining interviewees were opportunistically approached during conversation; they had

expressed an interest in and were highly knowledgeable on my research topic; one, a

postgraduate law student, another, a professor at both universities and the third, a

representative from Proninciar, an internal quota student support service in UERJ.26

After negotiations with the head of each respective department, written permission was

obtained to conduct the questionnaire,27 completed during either the first or last fifteen

minutes of ordinary lectures at each university, as pre-arranged with the lecturer of each 24See appendix-E for a comprehensive ethical protocol. 25 See appendix-A1 26 Visit-http://www.caiac.uerj.br/cotas.html for details regarding Proniciar’s service. 27 See appendix-H for written permission examples.

12

respective class. All respondents received a Brazilian Portuguese version of documentation,

originally constructed in English.28 The translation was piloted and reviewed with Brazilians,

one notable alteration to the questionnaire related to self-classified race/colour; some of the

initial English categories directly translated were deemed problematic and, thus, were altered

to align with traditional Brazilian census categories.29 30 Students were briefly introduced to

the research/questionnaire content prior to distribution, at which point it was emphasised that

the questionnaire was being administered to ascertain their perceptions and that no right or

wrong answers existed; also explained to interviewees.

The co-ordination process was logistically challenging, partly due to the on-going

demonstrations occurring during the research period with each university closing down on

scheduled protest dates (see Economist 2013c). Conversely, the demonstrations are seen

as advantageous to this research. Underlying desires voiced during these mass protests

included the demand for increased government investment to initiate educational reform and

ensure equality and equity of opportunities for all; thus, my research topics were embedded

in the forefront of respondent’s minds. It is believed this heightened awareness made

respondents more receptive to the research topic, enhanced their willingness to complete the

questionnaire and/or participate in interviews and augmented openness, adding flavour and

sincerity to their responses.

The questionnaire collected data regarding respondent’s perceptions of race/colour

(including self-classification), their standard of living (SOL) compared to others, the university

system in relation to inequality, the fairness and effectiveness of RQs, their course of study

and other demographics. Moreover, strategically placed comment boxes were provided for

three questions enabling respondents to voluntarily elaborate upon their initial quantitative

answers. These produced a vast quantity of valuable qualitative data with many key themes

arising, as illustrated in Table 2,31 complemented further by the fruitful qualitative data

obtained from interviewees.

8.2-‐Approach Rationale

Obtaining data from students across differing disciplines was important; leading private and

public Brazilian education is often prided upon prestigious vocational degrees such as law,

medicine and engineering, commonly exclusive to the elite classes. Variation in the sample

28 See Appendix-F/G for English/Portuguese versions of the Questionnaire, Interview Information Sheet and Interview Consent Form 29 Standard census options: branco (white), preto (black), amarela (yellow/Asian) and indígena (Indigenous). 30 Particularly confusing were English options ‘black’ and ‘mixed-race’, initially translated as ‘negro’/‘mestiço‘. The politicised term ‘negro’ has already been discussed, the term ‘mestiço’ (when solely used) has ethnic connotations attached and therefore was replaced by ‘pardo’ (brown/mixed-race). See Francis and Pianto (2012:54) for an informative explanation. 31 See appendix-A2. Appendix-D contains qualitative exerts exemplifying these themes.

13

ensures that data were collected from numerous types of students, with regards to their

choice of course and differing demographic, enabling interesting comparative analysis.

Similarly, the above applies to selecting two universities as opposed to just one. While both

institutions are highly regarded, they were specifically chosen for their key differences,

namely that PUC-Rio is a private institution with a ProUni social quotas (SQs) policy in place

but no RQP; and UERJ, a public university and the first to implement RQs in Brazil. These

universities are being treated as individual case studies for comparative purposes and the

collected data are not seen as a general representation of Brazilian university student

perceptions regarding RQs.

8.3-‐Reflexivity, Positionality and Power Relations

“My other professors all look like you, white with blue eyes. One looks just like you, he has a

Masters, is very intelligent and has gringo characteristics.” (Interviewee-B)

Throughout the fieldwork it was necessary to remain aware of issues regarding reflexivity,

positionality and power relations (see Sultana, 2007); remaining sensitive to aspects of race,

gender and educational stature, given my position as a white British male, representing an

English university, conducting research about a policy regarding racial inequality in Brazilian

tertiary education. It is acknowledged that my origins and imperfect Portuguese restrict my

understanding of local contexts, may have affected the way in which some participants

interacted with me and/or influenced my interpretation of the collected qualitative data. For

instance, it was difficult to remain wholly objective and refrain from letting my own perception

interfere with my line of questioning and interpretation of interviewee responses when they

opposed RQs. Moreover interviewees often wanted to know where I positioned myself in the

debate; implicitly assuming our positions were allied and thus felt comfortable to speak freely.

Furthermore, on several occasions in both negotiations and interviews, reference was made

to my nationality and blonde-haired blue-eyed appearance in the context of social

ranking/mobility and opportunities, often leaving me without a response other than an

embarrassed “I understand”. Conversely, my exotic status may have been beneficial,

apparent in my class addresses where I noticed my exoticness and sound Portuguese

appeared to endear me to my audience and, almost certainly, positively impacted upon their

willingness to participate.

8.4-‐Analytical Methods

Results and analysis within this paper refer to topics that organically emerged as key themes

during data examination, questions focusing upon linkages between the university system

and inequality, self-perceived race/colour, PRSOL and concepts of fairness and

effectiveness. Quantitative data has been utilised using software to enable statistical analysis;

14

qualitative data obtained in questionnaires and interviews has been translated, quantified

(questionnaire data only) often transcribed32 and objectively analysed. Moreover, literature

and previous studies from sources pertaining to: Economics; Sociology; Race/Ethnicity, Latin

American History and Psychology have been utilised, enhancing insight to, and

substantiating the findings of, this research. Although a vast theoretical and empirical body of

literature concerning AA exists, recent primary data regarding perceptions is lacking, while

the RQP and affiliated debates have evolved. Moreover, studies often focus on either

qualitative or quantitative data; I provide a unique approach to data collection on this scale.

Furthermore, this study differs from others regarding perceptions, which tend to have

focused solely on public university students, where perceptions are heavily influenced by

their everyday encounters with the RQP (Schwartzman & Silva, 2012).

32 Reviewed by Brazilians to ensure accuracy.

15

9. Data Analysis and Findings

9.1.Perceptions-‐1: Race, Colour and Identity: Problematic for AA in Brazil

This section examines respondents’ perceptions of race/colour and identity, how these

perceptions interlink with the wider context of the subject in Brazil and how they may justify

RQs as an effective way of tackling racial inequality. Quantitative analysis will focus on data

generated by Q3, Q4 and Q14, as outlined in Table 1; further supported by qualitative data

collected in interviews and by Q4, Q13 and Q15, outlined in Sections 1 and 2 of Table 2.

Some direct quotation from qualitative data is provided when it is believed to add significant

analytical value.

Of particular interest -given the complexities described in Chapter 3- were questionnaire

responses to Q3 regarding self-classified race/colour. It is acknowledged that using self-

classification of race/colour as an analytical variable is potentially problematic, given that no

method of validation regarding respondent’s sincerity was employed. Indeed, adversaries to

RQs argue many students commit fraud, self-classifying their race/colour dishonestly in order

to obtain a RQ (Schwartzman, 2009:222), a view illustrated in Table 2 with 12.5% (n=48) of

respondents who commented in Q13 or Q15 mentioning ‘deceit’. Previously third-party

classification was employed to decide upon RQ admissions using applicant photographs;

however, the subjectiveness of racial classification also makes this method controversial.

Bailey (2008) shows that others could exclude would-be beneficiaries, highlighted by a well-

documented case in 2007 at UnB where only one of two identical twin brothers applying for a

RQ were admitted. Thus, self-classification is a key criterion embedded within the RQP

embracing the notion “…that people are racially what they say they are…” (Bailey & Telles,

2006:75)33 and, therefore, given its relevance in the RQP debate, the self-classification

model is deemed appropriate.

Figure-1 shows a breakdown of respondent’s (n=488) racial categories.

Figure-1

Race 0=branco

1=pardo (branco)

2=pardo (mestiço)

3=pardo (negro)

4=preto 5=amarela 6=indígena 7=other

% 57.2

(59.4) 15.8

(16.4) 13.9

(14.5) 4.3

(4.4) 5.1

(5.3)

1.45

0.8

1.45

Although these percentages are not nationally representative (IBGE, 2013),34 they do more

accurately characterise the racial composition of RJ state35 (IBGE, 2013).36 The additional

33 Cited from Perlmann and Waters (2002). 34 See-http://seriesestatisticas.ibge.gov.br/series.aspx?vcodigo=PD336 &_http://seriesestatisticas.ibge.gov.br/series.aspx?vcodigo=PD336 (2011)

16

breakdown is derivative of the first, illustrating the percentage of respondents self-classifying

their race/colour within Categories 0-4 (the branco/pardo/preto continuum); embodying 96.3%

(n=470) of all responses. As with other studies (Bailey & Telles, 2006; Bailey, 2008; Loveman

et al., 2011) this research focuses on this continuum and the branco/preto dichotomy, given

that the majority of controversy regarding self-classification within the admissions process

refers to these colour categories. However, also noteworthy is Category 7 (other), offered to

participants given that past research has exposed that many Brazilians prefer to describe their

race/colour outside the standard census categories (Sansone, 2003); however, just seven

respondents opted for this category.

9.1a-‐Self-‐Classification: A Simple Process?

Table 1 shows that 40.8% (n=186) of respondents answered ‘no’ to Q4, stating that the racial

categories offered in Q3 were not adequate parameters for distinguishing their race/colour.

Thus, the low usage of ‘other’ is surprising. There is no reason to believe this low usage was

due to respondents conveniently opting for conventional census categories, given the

popularity of non-census options (1-3), examined later in this section.

The adjoining comments to Q4 highlight why so many respondents answered ‘no’:

“I find it hard to distinguish, since each person is evaluated. Your category can be different

dependent on who's seeing.” (QR-A29, self-identified as Parda (Negra))

“It’s very difficult to specify what colour/race you consider yourself. Everyone might think that

their skin colour is different from what it really is. For example, I'm white, but I have some

relatives who are black and some who are Indians. I can’t consider myself Caucasian. What

am I?” (QR-E55, self-identified as Branca)

“It isn’t adequate as it is a subjective criteria, in which already existing prejudice can

influence the self-declaration of race and colour.” (QR-B83, self-identified as Branca)

Of those who answered ‘no’ to Q4, 30.1% (n=56) commented upon the ‘difficulty’ and

‘subjectivity’ of racial classification, both in terms of others and oneself; adding the

interrelated comments regarding ‘miscegenation’ and ‘no.racial.divsion’ 37 38 (examined in

Chapter 9.2), increases this figure to 76.3% (n=142). These figures indicate that respondents 35 Percentages if a post-hoc re-designation of categorical boundaries (2,3) and (4,5) are collapsed into pardo and negro/preto respectively with branco, amarela and indígena unchanged 36 See-http://seriesestatisticas.ibge.gov.br/series.aspx?vcodigo=PD336 &http://seriesestatisticas.ibge.gov.br/series.aspx?vcodigo=PD336 Referring to all censuses excluding 2011. A noteworthy observation is the decrease and increase in the relative and absolute number of brancos and pardos respectively in RJ state between 2003 (year the RQP was fully enacted) and 2011. 37 Appendix-D contains additional exerts exemplifying these categories. 38 See appendix-A2 (Section-1). In total, of the 263 respondents commenting on Q4, 77.6% made comments referring to these four categories.

17

commonly perceive racial classification and identity as problematic in everyday life and, thus,

bring into question the appropriateness of the RQP self-classification model.

9.1b-‐Self-‐Classification: Effective Targeting of ‘Non-‐Whites’

Alongside the standardised Brazilian census options, three additional categories of pardo

were offered: pardo (branco), pardo (mestiço) and pardo (negro), enriching the colour

continuum by which respondents could self-classify. Their utilisation appears to demonstrate

their usefulness as parameters for self-classification, as emphasised by Interviewee-R, an

UERJ student who voluntarily identified himself as a RQ beneficiary:

“I think the categories come close to reality, for example, UERJ’s questionnaire only has 5

categories: white, black, brown, indigenous and yellow; you offering different options

simplifies it for me, but it’s still very difficult to pick one colour/race. (_) When I completed the

vestibular I said I was black not because of the quota but because of what I resemble best.

Now, I put down pardo (mestiço), why? Because I have German heritage, Portuguese, Black,

Indian, practically everything, so it’s very difficult to classify myself with one option, just like

many other Brazilians, they can’t do it.”

These additional categories were identified as alternative options to those used by previous

studies assessing the re-classification of pardos (Bailey, 2008; Loveman et al., 2011). Bailey

found that 44% of individuals self-identifying as pardo/mixed-race within census (ternary

categorisation) and open-ended classification, shifted to whiteness when constrained to a

white/black dichotomy. Moreover, Rosemborg (2004) -who conducted an AA oriented survey

amongst candidates for a Ford Foundation fellowship- also found that approximately half of

pardos39 reacted similarly when asked to self-classify as negro for beneficiary inclusion.

This research generates a similar finding; the data outlined in Table 1 indicates that 46.4%

(n=77) of respondents self-classifying as a type of pardo would likely opt for whiteness

(branco) if constrained to a dichotomous white/black classification model. However, this

figure could potentially increase to 87.3% (n=145) if all those who chose pardo (mestiço)

opted for whiteness. Moreover, these data suggest that 12.7% (n=21) of pardos would likely

opt for blackness (negro), potentially increasing to 53.7% (n=89) if joined by those who chose

pardo (mestiço). Respondents such as Interviewee-R -self-classifying as pardo (mestiço)-

who could opt for either, may represent a key target group for AA activists as they attempt to

attract non-white Brazilians to affiliate with the negro movement. Therefore -in order to

substantiate the actual outcome- inclusion of an additional branco/preto and/or branco/negro

dichotomy question would have been preferable. In any case, considering that just 12.7% of

self-classified pardos chose pardo (negro), it may be unwise for AA administrators to

39 n=304

18

continue under the premise that negro incorporates both the pardo and preto census

categories (Bailey, 2008:596) an idea reinforced by Telles (2004:86-87) who claims “‘Negro

in the popular system, like preto, refers only to those at the darkest end of the colour

continuum” and Schwartzman (2008:5) who found that many UERJ students interviewed in

her research “…did not understand the label negro as including pardos.”

The ‘whitening effect’ described above is potentially problematic. A significant number of

socioeconomically disadvantaged mixed-race Brazilians may demonstrate reluctance

towards affiliating themselves to the negro category even when potential benefits are at

stake (See Rosemborg, 2004); potentially resulting in many would-be beneficiaries suffering

“…exclusion from racial quotas…” (Bailey 2008:604). Thus impacting negatively upon the

overall effectiveness of RQs in tackling racial inequality. Conversely, Interviewee-R’s above

comments suggest that RQs have caused students to self-classify ‘reflexively’ “…according

to a conscious assessment of their place within the policy and within the structure of Brazilian

society.” (Schwartzman, 2009:247), a potentially encouraging sign for the negro movement

as it attempts to stimulate a sense of black consciousness and create solidarity among the

subjugated non-white population. Moreover, Loveman et.al (2011:15), identify that the

‘whitening effect’ helps to distinguish pardo’s socioeconomic status as those re-classifying as

white in a dichotomous model “…have, on average, higher incomes than those who identify

as black.” The data obtained by Q14 may support this finding.

The relationship between self-classified race/colour (as measured by the pardo 1-3 scale in

Q3) and PRSOL in comparison to other Brazilian families (as measured by the PRSOL 1-5

scale in Q14) was investigated using Spearman’s correlation coefficient. Preliminary

analyses were performed to ensure no violation of the assumptions of normality and linearity.

There was a small but significant negative correlation between the two variables (Figure-2),40

with darker levels of self-classified race/colour within pardo categories associated with lower

levels of PRSOL.

40 See Appendix-B1 for full results. 41 branco mean =4.02

Figure-2

PRSOL in Comparison to Other Brazilian Families

Correlations RACE PRSOL

Race Mean N

RACE

Correlation Coefficient

1.000 -.223**

pardo (branco)41 3.6711 76 Sig. (2tailed) . .004 pardo (mesitco) 3.4000 65

pardo (negro) 3.1500 20 N 166 161

**.Correlation significant at 1% (2-tailed).

19

Interpretation of these data should be made cautiously, given the relatively small sample

within each category, especially for pardo (negro). Moreover, here it is assumed RSOL

represents income; however, it is acknowledged that SOL could refer to multiple aspects of

one's life. It may be that attending reputable universities influences respondents to perceive

their own SOL as ‘better’ or ‘very much better’ (options 4 and 5 in Q14) in comparison to

others, with their present and/or expected social status/mobility after graduating affecting

their responses. Therefore perceptions of SOL (much like race/colour) could be dependent

upon to which segment(s) of society an individual belongs (Telles, 2004), as illustrated by

Interviewee-R who explains his reasons for choosing blackness during the vestibular process:

“I think I chose it for historic reasons rather than because I classify myself a certain colour,

because I identify more with black than white; I think more like a black than a white. ( )

Judging by the history of the country, the blacks were denied and excluded from society and

therefore I prefer to consider myself more black than white.”

In sum, although causality is not assumed, in concurrence with findings of Loveman, et.al

(2011), it is reasonable to suggest that pardos opting for whiteness or blackness may be

influenced by their perceived/actual SOL; a trend that may better enable the RQP to

effectively target the most socioeconomically disadvantaged non-white Brazilians.

9.1c-‐A Preference for Whiteness?

A wide body of historical and ethnographic scholarship documents the “…the ideological

privileging of whiteness.” (Loveman et al., 2011:11). Therefore, the preference for whitening

is unsurprising; Illustrated by comments made by Interviewee-T, a self-identified pardo

(mestiço) and beneficiary of the Pro-Uni SQ system at PUC-Rio:

“The colour people say they are is more influenced by how it’s perceived in society and the

status it will bring. For example, if a person says, "You’re brown" you'll think, "Oh, you’re

saying that because you think I’m black, but I also have a white part, a good thing in me."

Because here being black is a bad thing. After all, the blacks are the slaves, the poor, the

slum residents, the criminals.”

Interviewee-T’s comments align to the notion of ‘internalised racism’ as posited by Donna

Goldstein (2003:128) who conducted an ethnographic study of favela communities in Rio.

She extends the concept beyond the issue of race, attesting the condemned self-attitudes of

the poor black community toward a set of ‘internalised beliefs’. Interviewee-S, a self-identified

20

preto and another beneficiary of the Pro-Uni SQ system at PUC-Rio, describes these beliefs:

“Even though the government opened the quotas, the majority think they won’t be able to get in…I used to teach in the downtown, giving lessons in various neighbourhoods of Rio. The students didn’t believe in themselves. Many students have low self-esteem and that’s why they don’t pass or even take the exam to get into the university; they think, “I’m black, I’m poor, I’m not going to make it.”

With Interviewee-T’s comments in mind, Telles (2004:96) raises an important question

“Whitening in relation to what?” stating that “…since there is no objective way to classify race,

then one cannot whiten in relation to what one’s race ‘is’.” He provides three possible

interpretations, linked to social mobility, appearances signifying wealth and money, which

“…whitens a person in relation to how one identifies oneself or vice versa.” (ibid.) Further

comments, made by Interviewee-T align to his theory:

“Black people want to be white, want to have the level of white people, in fact, rich people. No one in the favela communities is studying to go back there and change it. People are studying in order to live in Leblon, Barra, etc,42 to stop being black and having black habits and become white, with white habits. The blacker you are the less money you have, the whiter you are the richer you are.”

Interviewee-T’s comments regarding her studies reflect theories regarding the correlation

between educational status and racial classification in Brazil. Some have claimed that higher

educational status, in similar fashion to social status generally, may influence darker

individuals to opt for whiteness (Harris, 1964; Wade, 1997). Conversely, the negro

movement is often regarded as one for the middle-class (Bacelar, 1999) with the negro term

resonating strongly among university-educated non-white Brazilians (Hanchard, 1994;

Schwartzman, 1999) who are more exposed to the movement’s discourse and/or have a

greater awareness of racial discrimination (Sansone, 1995). 43 Thus it is claimed that

education and its concomitant benefits may, in fact, darken (Telles, 2004; Marteleto, 2012); a

theory supported by comments made by Interviewee-S:

“I was against racial quotas in the beginning because I listened to the arguments of the white and rich people and they influenced my opinion. However, I went to a 3 day conference in Minas, with a very well-renowned professor from South Africa talking about quotas and it got black people talking about the policy and it made me and others realise “wow, this is really cool” at that point, I saw myself in the situation. When I realised that the people from my race had this debt, I realised that it needed to be this way.”

This section has illustrated the ambivalence and vacillation demonstrated by Brazilians towards race/colour and identity, highlighting the complexities that make racial classification 42 Leblon and Barra are two of RJ’s most affluent areas. 43 References in this paragraph cited by Bailey and Telles (2006:79-80)

21

of both others and oneself highly difficult and controversial; complexities that have made/continue to make the effective targeting of potential RQ beneficiaries highly problematic for AA administrators.

9.2.Perceptions-‐2: Education’s Contribution to Racial Inequality in Brazil

This section focuses upon respondent’s perceptions of the Brazilian education system,

specifically those regarding its contribution to racial inequality. Quantitative analysis will focus

on data generated by Q7, Q8 and Q10, as outlined in Table 1, further supported by qualitative

data collected in interviews and by Q4, Q13 and Q15, as outlined in Table 2.

Table 1 illustrates that of all those who answered Q7 (n=479) 62% perceive that, prior to the

RQP’s introduction, the university system contributed to racial inequality. Surprisingly, less

(55.9%) perceive the RQP to have mitigated the university system’s contribution to racial

inequality. 15 students -5% of those who chose ‘produced more’ in Q7- answered ‘increased’

in Q8. This indicates some students who acknowledge the university system contributed to

racial inequality historically perceive the introduction of RQs to have amplified racial

inequality. More noteworthy, these data highlight that 44.1% (n=212) 44 of respondents perceive the policy is not achieving its primary objective.

Section 2 of Table 245 shows that of 383 respondents commenting on Q13 and Q15, 90.1%

made a total of 544 comments referring to the categories shown in Figure-3.

44 Options 2/3 combined in Q8 45 See appendix-A2 for full version.

Figure-3 Categories Number of comments

Q13. Q15. A Problems in Basic Education/Investment needed 150 19 B Just SQs 94 8 C Palliative Measure 37 13 D Deceit 36 12 E Temporary Measure 24 5 F RQs and SQs 22 G Race/Subordinate/ Capacity/Prejudice 19 2 H Racism/Discrimination 18 2 I Meritocracy 16 2 J Race does not define Capacity 15 K Racial Quotas with Basic Education 10 2 L Capacity Problems 8 3 M Increases Prejudice 8 3 N Racialisation 7 1 O More Social than Racial 7 1

22

The qualitative data is useful to ascertain why 44.1% of respondents may perceive the RQP

to have augmented (n=109) or had no effect (n=103) upon racial inequality. Of those

commenting on Q13 and Q15, 23.6% (n=50) made reference to categories G, H, J, M and N,

including:

“The intention of the racial quotas system is to diminish racial inequality and prejudice, but a

racial quota is a prejudice in itself. In my opinion, this type of attitude detracts the afro-

descendant race, treating them as an inferior race in comparison with Caucasians. Black

people are as capable as White people. We are all the same.” (QR-A45)

“I believe that only thinking about the racial issue reaffirms and sustains racial prejudice. We

should think, therefore, about socioeconomic quotas.” (QR-D78)

“It’s common to see white people with a much worse economic situation than many people

with black skin. Skin colour is no indication of the need for assistance. Now that’s

racism.”..(QR-A36)

In total, 75 respondents made comments pertaining to categories G, H, J, M and N. These

comments are partially based upon a theoretical concept of equality in terms of believing all

to be equal and, therefore, treating all as equal, as opposed to objective indicators of

inequality such as income distribution, social mobility and exclusion from public services.

Moreover, they signify that some respondents perceive RQs as generating prejudice and

racism within Brazilian society, aligning to findings of a Datafolha survey (2006) showing that

48% of respondents46 agree that "to reserve quotas for Blacks at universities is humiliating

for them" and 55% consent that “to reserve quotas for Blacks at universities can generate

acts of racism" (Guimarães, 2007:11-12). These perceptions also pertain to the theory of

racial democracy, as described in Chapter 6; illustrating class-centred justifications (see

Schwartzman & Silva 2012:40) which support the concept that while Brazilians acknowledge

racial prejudice/discrimination exists, they do not perceive it to define the hardships

experienced by non-whites and instead perceive them to result from socioeconomic

factors/status (Santos, 2006). Moreover, the anti-racialism aspect of racial democracy is

illustrated by responses to Q4 with 46.8% of all respondents who commented (n=263)

referring to aspects of ‘miscegenation’ and ‘no racial division’ writing comments such as:

“The race categories are failed because Brazil is one of the most racially mixed countries in

the world. Including me, as even being white I have black members in my family.”o

……………………..…………………………………………... (QR-C15, self-identified as Branca)

46 n=unknown.

23

“Colour/Race is a concept that is hard to be applied in Brazil because our country went

through an intense miscegenation process. It is impossible to define someone’s ethnicity. I

have Portuguese, African and Indigenous relatives. My great-grandparents are Portuguese

and Tupi. My grandfather is black, as is all his family.”oooooooooooooooooooooooooooo

0………………………………………………………...(QR-B43, self-identified as Parda (Branca))

“My mum is white, her mother is Indian and her father is white. My father and his mother are

black. His father is white. So, I consider myself “pardo” (mestiço).” (QR-B52)

The commonness of these race-centred responses (see Schwartzman & Silva, 2012),

highlight how deeply internalized the notion of anti-racialism remains within Brazilian society

(Twine, 1998:8). Combined, the perceptions above illustrate how perceptions of race/colour

and identity subsequently affect student’s perceptions of inequality in Brazil. Whilst

addressing causality is beyond the scope of this paper -of the 38% (n=182) refuting the idea

that the university system historically contributed to racial inequality in Q7- many

respondents may be influenced by the concepts of racial democracy; including the lack of

recognition of race as a meaningful parameter to comprehend the socioeconomic dynamics

of Brazilian society.

9.2a-‐A Basic Education Problem

44.2% (n=169) of those respondents who commented on Q13 and Q15 referred to Category

A, stating that policy should focus upon augmenting quality/investment in the Brazilian basic

education system, which they perceive as the primary/sole cause of racial inequalities of

access to university education,47 thus helping to partially explain why 51.5% (n=87) of these

respondents perceive RQs in universities as ineffective in tackling the problem (options 1 and

2 combined in Q10).48 Moreover, 37.8% (n=64) of those referring to Category A, also referred

to Categories C, E and K, perceiving the policy as a palliative measure and/or one that

should only be a temporary initiative with structural reform in basic education taking priority.

“I believe the quotas system is extremely fair, however, I think it is an emergency measure

and other measures such as the improvement of basic education, should be taken.”

O…………………………………………………………………………......... ……………..(QR-A51)

47 13.6% (n=23) of these answered produced less or had no effect to Q7, suggesting they perceive basic education as the sole ‘educational’ cause of racial inequality. 48 Just 15.4% (n=26) of these respondents perceived RQs as effective (options 4/5 combined in Q10). In total 49.1% (n=237) deemed RQs as ineffective and just 18.2% (n=88) as effective. Although not explicitly stated upon the questionnaire, respondents were informed of what categories 2-4 represented in Q10/1112 during the briefing.

24

This was also emphasised within several interviews:

“Quotas are a palliative measure. It will not solve the issue of inequality. The problem is that

Brazil enjoys ‘painkillers’; it's just an attempt to make up for the problem. No one is really

interested in solving the problem, only to hide it; pretending it does not exist, as though

everything is fine. So, in the case of quotas, you reduce the level of education required for

people to enter; instead of improving their education overall.” (Interviewee-T)

In total, 23.8% (n=91) of respondents who commented referred to Categories C, E and K

suggesting that although RQs are perceived as necessary, they are not considered a

panacea to reduce racial inequalities of access to university education, due to fundamental

problems within the country’s basic education system; thus helping to explain why 14

individuals who chose ‘produced more’ in Q7 answered ‘had no effect’ in Q8.

9.3.Perceptions-‐3: Fairness

What defines and encompasses fairness is a broad topic of discussion, one beyond the

scope of the analysis presented in this paper. As noted by the World Bank (2004:5) “The

concept of social justice is inherently normative, which means that departing from different

views about what constitutes fairness could very well lead to radically different perceptions of

whether the Brazilian society, unequal as it is, is or is not fair.”

This section briefly examines respondent’s perceptions of fairness in relation to the RQP as

an intervention to tackle racial inequalities of access to Brazilian university education and

what may influence those perceptions. Quantitative analysis will focus on data generated by

Q11 and Q13, as outlined in Table 1; further supported by qualitative data collected by Q13

and Q15, as outlined in Table 2.

Table 1 shows that 49.1% (n=237) of respondents perceive RQs in university admissions as

extremely unfair (20.6%) or unfair (25.5%)49 and just 20.7% (n=100) as fair (12.7%) or

extremely fair (8%)50 These results suggest that RQs have less support than previously

thought (Bailey 2004),51 also appearing to contradict the idea that support for AA amongst

Brazilians with a higher education level is growing (Guimarães, 2007). 52 Whilst

acknowledging sampling and questions differ between Bailey’s study (RJ State/regarding

approval) and this research (two RJ universities/regarding fairness), their similarity in

respondent profile and sentiment make the aforementioned comparisons reasonable,

49 (Options 1/2 in Q11) 50 (Options 4/5 in Q11). 51 Bailey found 30% of respondents approved of RQs. See footnote 13. 52 It is acknowledged Guimarães’ findings were generated by a national survey. See footnote 12.

25

supported by data collected in Q11. Of those who answered ‘1 or 2’ in Q11, 49.8% (n=118)53

marked ‘abolished’ in Q13 while just 2.5% (n=6)54 marked ‘continue in current format’;55

whereas for those who answered ‘4 or 5’ in Q11 (n=88) these figures were 2.1% (n=5) and

34.3% (n=39) respectively. These figures align with intuitive thought; the aspect of fairness

appears to be a strong predictor of (lack of) support for RQs. In order to substantiate this

claim, inclusion of a dichotomised question regarding approval would have been preferable.

In any case, examining and understanding what Brazilians constitute as fair and what

contravenes this fairness appears very relevant in the RQP debate; therefore, further

research upon the matter is desirable.

“You are seeing justice but people are with more prejudice with these black people entering

university because they’re, well it's a survey, I didn't take part, but it's a common consensus

here that they’re stealing the university places…I don't think they’re stealing, I think they’re

being provided wrongly because, as I said, it's not a racial problem, it's an economic

problem.” (Interviewee-M, a self-identified branco student at PUC)

The above exemplifies how perceptions of race/colour and identity and subsequent

perceptions of inequality, may be influential factors as to whether respondents perceive RQs

as (un)fair. These include the discussed concepts of racial democracy and anti-racialism; for

Brazilians not seeing differences in race/colour negates the possibility of them determining

inequality as a racial problem and, therefore, racial inequality need not be actively and

unfairly redressed through race-oriented policy (see Santos 2006). This may help explain

why 25.7% (n=61) of respondents who answered ‘1 or 2’ in Q11 commented that quotas

should be based solely upon socioeconomic criteria.56 Moreover, Interviewee-M’s remarks

indicate that even though respondents perceive the RQP as justice for beneficiaries, some

may still perceive RQs as unfair, rejecting the policy due to their own self-interests with RQs

intensifying competition amongst, and fundamentally reducing the chances of, the elite

classes obtaining a prized public university place.

“The whites don’t want their children to not enter university through the vestibular because

the racial quota students are getting in. But the blacks want to have the quota otherwise they

won’t be able to get into university.” mmmmOooooooooooooooooooooooooooooooooooooo

o…………………………((Interviewee-R, a self-identified pardo (mestiço) RQ student at UERJ)

53 Representing 93.6% of all those who marked ‘abolished’ in Q13. 54 Representing 8% of all those who marked ‘continue in current format’ in Q13. 55 47.3% (n=112) of respondents marked modified. Of these 97 commented in Q13/15 demonstrating disapproval for RQs; for instance 30.1% (n=31) commented on Category-B (just SQs) and 19.6% (n=19) on Category-A (Basic Education Quality/Investment) and only 16.5% on Categories C (Palliative Measure), E (Temporary Measure) and K (RQs with Basic Education) as described in Chapter 9.2a. 56 See appendix-D/Category-B

26

The references: “these black people” “the whites” “the blacks” used by Interviewees M and R,

exemplify common descriptors used by many interviewees and questionnaire respondents

and are potentially insightful. It has been noted that racial classification in Brazil is both

complicated and ambiguous, however it appears respondents tacitly acknowledge that the

‘for’ and ‘against’ dichotomy toward the RQP is identifiable along white and black lines. With

this in mind, the relationship between self-classified race/colour (as measured by the

branco/pardo/preto 0-4 scale in Q3) and perceived level of ‘fairness’ in relation to RQs as an

intervention to reduce racial inequalities of access to university education. (as measured by

the perceived fairness 1-5 scale in Q11) was investigated using Spearman’s correlation

coefficient as presented in Chapter 9.1b. There was a small but significant positive

correlation between the two variables (R=.144, n=463, p<.002), with darker levels of self-

classified race/colour within categories 0-4 associated with higher levels of perceived

fairness.

Additionally, a multivariate analysis was performed using an ordered-probit model.

Considering race as an ordinal variable and controlling for: 1.whether students attend a

public/private university; 2.gender; 3.age; 4.PRSOL; 5.how effective respondents perceive

RQs in reducing racial inequality in the university system, the previous finding remains highly

significant with control variables 1, 3 and 4 also significant (Figure-4). However, here race is

considered as an interval variable and may not strictly fulfil the assumptions of the model

given that differences between the racial categories cannot be measured in this way.

Representing race as dummy variables reveals that in comparison to Category 0 (branco) all

non-white categories (1-4) perceive the RQP as fairer, although only two are significant

(Figure-5).57

Figure-4

Fairness

Coefficient

P>|z| 1.Private -.2623576 0.015** 2.Female -.0693786 0.511 3.Age .02825 0.005*** 4.Effectiveness .8148919 0.000*** 5.PRSOL .0644169 0.307 6.Race .1229083 0.010** Figure-5 Figure-5

1.Pardo-(Branco) .117786 0.420 2.Pardo-(Mestiço) .2716659 0.080* 3.Pardo-(Negro) .2841654 0.269 4.Preto .5141827 0.035**

*significant at 10% **significant at 5% ***significant at 1%

57 See appendix-C1/C2 for full results.

27

Dichotomising the dependent variable (options 1 and 2 combined in Q11), a standard-probit

using the same control variables shows that the race coefficient is again highly significant as

well as control variables 1 and 4 (Figure-6), revealing that the darker respondents are, the

less likely they are to deem RQs as unfair. With race as dummy variables, as before, the

darkest categories (3 and 4) are highly significant (Figure-7).58

Figure-6

Fairness

Coefficient

P>|z|

1.Private -.308986 0.025**

2.Female -.0595317 0.662

3.Age .0184225 0.165

4.Effectiveness .701607 0.000***

5.PRSOL .0630745 0.441

6.Race .1827351 0.004***

Figure-7

1.Pardo-(Branco) .1956027 0.289

2.Pardo-(Mestiço) .2804135 0.160

3.Pardo-(Negro) .6928187 0.048**

4.Preto .7422576 0.028**

*significant at 10% **significant at 5% ***significant at 1%

While these results align to intuitive thought, R-values are relatively low. 41.5% (n=78) of

self-classified non-whites perceive RQs as unfair; of these, considering PRSOL, 35.9%

(n=28) answered ‘equal’ or ‘less’ in Q14. Moreover, while many remain neutral,59 almost as

many self-classified pardo (negros) and pretos perceive RQs as unfair/fair, suggesting even

potential RQ beneficiaries deem RQs as unfair.

However, these data should, again, be interpreted cautiously; while the sample is large

enough for Category 0 (branco) it is relatively small for Categories 1-4. Therefore,

complementary research with a larger representative sample of non-whites could help

assess the reliability of this significant finding. Overall -across all race/colour categories-

many respondents60 appear neutral in their perceptions regarding the fairness of RQs.

Moreover, these data demonstrate that while darker individuals perceive RQs as fairer, there

is no evidence to suggest perceptions of fairness are clearly divided along white/black lines.

58 See Appendix-C3/C4 for full results. 59 Option 3 in Q11. See Appendix-B2 Report-1 60 n=145

28

10. Conclusion

Before discussing the implications of this research, a few points of caution are in order. The

sample is substantial yet still relatively limited, conducted at only two universities out of the

hundreds available within Brazil. Moreover, both universities were located in just one city

(RJ) and there are reasons to expect that issues regarding AA would play out differently

elsewhere. That said, the data still provide some interesting findings; as discussed below.

Chapter 9.1 highlights that respondent’s perceptions of racial classification are as

contentious, complex and inconsistent as the literature and its adjoining theories propose;

therefore, it is difficult to make conclusive interpretations of the data. Findings highlight that

many students perceive race/colour classification of others and oneself as difficult and

subjective; thus, the appropriateness of self-classification as a method by which students

apply for RQs appears questionable. The self-classification model used in this study

illustrates this problem; responses indicate that many pardos (46.7%, n=77) would likely opt

for whiteness if constrained to dichotomous classification. Therefore, AA administrators

hoping to effectively target RQs may need to rethink continuing under the premise that negro

incorporates both the pardo and preto census categories (Bailey, 2008; Schwartzman, 2008).

Conversely, A significant negative correlation between darker levels of self-classified

race/colour within pardo categories and lower levels of PRSOL indicates that using a

dichotomous white/black model of self-classification for RQ admissions may reveal greater

levels of racial inequality amongst pardos. Therefore, without a robust problem-free

alternative, this method, alongside socioeconomic criteria testing, is justified as an effective

way as any at present in targeting Brazil’s most socioeconomically disadvantaged ‘non-

whites’ (Loveman, et al., 2011).

Despite perceptions of race/colour and identity being multifaceted and inconsistent, it is

interesting that three-quarters of respondents acknowledge that education has been/is a

contributing factor to racial inequality in Brazil 61 with broadly two-thirds perceiving the

university system as a contributing factor. Therefore, unsurprisingly, a majority of

respondents believe RQs have, to some extent, reduced racial inequality since their

introduction. However, noteworthy is that a significant proportion of respondents (38%,

n=212) perceive RQs to have had no effect upon, or even increased racial inequality since

their implementation. Moreover, half of respondents critically evaluate the RQP as an

intervention to tackle the problem, claiming it is ineffective and/or unfair.62 Aligning to intuitive

thought, I find a series of significant correlations between darker levels of self-classified

61 Aligning to findings of (Bailey, 2004; Bourguignon et al., 2008) 62 Of these, 78% (n=185) claimed RQs as both ineffective and unfair. Suggesting perceptions of effectiveness and fairness are not always aligned.

29

race/colour and their perceived level of fairness regarding RQs. These and the finding

regarding PRSOL suggests respondent’s perceptions are, to some extent, identifiable along

racial lines; however, no distinct racial divide is found within these perceptions with many

darker individuals, with lower levels of PRSOL -potential beneficiaries of RQs- deeming

them as unfair.

Qualitative data examined in Chapters 9.2 and 9.3 helps to explain why -despite the majority

of respondents recognising the university system has historically contributed to racial

inequality- many deem the RQP as an inappropriate intervention. Many students commented

that RQs create an unnecessary racial divide, reaffirming/generating prejudice and racism

within Brazilian society as opposed to embracing the notion that all races/colours are equal

and, therefore, should be treated as equal. These respondents display a preference to

perceive the hardships experienced by non-whites to result from socioeconomic

factors/status (Santos, 2006) and, therefore, prefer colour-blind policies (Schwartzman &

Silva, 2012:32), which focus primarily/solely upon socioeconomic criteria. Moreover, a

significant number of respondents commenting (n=124) exemplified a stance of anti-racialism,

referring to aspects of miscegenation and no racial division. Combined these factors highlight

how certain concepts of the racial democracy myth remain deeply internalized within

Brazilian society (Telles, 2004:77). Also, over a third of all respondents commented that

insufficient quality/investment in the basic education system drives racial inequalities of

access to university education. Therefore, many see the RQP as a palliative measure and/or

should be a temporary initiative, one that alone is unlikely to effectively mitigate the access

problem.

The above signifies why, a decade after the enactment of RQs in university admissions,

public debates remain strong and polarised. By definition, perception cannot be determined

as right or wrong and therefore debate aiming to convince proponents/opponents of one

another’s stance could prove futile. Instead, public debates should focus upon what kind of

society Brazilians wish to build (Santos, 2006:43). Alongside social justice -the principal

rationale behind Brazil’s AA policies- (Moses, 2009:221), there exists a clear set of economic

motivations. Gary Becker (1971; 1993) noted that discrimination against large minorities (the

case in Brazi) reduces the overall incomes of all. Severe and pervasive inequality often

perpetuates economic inefficiency, leads to social instability and reduced social solidarity,

fosters political corruption (World Bank, 2004:5; Todaro & Smith, 2009:222-23) and incites

crime and violence (Fajnzylber et al., 2002; Sachsida et al., 2010). This describes a Brazil of

which its people have long grown tired, signified by the recent countrywide demonstrations.

30

New data in this study highlights that justifications opposing RQs throughout Brazilian society

prevail in the arguments of many students at UERJ and PUC-Rio; driven by perceptions of

race/colour, identity and inequality that conflict with the reality many Brazilians face.

University students of today are society’s elite of tomorrow, those who will possess the

economic, social, and political capital necessary to enact positive change; propagating both

equity and equality, amongst a Brazilian population that has persistently remained highly

stratified.

Overall this study shows that racial classification in Brazil is highly complex. Despite that

there is a general strong feeling amongst the majority of respondents that education

(including the university system), has been/is a contributing factor to racial inequality.

Therefore, broadly speaking, respondents believe that some form of intervention is required;

however a significant proportion of respondents do not positively evaluate RQs, questioning

their fairness and effectiveness, regardless of their race. To explain this puzzlement we

return to the complexity of race, complexity that the policy struggles to engage with at a

societal level and ultimately influences respondent’s conceptualisation of inequality and

fairness; therefore shaping their overall attitude toward the policy.

The complications regarding the formulation and implementation of RQs are clearly evident,

unique in the Brazilian experience of AA in comparison to other nations. Although the RQP is

not considered a panacea for resolving Brazil’s racial inequality, it appears to be a policy of

progressive change. Therefore, if policymakers wish to generate widespread support for -and

ensure the effectiveness of- RQs in university admissions, they should consider ways of

modifying the policy so that it better engages with the discussed race-related issues,

particularly those pertaining to the category of pardo.

31

11.References

AZZONI, C. R. (2001). Economic growth and regional income inequality in Brazil. ANNALS OF REGIONAL SCIENCE. 35, 133-152.

BACELAR, J. (1999). ‘Blacks in Salvador: Racial Paths’, in CROOK, L., & JOHNSON, R. (eds) Black Brazil: culture, identity, and social mobilization. UCLA Latin American Center Publications, 85-101.

BAILEY, S. (2002) ‘The race construct and public opinion: understanding Brazilian beliefs about racial inequality and their determinants’, American Journal of Sociology, vol. 108, pp. 408–439.

BAILEY, S. R. (2004). Group Dominance and the Myth of Racial Democracy: Antiracism Attitudes in Brazil. AMERICAN SOCIOLOGICAL REVIEW. 69, 728-747.

BAILEY, S., & TELLES, E. (2006). Multiracial versus collective black categories. Ethnicities. 6, 74-101.

BAILEY, S. R. (2008). Unmixing for Race Making in Brazil. The American Journal of Sociology. 114, 577-614.

BAILEY, S. R. (2009). Legacies of race: identities, attitudes, and politics in Brazil. Stanford, Calif, Stanford University Press.

BAILEY, S. R., & PERIA, M. (2010). Racial Quotas and the Culture War in Brazilian Academia. Sociology Compass. 4, 592-604.

BARROS, R., HENRIQUES, R., & MENDONÇA, R. (2000). Desigualdade e pobreza no Brasil: retrato de uma estabilidade inaceitável (Inequality and poverty in Brazil: Portrait of an unacceptable stability). Revista Brasileira De Ci ê ncias Sociais. http://www.scielo.br/scielo.php?script=sci%5Farttext&pid=S0102-69092000000100009.

BASTOS, E. R. (2001). Brasil: um outro Ocidente? Gilberto Freyre ea formação da sociedade brasileira. (Brazil: another West? Gilberto Freyre and the formation of Brazilian society) Ciência & Trópico, 33-60.

BECKER, G. S. (1971). The economics of discrimination. Chicago, University of Chicago Press.

BECKER, G. S. (1993). The economic way of looking at life. [Chicago, IL], University of Chicago Law School.

32

BOURGUIGNON, F., FERREIRA, F. H. G., & LEITE, P. G. (2008). Beyond Oaxaca–Blinder: Accounting for differences in household income distributions. The Journal of Economic Inequality. 6, 117-148.

CASTRO, C. (2001). Educação superior e equidade: inocente ou culpada. Ensaio: avaliação de políticas públicas em Educação , 30, 110-120.

CICALO, A. (2008). What Do We Know About Quotas? Data and Considerations About the Implementation of the Quota System in the State University of Rio de Janeiro (UERJ). univ. humanist, 262-280.

CICALO A. (2012). Nerds and Barbarians: Race and class encounters through affirmative action in a Brazilian university. Journal of Latin American Studies. 44, 235-260.

SILVA, D. V. N., & HASENBALG, C. A. (1999). Race and educational opportunity in Brazil. in REICHMANN, R. L., ‘Race in contemporary Brazil: from indifference to inequality’, The Pennsylvania University, University Park, PA, 53-65.

FAJNZYLBER, P., LEDERMAN, D., & LOAYZA, N. (2002). Inequality and Violent Crime*. The Journal of Law and Economics. 45, 1-39.

FIESS, N. M., & VERNER, D. (2003). Migration and human capital in Brazil during the 1990s. Washington, D.C., World Bank.

FRANCIS, A. M., & TANNURI-PIANTO, M. (2012). Using Brazil's Racial Continuum to Examine the Short-Term Effects of Affirmative Action in Higher Education. Journal of Human Resources. 47, 754-784.

FREYRE, G. (1956). The masters and the slaves: (Casa-grande & senzala) A study in the development of Brazilian civilization. New York, Knopf.

FRIAS, L. (2012) As cotas raciais e sociais em universidades públicas são injustas? (Are the social and racial quotas in public universities unfair?) Direito, Estado e Sociedaden. 41, 130-156. http://direitoestadosociedade.jur.puc-rio.br/cgi/cgilua.exe/sys/start.htm?infoid=245&sid=23

GOLDSTEIN, D. M. (2003). Laughter out of place: race, class, violence, and sexuality in a Rio shantytown. Berkeley, University of California Press.

GUIMARãES, A. S. A. (2007, February). New ideological inflections in the study of racism in Brazil. In UCLA-Havard Interdisciplinary Roundtable Conference on Racial, Ethnical and Caste Discrimination and Remedial Measures in Global Perspective, Los Angeles.

HANCHARD, M. (1994) Orpheus and Power: The Movimento Negro of Rio de Janeiro and São Paulo, Brazil, 1945–1988. Princeton, NJ: Princeton University Press.

33

HARRIS, M. (1964). Patterns of race in the Americas. New York, Walker.

HERNANDEZ, T. K. (2005). To Be Brown in Brazil: Education and Segregation Latin American Style. Review of Law and Social Change. 29, 683-718.

HOLZER, H. J., & NEUMARK, D. (2006). Affirmative action: What do we know?. Journal of Policy Analysis and Management, 2, 463-490.

HTUN, M. (2004). From "Racial Democracy" to Affirmative Action: Changing State Policy on Race in Brazil. Latin American Research Review. 39, 60-89.

LAESER (Laboratório de Análises Econômicas, Históricas, Sociais e Estatística das relações raciais), (2012). “Estado da arte das ações afirmativas no Brasil, partes 1 e 2.” Tempo em Curso 4 (7 e 8), www.laeser.ie.ufrj.br

LIMA, M. (2011). Access to higher education in Brazil: Inequalities, educational system and affirmative action policies. Retrieved from http://www2.warwick.ac.uk/fac/soc/ier/research/glmf/heeer/acces_higher_education_i n_brazil_marcia_lima.pdf

LLOYD-SHERLOCK, P. (2009). Social Policy and Inequality in Latin America: A Review of Recent Trends. Social Policy & Administration. 43, 347-363.

LOVEMAN, M., MUNIZ, J. O., & BAILEY, S. R. (2011). Brazil in black and white? Race categories, the census, and the study of inequality. Ethnic and Racial Studies. 1-18.

GARCIA, M. L., ÑOPO, H., & SALARDI, P. (2009). Gender and racial wage gaps in Brazil 1996-2006: evidence using a matching comparisons approach. Washington, DC, Inter-American Development Bank, Research Dep. http://hdl.handle.net/10419/51472.

MARTELETO, L. J. (2012). Educational Inequality by Race in Brazil, 1982-2007: Structural Changes and Shifts in Racial Classification. Demography. 49, 337-358.

MCCOWAN, T. (2007). Expansion without Equity: An Analysis of Current Policy on Access to Higher Education in Brazil. Higher Education. 53, 579-598. http://eprints.ioe.ac.uk/2526/1/McCowan2007Expansion579.pdf

MOSES, M. S. (2010). Moral and Instrumental Rationales for Affirmative Action in Five National Contexts. Educational Researcher. 39, 211-228.

O GLOBO. (2013). ‘Candidato de pele branca é aprovado por cotas raciais na 1ª fase do Itamaraty’ http://oglobo.globo.com/educacao/candidato-de-pele-branca-aprovado-por-cotas-raciais-na-1-fase-do-itamaraty-9908199 (Published September 11th), accessed 3/1/2014

PERLMANN, J., & WATERS, M. (2002). ‘Introduction’, in J. Perlmann and M. Waters (eds)

34

The New Race Question, pp. 1–30. New York: Russell Sage Foundation.

PIZA, E., & ROSEMBERG, F. (1999). ‘Color in the Brazilian census’, in REICHMANN, R. L., ‘Race in contemporary Brazil: from indifference to inequality’, The Pennsylvania University, University Park, PA, 37-52

RAWLS, J. (1971). A theory of justice. Cambridge, MA, Belknap Press of Harvard University Press.

REICHMANN, R. L. (1999). Race in contemporary Brazil: from indifference to inequality. The

Pennsylvania University, University Park, PA,.

ROEMER, J. E. (1998). Equality of opportunity. Cambridge, Mass, Harvard University Press.

ROSEMBERG, F. (2004). O branco no IBGE continua branco na ação afirmativa? Estudos avançados, (White in IBGE continues White in affirmative action? Advanced Studies) 50, 61-66.

SANSONE, L. (1995) ‘O Local e o Global na Afro-Bahia Contemporânea’ (The Local and the Global in Contemporary Afro-Bahia), Revista Brasileira de Ciencias Sociais 29, 65–84.

SANSONE, L. (2003). Blackness without ethnicity: constructing race in Brazil. New York, Palgrave MacMillan.

SANTOS, S. A. D. (2006). Who Is Black in Brazil? A Timely or a False Question in Brazilian Race Relations in the Era of Affirmative Action? Latin American Perspectives. 33, 30-48.

SACHSIDA A., DE MENDONCA M.J.C., GUTIERREZ M.B.S., & LOUREIRO P.R.A. (2010). Inequality and criminality revisited: Further evidence from Brazil. Empirical Economics. 39, 93-109.

SCHWARTZMAN, S. (1999) ‘Fora de Foco: Diversidade e Identidades Étnicas no Brasil’ (Out of Focus: Diversity and Ethnic Identities in Brazil), Novos Estudos 55, 83–96.

SCHWARTZMAN, L. F. (2008). Who are the Blacks?. The Question of Racial Classification in Brazilian Affirmative Action Policies in Higher Education. Cahiers de la Recherche sur l’Education et les Savoirs, 7, 27-47.

SCHWARTZMAN, L. F. (2009). Seeing like Citizens: Unofficial Understandings of Official Racial Categories in a Brazilian University. Journal of Latin American Studies. 41, 221-250.

SCHWARTZMAN, S. (2011). Academic Drift in Brazilian Education La tendencia academicista en la educación brasilera. Pensamiento Educativo. Revista de Investigación Educacional Latinoamericana 48, 15-26 http://pensamientoeducativo.uc.cl/files/journals/2/articles/380/public/380-1180-1-PB.pdf , accessed 30/11/2013

35

SCHWARTZMAN, L. F., & DA SILVA, G. M. D. (2012). Unexpected Narratives from Multicultural Policies: Translations of Affirmative Action in Brazil. Latin American and Caribbean Ethnic Studies. 7, 31-48.

SULTANA, F. (2007). Reflexivity, positionality and participatory ethics: Negotiating fieldwork dilemmas in international research. ACME: An International E-Journal for Critical Geographies, 3, 374-385.

TELLES, E. E. (2004). Race in another America: the significance of skin color in Brazil. Princeton, N.J., Princeton University Press.

TELLES, E., & PAIXãO, M. (2013). Affirmative Action in Brazil. LASAFORUM (Published spring 2013, volume xliv, issue 2) http://lasa.international.pitt.edu/forum/files/vol44-issue2/Debates4.pdf , accessed 12/12/2013

THE BBC. (2013). ‘Brazil’s Universities Take Affirmative Action’ http://www.bbc.co.uk/news/business-23862676 (Published August 28th), accessed 3/1/2014 THE ECONOMIST. (2013). ‘Affirmative-action bans ‘Minority report’ http://www.economist.com/blogs/democracyinamerica/2013/10/affirmative-action-bans (Published October 23rd), accessed 3/11/2013 THE ECONOMIST. (2013a). ‘A Never Ending Policy’ http://www.economist.com/news/briefing/21576654-elections-may-could-mark-turning-point-never-ending-policy (Published April 27th), accessed 3/11/2013

THE ECONOMIST. (2013b). ‘Affirmative Action in Brazil – Slavery’s Legacy’ http://www.economist.com/blogs/americasview/2013/04/affirmative-action-brazil#comments (Published April 26th), accessed 3/11/2013

THE ECONOMIST. (2013c). Protests in Brazil: ‘Taking to the Streets’ ‘http://www.economist.com/news/americas/21579857-bubbling-anger-about-high-prices-corruption-and-poor-public-services-boils-over (Published June 22nd), accessed 5/11/2013

THE NY TIMES. (2010). ‘Campus That Apartheid Ruled Faces a Policy Rift’ http://www.nytimes.com/2010/11/23/world/africa/23safrica.html?pagewanted=all&_r=0 (Published November 22nd), accessed 3/11/2013

THE RIO TIMES. (2012). Brazil’s University Affirmative Action Law http://riotimesonline.com/brazil-news/front-page/brazil-senate-passes-university-affirmative-action-law/ (Published August 14th), accessed on 4/11/2013

TODARO, M. P., & SMITH, S. C. (2012). Economic development. Boston, Mass, Addison-Wesley.

36

TWINE, F. W. (1998). Racism in a racial democracy: the maintenance of white supremacy in Brazil. New Brunswick, NJ [u.a.], Rutgers University Press.

UNESCO (2004) Statistical Tables: Education. http://www.uis.unesco.org/DataCentre/Pages/country-profile.aspx?code=ARG&regioncode=40520 , accessed 3/1/2014

WADE, P. (1997). Race and ethnicity in Latin America. Chicago, Ill, Pluto Press.

WORLD BANK STAFF. (2004). Inequality and Economic Development in Brazil. T WORLD BANK (E-Library). Washington, World Bank Publications. http://www-wds.worldbank.org/servlet/WDSContentServer/WDSP/IB/2004/10/05/000012009_20041005095126/Rendered/PDF/301140PAPER0Inequality0Brazil.pdf

WORLD BANK (2013) http://www.worldbank.org/en/news/feature/2013/01/25/Brazil-Argentina-Mexico-lead-fight-against-inequality Accessed on 17/10/2013.

WORLD BANK (2013a). http://data.worldbank.org/indicator/SI.POV.GINI Accessed on 17/10/2013.

37

Appendix-‐A1: Table_1-‐Quantitative Data Variable

Description N.

obs. Mean

%

Private Vs.

Public (Known)

Dummy = 1 if subject is a public university student; Dummy = 1 if subject is a private university student (baseline). No specific question required to obtain data.

489 0.542 0.458

Q1. Male/ Female

Dummy = 1 if subject is male; Dummy = 1 if subject is female (baseline)

489 0.597 0.403

Q3.ord SCR/C

Ordinal according to Self-‐Classification of Race/Colour subject’s response exhibited in question 3

0 = Branca, 1 = Parda (Branca), 2 = Parda (Mestiça), 3 = Parda (Negra), 4 = Preta, 5 = Amarela, 6 = Indígena, 7 = Other.63

Subjects choosing categories 5, 6 & 7 = (7)+(4)+(7) = 18/0.037%

488 (470)

0.99 (0.80)

0 = 57.2 1 = 15.8 2 = 13.9 3 = 4.3 4 = 5.1 5 = 1.45 6 = 0.8 7 = 1.45

(0= 59.4) (1= 16.4) (2= 14.5) (3= 4.4) (4= 5.3)

Q4(a). A NA

Dummy = 1 if subject believes categories provided are not adequate to distinguish their race/colour. Dummy = 1 if subject believes categories provided are adequate to distinguish their race/colour (baseline).

456 0.408 0.592

Q5. Age

Age in years (17;58)

a = 17;19, b = 20;22, c = 23;25, d = 26;29, e = 30>

488 21.90

a= 43.647 b= 31.148 c=11.065 d= 5.533 e= 8.607

Q6. VC

Dummy = 1 if subject studies prestigious vocational course (Accounting Sciences (N=35), Administration (11), Economics (5), Engineering (6), International Relations (55), Law (69), Medicine (84), Psychology (41), Philosophy (1).

307 0.633

Q6. NVC

Dummy = 1 if subject studies any other discipline other than prestigious vocational degree (Art (N=2), Cinema (5), Design (14), Education (14), Geography (1), History (19), Journalism (4), Languages (8), Literature (2), Publicity (7), Social Communication(81), Sociology (19), Visual Arts (2))

178 0.367

Q.7.ord Ordinal according to how the Brazilian university system prior to racial quotas contributed to racial inequality, subject’s responses exhibited in question 7. 1 = produced less, 2 = produced more, 3 = had no effect

479 2.451

1= 16.075 2= 62.004 3= 21.921

Q8.ord Ordinal according to how the Brazil’s university system since the introduction of racial quotas contributed to racial inequality, subject’s responses exhibited in question 8. 1 = reduced racial inequality, 2 = increased racial inequality, 3 = had no effect

481 1.66

1= 55.925 2= 22.661 3= 21.414

Q10.ord Ordinal according to the degree of effectiveness subject’s responses exhibited in question 10. 1 = extremely effective, 2 = ineffective, 3 = neutral, 4 = effective 5 = extremely effective

483 2.54

1= 19.462 2= 29.607 3= 32.712 4= 13.250 5= 4.969

Q11.ord Ordinal according to the degree of fairness subject’s responses exhibited in question 11. 1 = extremely unfair, 2 = unfair, 3 = neutral, 4 = fair, 5 = extremely fair

482 2.56

1= 23.651 2= 25.519 3= 30.083 4= 12.656 5= 8.091

Q13.ord Ordinal according to the trajectory of the RQP, subject’s responses exhibited in question 13. 0 = Abolished, 1 = Modified, 2 = Continue in its current format

479 0.89

0= 26.304 1= 58.037 2= 15.657

Q14. PRSOL

Perceived Relative Standard of Living to other families in Brazil. Ordinal: 1 = very much lower, 2 = lower, 3 = equal, 4 = higher, 5 = very much higher;

477 3.782

1= 0.419 2= 9.853 3= 20.126 4= 50.315 5= 19.287

63 Other: None specified (3 subjects), Mestico/Pardo/Papel (1), Morena (1), Branca/Preta/India (1), Negra (1)

38

Appendix-‐A2: Table_2-‐Quantified Qualitative Data

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! !!!!!!!! !1!Overlaps refer to comments made by respondents across multiple categories in either question (no respondent made comments across more than 4 categories).

SECTION 1: Q.4 (b). Do you think the above categories are adequate to distinguish your race/colour? (Please Comment)

Total No. of student respondents who commented

263

Total No. of student respondents whose comments relate to categories 1-5

206

1 Miscegenation 72

2 No Racial Division 52

3 Difficult to Classify 52

4 Subjective 28

5 Use Fewer Categories 8

Total No. of comments for categories 1-5

212

Total No. of comment overlaps1 6

SECTION 2: Q13. In your opinion the racial quotas system should be: (1) Abolished (2) Modified (3) Continued in Current Format (Please Comment)

Q15. Any other comments?

Q13 & Q15. Total no. of student respondents

No. of student respondents who commented N = 378 N = 102

383 (480 minus 97 overlaps)

No. of student respondents whose comments relate to categories A-O:

N = 341 N = 57 345 (398 minus 53 overlaps)

Categories No. of comments

Total N. of comments Q13 & Q15

Q13. Q15.

A Problems in Basic Education/Investment needed 150 19 169

B Just Social Quotas 94 8 102

C Palliative Measure 37 13 50

D Deceit 36 12 48

E Temporary Measure 24 5 29

F Racial and Social Quotas 22 22

G Race/Subordinate/ Capacity/Prejudice 19 2 21

H Racism / Discrimination 18 2 20

I Meritocracy 16 2 18

J Race does not define Capacity 15 15

K Racial Quotas with Basic Education 10 2 12

L Capacity Problems 8 3 11

M Increases Prejudice 8 3 11

N Racialisation 7 1 8

O More Social than Racial 7 1 8

Total No. of comments for categories A-O

471 73 544

Total No. of comment overlaps

130 16 146

39

Appendix-‐B1: PRSOL*Race-‐Crosstabualtion Report 1: Perceived Relative Standard of Living * RACE Crosstabulation

RACE

Total Pardo

(Branco) Parda

(Mesitca) Pardo

(Negro)

RELATIVE_S.O.L Less Count 7 15 5 27

% of Total 4.3% 9.3% 3.1% 16.8%

Equal Count 20 15 7 42

% of Total 12.4% 9.3% 4.3% 26.1%

Better Count 40 29 8 77

% of Total 24.8% 18.0% 5.0% 47.8%

Very much

better

Count 9 6 0 15

% of Total 5.6% 3.7% .0% 9.3%

Total Count 76 65 20 161

% of Total 47.2% 40.4% 12.4% 100.0%

Correlations

RACE PRSOL

Spearman's rho RACE Correlation Coefficient 1.000 -.223**

Sig. (2-tailed) . .004 N 166 161

PRSOL Correlation Coefficient -.223** 1.000

Sig. (2-tailed) .004 .

N 161 161

**. Correlation is significant at the 0.01 level (2-tailed).

64 branco mean =4.02

Report 2: Perceived Relative Standard of Living in Comparison to Other Brazilian Families

Race Mean N

Std.

Deviation

Std. Error of

Mean Range Variance

% of Total

N Skewness

Pardo

(Branco)64

3.6711 76 .80644 .09251 3.00 .650 47.2% -.422

Parda

(Mesitca)

3.4000 65 .94868 .11767 3.00 .900 40.4% -.213

Pardo (Negro) 3.1500 20 .81273 .18173 2.00 .661 12.4% -.296

40

Appendix-‐B2: Fairness*Race-‐Crosstabulation Report 1: FAIRNESS * RACE Crosstabulation

FAIRNESS RACE

Total Branco Pardo (Branco) Parda (Mesitco) Pardo (Negro) Preto

Extremely

Unfair

Unfair

Neutral

Fair

Extremely Fair

Count 70 19 12 4 4 109

% within Race 25.5% 25.0% 17.9% 19.0% 16.7% 23.5%

% of Total 15.1% 4.1% 2.6% .9% .9% 23.5%

Count 83 19 16 1 3 122

% within Race 30.2% 25.0% 23.9% 4.8% 12.5% 26.3%

% of Total 17.9% 4.1% 3.5% .2% .6% 26.3%

Count 73 23 22 9 10 137

% within Race 26.5% 30.3% 32.8% 42.9% 41.7% 29.6%

% of Total 15.8% 5.0% 4.8% 1.9% 2.2% 29.6%

Count 30 10 10 5 2 57

% within Race 10.9% 13.2% 14.9% 23.8% 8.3% 12.3%

% of Total 6.5% 2.2% 2.2% 1.1% .4% 12.3%

Count 19 5 7 2 5 38

% within Race 6.9% 6.6% 10.4% 9.5% 20.8% 8.2%

% of Total 4.1% 1.1% 1.5% .4% 1.1% 8.2%

Total Count 275 76 67 21 24 463

% within Race 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

% of Total 59.4% 16.4% 14.5% 4.5% 5.2% 100.0%

Correlations RACE FAIRNESS

Spearman's rho RACE Correlation Coefficient 1.000 .144** Sig. (2-tailed) . .002 N 470 463

FAIRNESS Correlation Coefficient .144** 1.000

Sig. (2-tailed) .002 .

N 463 463

**. Correlation is significant at the 0.01 level (2-tailed).

Report 2: FAIRNESS * RACE Crosstabulation Summary

RACE Mean N

Std.

Deviation

Std. Error

of Mean Minimum Maximum

% of Total

Sum

Branco 2.4364 275 1.18018 .07117 Extermely Unfair Extremely Fair 56.7%

Pardo (Branco) 2.5132 76 1.19436 .13700 Extermely Unfair Extremely Fair 16.2%

Parda (Mesitca) 2.7612 67 1.21966 .14900 Extermely Unfair Extremely Fair 15.7%

Pardo (Negro) 3.0000 21 1.22474 .26726 Extermely Unfair Extremely Fair 5.3%

Preto 3.0417 24 1.33447 .27240 Extermely Unfair Extremely Fair 6.2%

Total 2.5529 463 1.20825 .05615 Extermely Unfair Extremely Fair 100.0%

41

Appendix-‐C1: Fairness*Race (Ordered-‐Probit)

Number of obs. = 453 LR chi2(6) = 270.38 Prob > chi2 = 0.0000 Pseudo R2 = 0.1975

Fairness Coef. Std. Err. z P>|z| [95% Conf. Interval

Private -.2623576 .1083351 -2.42 0.015 -.4746904 -.0500248

Female -.0693786 .1054401 -0.66 0.511 -.2760374 .1372802

Age .02825 .0100392 2.81 0.005 .0085735 .0479266

Effectiveness .8148919 .0560214 14.55 0.000 .705092 .9246917

PRSOL .0644169 .0631172 1.02 0.307 -.0592905 .1881243

Race .1229083 .047792 2.57 0.010 .0292377 .216579

Appendix-‐C2: Fairness*Race (Ordered-‐Probit)



Private -.2652564 .1086521 -2.44 0.015 -.4782106 -.0523023

Female -.066593 .105663 -0.63 0.529 -.2736886 .1405026

Age .0281616 .0100851 2.79 0.005 .0083951 .0479281

Effectiveness .8157109 .0561221 14.53 0.000 .7057136 .9257081

PRSOL .0639912 .0633905 1.01 0.313 -.0602519 .1882343

Pardo (Branco) .117786 .1459795 0.81 0.420 -.1683286 .4039006

Pardo (Mestiço) .2716659 .1550307 1.75 0.080 -.0321888 5755205

Pardo (Negro) .2841654 .2573331 1.10 0.269 -.2201982 .7885289

Preto .5141827 .2441566 2.11 0.035 .0356445 .9927209

Appendix-‐C3:_Fairness*Race (Standard-‐Probit)



Private -.308986 .1374734 -2.25 0.025 -.578429 -.039543

Female -.0595317 .1363946 -0.44 0.662 -.3268603 .2077969

Age .0184225 .0132713 1.39 0.165 -.0075888 .0444338

Effectiveness .701607 .0691972 10.14 0.000 .565983 .8372309

PRSOL .0630745 .0819038 0.77 0.441 -.097454 .223603

Race .1827351 .0631034 2.90 0.004 .0590547 .3064154

42

Appendix-‐C4: Fairness*Race (Standard-‐Probit)



Private -.3056966 .1376867 -2.22 0.026 -.5755575 -.0358357

Female -.0650203 .1367397 -0.48 0.634 -.3330253 .2029846

Age .0184568 .0133212 1.39 0.166 -.0076523 .044566

Effectiveness .7016709 .0693671 10.12 0.000 .565714 .8376279

PRSOL .0623911 .0819948 0.76 0.447 -.0983158 .223098

Pardo (Branco) .1956027 .1843684 1.06 0.289 -.1657527 .5569581

Pardo (Mestiço) .2804135 .1996392 1.40 0.160 -.1108722 .6716992

Pardo (Negro) .6928187 .351089 1.97 0.048 .004697 1.380941

Preto .7422576 .3373165 2.20 0.028 .0811294 1.403386

43

Appendix-‐E: Ethical Protocol

44

1.OVERVIEW OF STUDY Describe the purposes of the research/project proposed. Detail the methods to be used and the research questions. Provide any other relevant background which will allow the reviewers to contextualise your research or project activities. Include questionnaires/checklists as attachments, if appropriate.

Introduction

The research focuses on how University students in Brazil perceive the ‘affirmative action’ social policy; in particular, whether their perceptions have altered over time. The policy was introduced in 2001 in the hope of mitigating racial inequality in the Brazilian University system, whereby a percentage of University places were set aside for black students along with a “social quota” of pupils from state schools whose parents’ income is less than twice the minimum wage – again, often who are black. Since then, the contentious policy continues to evolve, creating a division in opinion over the appropriateness, effectiveness and fairness of the University system in Brazil in reducing racial inequalities.

My interest in social perceptions for this study originates from the time I have spent in Brazil previously (14 months). It was noticeable that social disparities within the country were often distinguished by race and, therefore, I am keen to examine the ‘affirmative action’ policy’s role in accentuating or indeed extenuating racial inequalities both in the University system and the wider society. In addition, I am interested to explore the relationship between social perceptions of racial inequality and economic well-being / development in Brazil; more specifically how social perceptions may affect the country’s progress in fulfilling its full economic potential.

Research Questions

1. How is the ‘affirmative action’ policy perceived by University students in Brazil?

2. What influences these perceptions?

Research Design

The main research element within my study will be in the form of a simple, non-intrusive, opinion based questionnaire comprising of 17 questions regarding the ‘affirmative action’ social policy in Brazil (please see appendix: 1A). Questionnaires and interviews will be administered at 2 separate Universities in one city, Rio de Janeiro on designated days during a month long period in Brazil. It is hoped that the research will include as diverse a sample as possible including individuals of different gender, ethnicity and University degree. It is hoped that as large a sample possible can be obtained with a minimum of 300 respondents to the questionnaire and 20 audio recorded face to face key informant interviews with University students and staff, once again relating to general opinions of the ‘affirmative action’ policy as opposed to personal experiences of the policy itself (please see appendix 1B).

Combined, the type of data generated from these research methods will provide me with a detailed source of quantitative and qualitative evidence which will be used to compliment any desk based research I may conduct, while enriching my analysis in answering my chosen research questions. The questionnaire will primarily produce numerical data that will be entered into SPSS for comparative statistical analysis; specifically highlighting students’ knowledge of the policy as well as their perceptions of its appropriateness, effectiveness and fairness over time.

The nature of the data generated through semi-structured interviews will be non-numerical and will instead generate a more in-depth data set, highlighting perspectives and perceptions regarding general aspects of the ‘affirmative action’ policy; including its objectives, implementation and outcomes.

45

Field work assistants/translators have been identified who, alongside me, will be instrumental in t

he delivery of the research topic, the distribution and collection of questionnaires as well as initiating the semi structured interviews with staff and students alike. Verbal and / or written consent from participating Universities and individual respondents will be sought for both the questionnaire completion and digitally recorded interviews. If necessary, all recordings will be translated into English and transcribed.

2. SOURCES OF FUNDING

The organisation, individual or group providing finance for the study/project.

3. RISKS OR COSTS TO PARTICIPANTS

What risks or costs to the participants are entailed in involvement in the research/project? Are there any potential physical, psychological or disclosure dangers that can be anticipated? What is the possible benefit or harm to the subject or society from their participation or from the project as a whole? What procedures have been established for the care and protection of participants (e.g. insurance, medical cover) and the control of any information gained from them or about them?

A series of issues and measures to address these will be adhered to during field work to ensure that the researcher, research assistant(s) and respondents are not subjected to increased risk of physical or psychological harm while taking part in the study. There are a number of issues for consideration here in relation to the risk to participants.

The nature of the questionnaire and interview questions helps to ensure that risks to participants in the form of disclosing sensitive information are significantly reduced. The questions are non-intrusive and based upon general opinions as opposed to individual experiences with personal information being limited to gender, race/ethnicity, age and home state. Due to the sensitivity over race within the topic being researched an additional question relating to race/ethnicity has been included in the questionnaire; allowing respondents to highlight whether such racial categorisation is appropriate.

The research is primarily targeting fellow University students of a fairly similar age to the researcher; therefore I do not envisage their being any boundaries or risks related to age. Barriers which could pose a problem are those of race and gender, given that I am a white western male conducting research about a social policy regarding racial inequality. I must be aware of my own position at all times and remain sensitive to any race /gender related issues that may result from interaction with respondents. It is difficult to completely eradicate potential problems; however I hope to mitigate them through maintaining a friendly disposition, interacting with participants in an informal yet professional manner, partially in the local language, providing a concise yet informative introduction of myself and briefing of my research.

It is vital that respondents do not feel there are any right or wrong answers or that they will be judged by their responses. Depending upon available facilities, there may be an option to complete the survey electronically via survey monkey; however, should this not be possible, paper copies of the questionnaire will be distributed and collected without any immediate

46

assessment. Any audio recording of interviews will require consent of interviewees, along with whether or not they wish to remain anonymous. Moreover, additional consideration will be made toward the presence of any assistant/translator during any interview process.

The questionnaire will be completed anonymously and the researcher/assistant(s) will ensure that any evidence obtained will remain strictly confidential. Issues of disclosure are addressed by giving each respondent verbal and written assurance of their anonymity and that all information exchange will be confidential. The safety of respondent information is also important here whereby hard copies of notes and completed questionnaires will be transported in hand luggage with combination locks and stored in a locked safe until they can be transferred to a secure filing system on academic premises and entered into computer software/documents which are password protected.

I will not look to ascertain information regarding the backgrounds of potential respondents to the questionnaire except for their University course. In addition, I will not be targeting any specific groups of individuals with regards to their gender, ethnicity or general background. All respondents will be those I am introduced to through University staff at each research location – most likely in a classroom scenario.

The above applies to any interviews the researcher plans to conduct also, however additional cautionary steps will be taken to mitigate any potential risks – all of which are to be made known during the introduction phase.

An informative introduction to my research and distribution of the questionnaires will take place during class time, as negotiated with the University and lecturers. At the time questionnaires are handed out I will also distribute information sheets regarding the key informant interview process which will contain my contact details. At this point I will enquire as to whether anyone would be interested to participate in the interview process. Students will be instructed to either complete the contact details section on the information sheet or contact me via email if they wish to volunteer. This will ensure that any potential interviewees can volunteer discreetly, unknown to their peers. It will be explained that in the event of an excess number of volunteers not all may be able to participate. From the list interview volunteers a diverse sample will be selected.

The researcher will offer a copy of the questions to be asked during in the interview to any willing participants so that they may familiarise themselves with the interview content. Following this, the researcher will attend at a later date (within a few days) to conduct interviews with those who have volunteered. This process will be repeated with any University staff that may wish to participate. In the first instance face to face interviews will be offered, however, the option of skype or phone interview will also be offered. Potential interviewees will be given the option of having an additional person present at any face to face interview in addition to any research assistant/translator. The researcher will also offer the option of whether participants wish the interview to take place in a private or quiet public area, with the use of any private room being confirmed with the University. Furthermore, a copy of the full interview recording will be offered to all interviewees who may review and then highlight any comments they wish to edit and/or indicate to the researcher any information they do not wish the researcher to use as part of the study.

Participation in this research is completely voluntary both in completing the questionnaire and being interviewed. This will be strongly emphasised during the introduction phase to all potential students. All students will have to give full written consent in order to participate and the opportunity to withdraw from the research at any moment without explanation. Potential

47

participants will not be approached personally or individually by either the researcher or the assistant. No key informant is to be a personal friend or acquaintance of the assistant / translator present in the interview process as this may cause a conflict of interests to both the participant and the assistant as well as potentially affect the data collected. The risk of such an event occurring will be minimised further as it is hoped the assistant / translator will not be a student from any of the classes / courses at either of the Universities visited.

It is hoped that these options will ensure that the interviewee has an element of control over the interview process, control which may make any potential participant feel less anxious or apprehensive over their involvement.

This anonymity and confidentiality helps to ensure respondents can feel comfortable in declaring their views and opinions about the research topic without the worry of being judged or reprimanded by their peers or professors.

4. RECRUITMENT/SELECTION PROCEDURES

How will study/project participants be selected? Is there any sense in which participants might be ‘obliged’ to participate – as in the case of students, prisoners or patients – or are volunteers being recruited? If participation is compulsory, the potential consequences of non-compliance must be indicated to participants; if voluntary, entitlement to withdraw consent must be indicated and when that entitlement lapses.

Recruitment

Access to the University and its students will be authorised by an official of the University. I have already made contact with several Professors who have indicated that they are willing to help but each have reiterated that this process will be more efficient and effective once I have arrived in Brazil. In addition, they have highlighted particular individuals that have a key interest in my chosen area of research and would most likely be willing to participate / assist. In any instance, written confirmation from each participating University will be sought stating that they are happy for me to undertake my research. In addition, permission for me to enter individual classes will be sought from individual lecturers.

This research will not be conducted until a formal invitation has been received after having shared my ethical protocol with each participating University. I am soon to be in a position to send more detailed information to contacts which shall include a more thorough introduction to the nature of my research interests, the duration of my visits to the University and the level of participation sought; assurance over respondent anonymity and data confidentiality; notification that a research assistant/translator will be present; an explanation of the need for their presence given that my intermediate level of Portuguese is not adequate to conduct this study alone; and finally, contact information for the University of East Anglia and a staff member who can vouch for the researcher and research and be able to answer any questions they may have in this regard. Should respondents consent then they can introduce me to students and additional staff members alike who may be willing to participate in the study.

Involvement in this research will be voluntary and consent from each individual participant will be obtained prior to participation. I should ensure that all those involved are happy to take part and, if signs of discomfort or coercion are apparent, then respondents should be reminded of their right to withdraw at any point.

48

All further information prior to the researchers’ arrival in Brazil will be sent via email. In the first instance, this will be in English although could be translated into Portuguese if necessary. Questionnaires will be produced in both English and Portuguese for participants to complete. Respondents will sign their name to highlight their willingness to participate. Prior to any interview, respondents will be asked whether or not they allow the researcher to digitally record proceedings.

Selection Procedures

The Universities involved in this research need not have the ‘affirmative action’ policy in place nor do students / staff completing the questionnaire / interviews need to match a particular profile.

As yet, Universities have not as much been selected but based upon contacts with which friends and acquaintances of the researcher have access to and to whom they have been able to introduce the research topic. These are a mixture of state and private universities, in some of which the ‘affirmative action’ social policy is active, located in the South Eastern region of Brazil in Rio de Janeiro. This contrast of Universities will allow for interesting comparative analyses of the primary data collected.

Student participants will not be randomly selected; however it is hoped that a diverse a range of students possible will be included in the study, including: different age, race/ethnicity, course of study as well as an even number of female and male participants. It is hoped that students from particular courses such as sociology and international development may complete the questionnaire in order to assess whether certain courses influence student perceptions of the ‘affirmative action’ social policy although this is not essential.

A random selection of responses from each University may be chosen for statistical analysis at a later stage. This considerably smaller sample would represent equal numbers of men and women from each University within each city.

5. PARTICIPANTS IN DEPENDENT RELATIONSHIPS

Specify whether participants will include students or others in a dependent relationship (this could affect their ability to decline to participate). If such participants will be included what will you do to ensure that their participation is voluntary etc.?

Respondents taking part in the questionnaire will be University students and will therefore not require written consent from parents or guardians to participate. Students will not be obliged to take part in this study and the voluntary aspect will be made clear during the introduction phase by both the researcher and research assistant(s). Consent from each individual participant will be obtained prior to participation. I should ensure that all those involved are happy to take part and, if signs of discomfort or coercion are apparent, then respondents should be reminded of their right to withdraw at any point.

6. VULNERABLE INDIVIDUALS

Specify whether the research will include children or people with mental illness. If so, please explain the necessity of involving these individuals as research participants and what will be done to facilitate their participation.

49

No children under the age of 18 will be part of this study. Neither will this study specifically target older people or disabled individuals who may be considered vulnerable.

7. PAYMENTS AND INCENTIVES

Will payment or any other incentive, such as a gift or free services, be made to any participant? If so, please specify and state the level of payment to be made and/or the source of the funds/gift/free service to be used. Please explain the justification for offering payment or other incentive.

No immediate benefit for the respondents will result from taking part in this study. For example, I will offer no financial reward in order to encourage people to participate. However, I hope to secure funding in order to cover food and / or transport costs as well as other expenses incurred by any research assistant/translator that may help with this study. In addition, at the end of the questionnaire / interview completion I will offer each participant and each assistant/translator a small token by way of showing my appreciation for the time and help they have given me.

The small token that I wish to offer to participants is likely to be UEA stationery, maybe a pen, pencil and eraser. However, the exact nature of my gift will be constrained to available finances. The research assistant / translator will receive payment to cover any expenses they may have incurred during the research process.

8. CONSENT

Please give details of how consent is to be obtained. A copy of the proposed consent form, along with a separate information sheet, written in simple, non-technical language MUST accompany this proposal form (do not include the text of the form in this space, attach with your submission as a separate document).

The letter of introduction will facilitate my entry into the Universities; I aim to ensure that all professors potentially involved in facilitating my research are made aware prior to my arrival through my lead contacts. Then, alongside my contact/assistant/translator I will further explain my project to professors and students alike in person, seeking their consent before distributing my research questionnaires for completion.

Involvement in this research will be voluntary and consent from each individual participant will be obtained prior to participation. I should ensure that all those involved are happy to take part and, if signs of discomfort or coercion are apparent, then respondents should be reminded of their right to withdraw at any point. Respondents will sign their name to highlight their willingness to participate. Consent will be made in advance of any interview with respondents being asked whether or not they will allow the researcher to digitally record proceedings and whether they are happy to have an assistant/translator present.

Consent forms in English have been attached but these will also be translated into Portuguese.

9. CULTURAL, SOCIAL, GENDER-BASED CHARACTERISTICS

Comment on any cultural, social or gender-based characteristics of the participants which have affected the design of the project or which may affect its conduct.

50

Respondents are chosen on the basis that they are current University students. In this way the researcher is able to answer the chosen research questions, identifying the potential shift or maintained perceptions of the relatively young and educated Brazilian population toward the ‘affirmative action’ policy over time – a policy which may directly affect / have affected those involved in the research.

Although, participants may have been affected by the policy in question, it is, once again, important to highlight that the information being asked for in the questionnaire / interviews pertains to general opinions as opposed to individual experiences directly occurring as a result of the policy, with personal information being limited to gender, race/ethnicity, age and home state. Due to the nature of the topic being researched an additional question relating to race/ethnicity has been included in the questionnaire; allowing respondents to highlight whether such racial categorisation is appropriate. In addition, an option ‘Prefer not to say’ has been included on any questions relating to the personal information highlighted above.

10. ENVIRONMENTAL IMPACT

Identify any environmental impacts arising from your research/project and the measures you will take to minimise risk of impact.

Although I do not envisage any significant differences being apparent, I will be sure to maintain a respectful way of undertaking my research in keeping with local cultural norms and values.

My own way of life and background could be very different to those who complete my research, although our shared University student status and relatively close age is likely to mitigate potential socio-cultural barriers, my white western appearance could result in prejudiced first impressions, however this is not anticipated to be a serious risk. Although difficult to completely eradicate this problem, I hope to mitigate it through maintaining a friendly disposition, interacting with participants in an informal way while providing a concise yet informative introduction of myself and briefing of my research. In addition, it is important that I take measures to minimise imposing my ideas and experiences onto other people, ensuring my introduction and briefing of my research remains neutral and non-leading. This is due to the fact that my questionnaire / interviews will relate to the opinions and perceptions of participants and I do not wish to cloud these personal judgements with my own subjective ideas.

The key environmental impact will be the flights I make to and from Brazil. I plan to make one return trip, thereby minimising my carbon-print.

11. CONFIDENTIALITY

Please state who will have access to the data and what measures which will be adopted to maintain the confidentiality of the research subject and to comply with data protection requirements e.g. will the data be anonymised?

The questionnaire will be completed anonymously and the researcher/assistant(s) will ensure that any evidence obtained will remain strictly confidential. I am confident that the privacy of the people participating in the research will not be violated and that all information they give will be protected. This will be achieved using alpha/numerical codes to anonymise all documentation and recordings once it has been generated. The original list of participants will be held by the primary researcher only. All research assistants/translators will also be briefed in matters of

51

anonymity and confidentiality. They will have access to primary data only when necessary and only in anonymous form – for example when discussing data or transcribing audio recordings. All documentation and recordings will be given a code according to the following: University initials and individual respondent number. These initials/numbers will have the date of completion or recording. Cases will be referred to by this code. The code would then look similar to the following: UFRJ:56:20.06.2013. Should exerts be taken directly from narratives as part of writing up the dissertation, or anecdotes used for the purposes of private or academic discussion or presentation then pseudonyms will be used.

As described in part three of this form, all written notes and questionnaires will be collected and stored in secure hand baggage with combination locks until they can be stored in a secure filing system on academic premises. If the questionnaires can be completed via survey monkey, information will be stored in my personal account which is password protected. In addition, all electronic recordings will be saved and stored on computerised equipment which are also password protected.

12. THIRD PARTY DATA

Will you require access to data on participants held by a third party? In cases where participants will be identified from information held by another party (for example, a doctor or school) describe the arrangements you intend to make to gain access to this information.

Although not essential to the study I will attempt to obtain data and general information relating to University admissions quotas and policy at each of the Universities that I conduct my research. If this information is not publicly available I will seek permission from the necessary department at each University and will follow any guidelines they wish me to follow with regards to the use and confidentiality of such information should access to be granted.

13. PROTECTION OF RESEARCHER (APPLICANT)

Please state briefly any precautions being taken to protect your health and safety. Have you taken out travel and health insurance for the full period of the research? If not, why not. Have you read and acted upon FCO travel advice (website)? If acted upon, how?

Measures to secure the well-being of the researcher and research assistant have been / will be carried out in a number of ways. I will ensure that various people are aware of my movements and whereabouts while residing in Brazil. In addition, I will be carrying upon my person a completed form containing personal details so that if an accident were to occur the correct people will be informed. I am able to speak Portuguese to an intermediate level and do not envisage any difficulties or risks occurring as a result of language barriers.

Any research assistants/translators will be either friends or acquaintances and are from the cities in which my research is to be conducted. Researcher has up-to-date vaccinations, comprehensive medical and travel insurance. Research is not to be completed in any dangerous settings and the researcher will not be exposed to any unfamiliar territory as a direct result of undertaking this research.

The carrying of baggage, electronic equipment and cash will be kept to a minimum so as to mitigate the chances of incurring criminal activity. Electronic copies of travel documentation and insurance cover will be available as well as access to additional credit if the researcher incurs any unforeseen financial outlays.

52

If deemed necessary, the UEA will be given the address and contact details of academic contacts in Brazil.

14. PROTECTION OF OTHER RESEARCHERS

Please state briefly any precautions being taken to protect the health and safety of other researchers and others associated with the project (as distinct from the participants or the applicant).

Any research assistants/translators will be local to the areas in which the research is to be conducted. No additional risks are thought to be apparent as a direct result of them helping with this study.

15. RESEARCH PERMISSIONS (INCLUDING ETHICAL CLEARANCE) IN HOST COUNTRY AND/OR ORGANISATION

The UEA’s staff and students will seek to comply with travel and research guidance provided by the British Government and the Governments (and Embassies) of host countries. This pertains to research permission, in-country ethical clearance, visas, health and safety information, and other travel advisory notices where applicable. If this research project is being undertaken outside the UK, has formal permission/a research permit been sought to conduct this research? Please describe the action you have taken and if a formal permit has not been sought please explain why this is not necessary/appropriate (for very short studies it is not always appropriate to apply for formal clearance, for example).

This study will be short in length taking place over designated days during a month long trip and therefore no formal permit will be sought.

16. MONITORING OF RESEARCH

What procedures are in place for monitoring the research/project (by funding agency, supervisor, community, self etc).

Research activities will be recorded on the day they have occurred. Due to the short amount of time this research will be conducted over it will not be necessary to send updates of progress to my supervisor or any third party.

17. ANTICIPATED USE OF RESEARCH DATA ETC

What is the anticipated use of the data, forms of publication and dissemination of findings etc.?

The data generated from this study will be used to produce an undergraduate dissertation and may be used in future essays of the researcher.

18. FEEDBACK TO PARTICIPANTS

Will the data or findings of this research/project be made available to participants? If so, specify the form and timescale for feedback. What commitments will be made to participants regarding

53

feedback? How will these obligations be verified?

Respondents will be made fully aware of my status as a UEA student who is currently conducting research and will be given an introduction to the research topic. The data generated from this research will not be made available directly to respondents or participating Universities; although, on completing of my dissertation I will produce a brief overview of my findings which could be offered to respondents and participating Universities alike.

No timescale, commitment or obligation has been discussed or put in place formally with any contacts thus far. However feedback will be given to respondents in a variety of ways. Firstly, the researcher will send interviewees (via email) recordings of their interviews for them to listen to within seven days of the interview taking place. In the email interviewees will be asked if they wish to withdraw, modify or develop information that they have given in their interview and if so to make it known to the researcher within thirty days of receiving the email.

In addition each participant and University will be offered an English copy of the dissertation once it is completed. If a Portuguese version is requested this could be supplied within 3 months; however the researcher is not in a position to translate the entire dissertation himself due to language barriers.

19. DURATION OF PROJECT

The start date should not be within the 2 months after the submission of this application, to allow for clearance to be processed. Start date End date 12th June 2013 12th July 2013 20. PROJECT LOCATION(S) Please state location(s) where the research will be carried out. Research sites are to be confirmed, however it is anticipated that two separate Universities in Rio de Janeiro will be visited.

54

Appendix-‐F: Questionnaire, Information Sheet, Interview Consent Form (English)

Research Questionnaire

This brief questionnaire contains questions to which I ask you to respond as subjectively as possible. Please read the questions carefully and remember that there are no “right” answers: I am interested in your view! Your responses will help to shed some light upon issues on which scholars and policymakers currently debate.

The questionnaire is anonymous. I would like to thank you very much for your participation.

1. What is your sex? ⃝ Female ⃝ Male

2. What is your nationality and home state? ___________________________________________________

3. Which race/ethnicity best describes you?

⃝ Yellow ⃝ Black ⃝ Mixed Race (Black) ⃝ Mixed Race (White) ⃝ Mixed Race (Mixed)

⃝ White ⃝ Indigenous ⃝ Other (Please Specify) _________________________________________

4. Do you find the above categories appropriate in distinguishing your racial/ethnic status? ⃝ Yes ⃝ No

Please comment

5. What is your age? ______

6. Which degree do you study and in which semester are you? ______________________________ _______

7. In your opinion, before the adoption of ‘Racial Quotas’, the University system in Brazil:

⃝ Produced less racial inequality ⃝ Produced more racial inequality ⃝ Had no significant effect either way

8. In your opinion, since its introduction in the university system ‘Racial Quotas’ have: ⃝ Reduced racial inequalities ⃝ Increased racial inequalities ⃝ Had no significant effect either way

9. Does your university currently have the system of ‘Racial Quotas’ ⃝ Yes ⃝ No

10. In your opinion, how effective is the ‘Racial Quotas’ system in reducing racial inequalities in the Brazilian University system?

55

⃝ 1 Highly ineffective ⃝ 2 ⃝ 3 ⃝ 4 ⃝ 5 Highly effective

11. The ‘Racial Quotas’ system attempts to reduce racial inequalities in access to University education; do you think it is a fair way in attempting to do this?

⃝ 1 Extremely Unfair ⃝ 2 ⃝ 3 ⃝ 4 ⃝ 5 Extremely Fair

Please read the following text, then answer question 12.

In 2000 the governor of RJ, approved a local law reserving half the available places in undergraduate courses in public universities based in RJ for students who had attended public high school. In 2001, the governor approved another law establishing that 40% of the vacancies under the 2000 law must be reserved for black and brown students. On April 26, 2012 the Brazilian Supreme Court unanimously ruled affirmative action policies can take a person’s race into account. The only national program designed and managed by the federal government to actively promote affirmative action is the “University for All,” ProUni, created in 2004. ProUni has adopted the quota system for candidates self-identified as black, brown or indigenous at the time of application for the scholarship. The percentage of scholarships reserved for those groups is determined by the percentage of these groups in the population of the candidate’s state (Lima, Revista, 2012).

12. Based upon the above text and thinking about what you already know, how would you rate your level of knowledge regarding the ‘Racial Quotas’ system?

⃝ 1 No knowledge ⃝ 2 ⃝ 3 ⃝ 4 ⃝ 5 Highly Knowledgeable

13. In your opinion, should the ‘Affirmative Action’ social policy be:

⃝ Abolished ⃝ Modified ⃝ Continued in its current format

If you commented

‘Modified’ Please

comment how so:

14. How would you compare the standard of living of your family with that of other families in Brazil?

⃝ 1 Very much lower ⃝ 2 Lower ⃝ 3 Average ⃝ 4 Higher ⃝ 5 Very much higher

15. Any other comments – Please feel free to note any additional comments you may wish to make; this could be regarding the survey itself, the questions or topics covered within the survey or indeed any other information you may wish to be considered.

56

Information Sheet Social Perceptions of the Affirmative Action Policy in Brazil

As part of my ‘International Development with Economics’ degree I have the option to complete a piece of independent research that will contribute towards my dissertation in my final year. With this in mind, I have decided to conduct some primary research regarding the ‘Affirmative Action’ social policy in Brazil, more specifically the way in which this policy is perceived by current university students.

I will be looking at a number of issues, including whether or not the policy is perceived to be effective in reducing inequality or deemed fair in practice. Much discussion surrounding such issues has already taken place, often producing divided opinions. I wish to compliment already existing literature regarding Affirmative Action in Brazil and around the world through generating primary data from students and professors who are currently enrolled / working in the University system which the policy targets.

Before you decide if you want to take part, please read this information sheet carefully and please feel free to ask any questions you may have.

What is the purpose of the study?

The main purpose of the study is to identify how the ‘Affirmative Action’ policy is perceived by current University students and professors, and what influences those perceptions. Generated data will be used as part of my final year dissertation.

Who is doing the study?

This study is led by Jonathan Stuart at the School of International Development, University of East Anglia, UK. Other members of the project are yet to be identified but all assistants / translators will be Brazilian nationals and privately sourced. Jonathan Stuart will have lead responsibility for the introduction to and distribution of questionnaires as well as key informant interviews; however, a research assistant / translator will be present also.

What would I be asked to do?

If you decide to take part in the study, you will be contacted by email to arrange a time and date that is convenient for you. You can choose to conduct the interview in person at a location in your university or via skype. You can stop the interview at any time, without having to explain why. We would like to audio record these interviews, if this is ok with you.

Is the research confidential?

Yes. Any information that you share with us will only be seen by members of the research team. We will not share it with other people. This means that whenever we write or talk about

57

anything you have told us we never use your real name. All information about you will be stored securely.

Unless you tell us otherwise, you will be understood to be speaking in a personal capacity, not as a representative of your organisation. We will not share the information you provide with other people in your organisation, unless you instruct us to do so.

How will the research be used?

The findings of this study will be reported anonymously in the lead researchers’ final year dissertation and may also be used in any additional essays or oral presentations completed by the lead researcher. Unpublished analysis will be emailed to you if it refers to information taken from your interview. You will be given 30 days to comment on, modify or withdraw the information you previously provided. Upon publication (or presentation), the finalised versions will be made available to everyone who agreed to participate in the study.

What are the benefits and risks of taking part?

Rather than view your participation as one-way, whereby we ask questions and you give answers, I plan to give an informative introduction to me and my research to all student participants; offering knowledge and advice from the perspective of a UK University student which they may find both interesting and useful.

Do I have to take part?

No, you do not have to take part in the study. If you do, it is up to you to decide what you want to tell us.

Will I receive anything for taking part?

No, we will not be able to pay you for taking part in this interview.

Can I have some more information before I decide?

If you would like to find out more before deciding, please let us know and we will be happy to help.

Ok, I’m happy to take part. Now what?

Once you have signed the consent form (see attached), I will contact you to arrange the visit / interview.

Who can you contact for more information?

If you have any questions or would like any more information before deciding to take part, please feel free to contact:

In the first instance Jonathan Stuart by email at [email protected] or by telephone at:

Pedro Bertacchini at [email protected] or by telephone at +55 (21)9158-0342 / 2527-6670

Thank you for reading this!

If you do decide to take part your help will be very helpful to us.

58

Social Perceptions of the Affirmative Action Policy in Brazil

Interviewee Consent Form

• I have read and understood the information sheet provided about this study.

• I understand that my participation in this interview is voluntary.

• I have the right to not answer any question I do not like or to stop the interview and

withdraw my answers, at any stage of the interview, without having to explain why.

• I understand that my answers will be kept confidential by the researchers and will only

be used for research purposes. My name will not be used in any research reports and

nothing will be published that might identify me.

• I agree to the interview being audio recorded YES / NO

• I understand that if I have any further questions I can contact one of the researchers

listed on the information sheet.

• I agree to some of my comments or statements being quoted in the report, provided

that I cannot be identified. YES / NO

• I would like to receive all publications and other publicly available material generated

by this study. YES / NO

Declaration:

I, ____________________________________________ agree to participate in an interview

for this study.

Signed: _____________________________(Participant) Date: ___/___/___

Signed: _____________________________(Researcher) Date: ___/___/___

59

Appendix-‐G: Questionnaire, Information Sheet, Interview Consent Form (Portuguese)

Questionário+da+pesquisa+

Este%breve%questionário%contém%perguntas%para%as%quais%pedimos%que%você%responda%da%forma%mais%subjetiva%possível.%Por%favor,% leia%as%perguntas%com%atenção%e% lembre@se%que%não%há% respostas% "certas":%estou% interessado%na%sua%opinião!%Suas%respostas%vão%ajudar%no%esclarecimento%sobre%questões%das%quais%estudiosos%e%politicos%atualmente%debatem.%

Este%questionário%é%anônimo%e%voluntario.%Eu%gostaria%de%agradecer%muito%a%sua%participação.%

1. Qual+é+o+seu+sexo?!!!!!!!!!!!!! ⃝!!!Feminino! ! ⃝!!!Masculino! !

+

2.++Qual+é+a+sua+nacionalidade+e+estado+de+origem?+________________________________________________+

3. Qual+raça+/+cor+melhor+descreve+você?+

+

!!!!!⃝!!!Amarela! !!!!!!⃝!!!Preto! !!!!!!!!⃝+++Pardo!(Negro)! ⃝!!!Pardo!(Branco)!!!!!⃝!!!Pardo!(Mestiço)!

!!

!!! !!!!!⃝!!!Branco! !!!!!!⃝+Indígena++!!!!!!!!!!!!!!⃝!!!Outro!(Por!favor!Especifique)!_________________________!!

4. Você+acha+que+as+categorias+acima+são+adequadas+para+distinguir+sua+raça+/+cor?!!!!!!⃝!!!Sim!!!!!!!!!!⃝+++Não+

+++++++Por+favor,++

+++++++Comente+

+

+

+

+

5. Qual+é+a+sua+idade?+_______!!

6. Qual+curso+você+estuda+e+qual+período?!________________________________________!!!!!______++

7. Na+sua+opinião,+antes+de+adotarem+as+“cotas+raciais+",+o+sistema+Universitário+no+Brasil:+

!

!!!!⃝+produzia!menos!desigualdade!racial!!!!! !⃝+produzia!mais!desigualdade!racial!!!

!!!!!⃝!Não!teve!papel!significativo!de!qualquer!forma!

!

8. Na+sua+opinião,+desde+a+sua+introdução++no+sistema+universitário+as+"cotas+raciais":+

⃝!Reduziram!as!desigualdades!raciais!!!! !⃝!Aumentaram!as!desigualdades!raciais!!!!

⃝!Não!tem!efeito!significativo!de!qualquer!forma!

9. A+sua+Universidade+atualmente+tem+um+sistema+de+cotas+raciais?++++++++++⃝!Sim!!!!!!!!! ! ⃝+Não!!

10. +Na+sua+opinião,+o+quão+eficaz+é+o+sistema+de+"cotas+raciais"+na+redução+das+desigualdades+raciais+no+sistema+universitário+

brasileiro?+

++++++⃝!1!Altamente!ineficaz!!!!!!!!!!!!!!!!!!!!!!⃝!2!!!!!!!!!!!!!!!!!!!!⃝!3!!!!!!!!!!!!!!!!!!!!⃝!4!!!!!!!!!!!!!!!!!!!!!!⃝!5!Altamente!eficaz!

!

!

+

+

60

11.#O#sistema#de#"cotas#raciais"#tenta#reduzir#as#desigualdades#raciais#no#acesso#ao#ensino#de#universitário;#você#acha#que#é##

uma#forma#justa,#na#tentativa#de#fazer#isso?#

#! ! #!!!!!!!⃝!1!Extremamente!injusto!!!!!!!!!!!!!!!⃝!2!!!!!!!!!!!!!!!!!!!!⃝!3!!!!!!!!!!!!!!!!!!!!⃝!4!!!!!!!!!!!!!!!!!!!!!!⃝!5!Extremamente!justo!!Por#favor#leia#o#texto#abaixo#e#responda#a#pergunta#12##

Em!2000,!o!governador!do!RJ!aprovou!uma!lei!local!reservando!metade!das!vagas!em!cursos!de!graduação!em!universidades!públicas!no!RJ!para!alunos!que!freqüentaram!escolas!públicas.!Em!2001,!o!governador!aprovou!outra!lei!que!estabeleceu!que!40%!das!vagas!nos!termos!da!lei!de!2000!deveriam!ser!reservadas!para!estudantes!negros!e!pardos.!Em!abril!2012,!o!Supremo!Tribunal! Federal! decidiu!por!unanimidade!que!o! sistema!de! cotas!pode! levar! a! raça!de!uma!pessoa!em!consideração.! Era!único!programa!nacional,!até!o!ano!passado,!concebido!e!gerido!pelo!governo!federal!para!promover!ativamente!o!sistema!de!cotas!é!o!"Universidade!para!Todos",!ProUni,!criado!em!2004.!O!ProUni!adotou!o!sistema!de!cotas!para!candidatos!autoPidentificados!como!preto,!pardo!ou!indígena!no!momento!da!aplicação!para!as!bolsas.!O!percentual!de!bolsas!reservadas!para!esses!grupos!é!determinado!pela!percentagem!desses!grupos!na!população!do!estado!do!candidato!(Lima,!Revista,!2012).!!

12.# Com#base#no#texto#acima,#e#pensando#no#que#voce#ja#sabe,#como#você#classificaria#o#seu#nível#de#conhecimento#sobre#o#

sistema#de#"cotas#raciais"?#

!!!!!!!⃝!1!Nenhum!conhecimento!!!!!!!!!!!!!⃝!2!!!!!!!!!!!!!!!!!!!!⃝!3!!!!!!!!!!!!!!!!!!!!⃝!4!!!!!!!!!!!!!!!!!!!⃝!5!Bastante!conhecimento#

13. #Na#sua#opinião,#o#sistema#de#'cotas#raciais'#deve#ser:#

!!!!!!!⃝!Abolido! !!!!!⃝!Modificado!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!⃝!Continuar!em!seu!formato!atual!

#####Por#favor,##

#####Comente#

#

#

#

14. #Como#você#compararia#o#padrão#de#vida#de#sua#família#com#a#de#outras#famílias#no#Brasil?###

!!!!!!!⃝!1!Muito!menor!!!!!!!!!!!!!!!⃝!2!Menor!!!!!!!!!!!!!!!!⃝!3!Igual!!!!!!!!!!!!!!!⃝!4!Melhor!!!!!!!!!!!!!!!⃝!5!Muito!melhor!

15. #Quaisquer#outros#comentários# P#Por#favor,# fique#à#vontade#para#oferecer#qualquer#comentário#adicional#que#você#possa#querer# fazer,# o# que# poderia# ser# sobre# a# própria# pesquisa,# o# tema# abordado# na# pesquisa# ou# mesmo# qualquer# outra#informação#que#você#considere#relevante.#

#

!

!

!!!!!!!!!!!! ! ! ##

61

Ficha de Informação Percepções sociais da política da ação afirmativa (cotas raciais) no Brasil

Como parte do meu curso de 'Desenvolvimento Internacional e Economia " tenho a opção de concluir um trabalho de pesquisa independente, que vai contribuir para a minha monografia no último ano. Com isto em mente, eu decidi realizar uma pesquisa preliminar sobre o sistema de "cotas raciais" no Brasil, mais especificamente, a maneira pela qual essa política é percebida pelos estudantes universitários atuais.

Estarei olhando para uma série de questões, incluindo como a política é percebida, se eficaz ou não na redução da desigualdade ou considerada justa na prática. Muita discussão em torno dessas questões já aconteceu, muitas vezes produzindo opiniões divididas. Gostaria de complementar a literatura já existente sobre cotas raciais no Brasil e ao redor do mundo através da geração de dados primários de alunos e professores que estão atualmente matriculados / trabalhando nas Universidades aonde este sistema se aplica/nao aplica.

Antes de decidir se você quer participar, por favor leia esta folha de informação com cuidado e sinta-se livre para tirar qualquer dúvida que possa ter.

Qual é o objetivo do estudo?

O principal objetivo do estudo é identificar como o sistema de "cotas raciais" é percebido pelos estudantes universitários e professores, e o que influencia esta percepção. Os dados gerados serão utilizados como parte de minha monografia de final de curso.

Quem está realizando o estudo?

Este estudo é liderado por Jonathan Stuart da Escola de Desenvolvimento Internacional da Universidade de East Anglia, no Reino Unido. Outros membros do projeto ainda estão para ser identificados, mas todos os assistentes / tradutores serão cidadãos brasileiros, participantes de forma particular. Jonathan Stuart terá a responsabilidade principal pela introdução e distribuição de questionários, bem como entrevistas com informantes-chave, no entanto, um assistente / pesquisador, tradutor estará presente também.

O que vou ser convidado a fazer?

Se você decidir participar no estudo, em contacto comigo por e-mail para organizar a hora e data que seja conveniente para você. Você pode optar por realizar a entrevista pessoalmente em um local em sua universidade, um outra lugar ou via skype. Você pode interromper a entrevista a qualquer momento, sem ter que explicar o porquê. Gostaría de gravar com áudio esta entrevistas, se isso é ok com você.

A pesquisa é confidencial?

Sim. Qualquer informação que você compartilhar com a gente só vai ser vista pelos membros da equipa de pesquisa. Não vou compartilhá-la com outras pessoas. Isto significa que sempre que escrever ou falar sobre qualquer coisa que você nos disse, nunca vou usar o seu nome real. Todas as informações sobre você serão armazenadas de forma segura.

A menos que você me dizer o contrário, entendemos que você esta falando de maneira pessoal, não como um representante de sua organização. Eu não ira compartilhar as

62

informações que você fornecer com outras pessoas em sua organização, a menos que você nos instruir a fazê-lo.

Como a pesquisa será usada?

Os resultados deste estudo serão relatados anonimamente no trabalho de monografia do pesquisador e também podem ser utilizados em quaisquer ensaios adicionais ou apresentações orais cumpridos pelo pesquisador. A análise inédita será enviada para você referente a informações extraídas de sua entrevista. Você terá 30 dias para comentar, modificar ou retirar a informação fornecida anteriormente. Após a publicação (ou apresentação), as versões finais serão disponibilizadas a todos os que concordaram em participar do estudo.

Quais são os benefícios e riscos de tomar parte?

Ao invés de ver a sua participação somente como, perguntas e respostas, eu pretendo dar uma introdução informativa sobre mim e sobre a minha pesquisa a todos os alunos participantes, oferecendo conhecimentos e conselhos da perspectiva de um estudante de uma Universidade no Reino Unido o que pode ser interessante e útil.

Eu tenho que participar?

Não, você não tem que tomar parte no estudo. Se você fizer isso, cabe a você decidir o que você quer nos dizer.

Vou receber algo em troca pela minha participação?

Não, eu não ira pagá-lo para participar desta entrevista.

Posso ter mais algumas informações antes de decidir? Se você gostaria de saber mais antes de decidir, por favor, avise-nos e teremos o maior prazer em esclarecer.

Ok, eu estou disposto a participar. Agora o quê?

Depois de ter assinado o termo de consentimento (ver anexo), entraremos em contato para marcar a visita / entrevista.

Com quem você pode entrar em contato para obter mais informações?

Se você tiver alguma dúvida ou quiser mais informações antes de decidir participar, por favor não hesite em entrar em contato:

Em primeira instância Jonathan Stuart pelo email [email protected] ou pelo telefone: +55 (21) 9160-3073

Pedro Bertacchini pelo email [email protected] ou pelos telefones:+55 (21)9158-0342 / 2527-6670

Obrigado pela sua Atenção!

Se você decidir participar, sua ajuda será muito útil para nós.

63

Percepções sociais da política da ação afirmativa (cotas raciais) no Brasil Termo de Consentimento do Entrevistado

• Eu tenho lido e compreendido a ficha de informações fornecida sobre este estudo..

• Eu entendo que a minha participação nesta entrevista é voluntária.

• Eu tenho o direito de não responder a qualquer pergunta que eu não gostar ou de

parar a entrevista e retirar as minhas respostas, em qualquer fase da entrevista, sem

ter que explicar o porquê.

• Eu estou ciente de que minhas respostas serão mantidas em sigilo pelos

pesquisadores e serão usadas apenas para fins da pesquisa. Meu nome não será

utilizado nos relatórios de pesquisa e nada será publicado que possa me identificar.

• Eu concordo com a entrevista sendo gravada em áudio SIM / NÃO

• Eu compreendo que se eu tiver alguma dúvida posso contactar um dos

pesquisadores listados na folha de informação.

• Eu concordo com alguns dos meus comentários ou declarações serem citados no

relatório, desde que eu não possa ser identificado SIM / NÃO

• Eu gostaria de receber todas as publicações e outros materiais à disposição do

público gerado por este estudo. SIM / NÃO

Declaração:

Eu, ____________________________________________ concordo em participar de uma

entrevista para este estudo.

Assinatura: _____________________________(Participante) Data: ___/___/___

Assinatura: _____________________________(Pesquisador) Data: ___/___/___

Dissertation_Final Draft_LI

Documents

Transcript of Dissertation_Final Draft_LI