Private Tutoring in English for Secondary School Students in Bangladesh

Private Tutoring in English for Secondary School Students in Bangladesh As English continues to spread across the globe, governments in “low profi ciency” English countries Graddol, 20

Private Tutoring in English for

Secondary School Students in Bangladesh

As English continues to spread across the globe, governments in

“low profi ciency” English countries (Graddol, 2006, p 110) are reappraising its importance and responding to the changing global order of English by updating their language and language-in-education policies (Tsui & Tollefson, 2007) It often happens that such responses

result in English gaining more space in the national curriculum: English

may be made a compulsory subject, where before it was only an

elec-tive; or it may be introduced in earlier grades, following the principle

of “the earlier, the better” (Medgyes, 2005, p 276) The limited space

in the curriculum (Kaplan, 2000) is thus forced to make room for more

English, and more resources are reallocated to promote its teaching

and learning, at the potential expense of other components of the

curriculum

It is not yet known whether extending the space for English in the

curriculum and increasing budgetary allocations for English teaching

actually deliver the desired effects Parents in many countries have

pre-empted this question by investing in private tutoring in English (PT-E)

for their children Whether this private investment in children’s English

learning pays off, and whatever the attitudes and perceptions of those

who participate in PT-E, the scale of its popularity worldwide suggests

that it is either successful or considered to be so (Greaney & Kellaghan,

1995, p 12)

PRIVATE TUTORING IN THE LITERATURE

PT-E is a special and important subclass of private tutoring (PT) PT,

or after-hour supplementary schooling, is a macro-phenomenon of

mod-ern education Recently, it has grown dramatically in Asia, Africa, Europe,

and North America (Baker, Akiba, LeTendre, & Wiseman, 2001; Bray,

2003, 2006; Ireson, 2004) This “shadow education” (Bray, 2003, p 21)

sits at the interface between education and commerce and has become

a multibillion-dollar industry (Aurini, 2004; Davies, 2004; Mischo &

Haag, 2002; Stevenson & Baker, 1992) Family investment in private

tutoring in wealthier East Asian countries like Korea, Japan, Hong Kong,

and Taiwan amounts to billions of dollars annually (Bray, 1999; Lee,

2005; Stevenson & Baker, 1992) Research suggests that in the Anglophone

and European countries PT tends to be principally remedial and

supple-mentary in nature (Mischo & Haag, 2002; Davies, 2004) In parts of East

Asia, however, it has established a position as an essential part of

educa-tion, culture, and society (Foondun, 2002; Kwok, 2004) This trend is

evidenced by the strong presence of juku, or private after-hour classes, in

Japan (Harnisch, 1994; Roesgaard, 2006; Rohlen, 1980) Private

tutor-ing is also embedded in South Korean society, where it is known as

ha-gwuan, or cram schools; US$12.4 billion was spent on this form of

education in 2003, equivalent to 56% of the country’s national

educa-tion budget (Lee, 2005) Private tutoring is an integral part of educaeduca-tion

in Hong Kong, which also consumes substantial family resources (Bray &

Kwok, 2003) Research shows that a complex set of factors—including

the value of education in the Confucian tradition, the parental role in

educating children, the system of high-stakes examinations for sion to higher education, the role of education in the job market, eco-nomic prosperity, and the high priority given to these factors in the overwhelming majority of middle class families—account for the popu-larity of PT in East Asia Not all these factors hold equally in less affl uent societies; nevertheless, PT is popular and well-subscribed in these societ-ies as well (Bray, 2006; Buchmann, 2002; Greaney & Kellaghan, 1995; Tansel & Bircan, 2006)

Despite some progress in research on PT, this phenomenon still

“remains very fi rmly in the shadows” (Ireson, 2004, p 110) Moreover, the existing research on PT mostly addresses issues related to its nature, scope, scale, and motivation Only a few studies have been directed at its evaluation or effectiveness, and they have reported only inconclusive

fi ndings (Bray, 2006) Furthermore, the small body of PT research has largely concentrated on relatively affl uent societies We know much less about the nature, practice, impact on achievement, and socio-educational implications of PT in less affl uent societies This article responds to this need in investigating the nature and practice of PT, and specifi cally PT-E, in a region, a country, and a context at the opposite end

of the affl uent scale: a disadvantaged rural area in Bangladesh

Research that exclusively focuses on PT-E, which is of substantial est to TESOL, applied linguistics, and language education, is apparently restricted to only two studies The fi rst, by Mischo and Haag (2002), eval-uates the effectiveness of PT in Germany in a pre– and post–control-group design The study involved 122 students in a PT group and the same number in a control group Both groups’ school marks in English, French, Latin, and mathematics in the pretest and the posttest were recorded and compared T-tests showed that the posttest scores of the PT group in all four subject areas were signifi cantly higher than the pretest

inter-scores ( p < 0.001), though whether this should be attributed to their

par-ticipation in PT is not certain

The other study on PT-E is Khuwaileh and Al-Shoumali (2001), which explores the reasons and conditions which contribute to the success of PT-E in Jordan Fifty students from two Jordanian universities partici-pated in a questionnaire survey These students and a sample of 10 par-ents were subsequently interviewed The study records positive attitudes

on the part of parents and students toward PT-E It also shows that the prevalence of PT-E was mainly due to such factors as the poor teaching of English at the school level, the students’ desire for good grades in their tertiary courses, the importance of English, family affordability, and fam-ily social class and prestige

This paucity of research on PT-E does not match the scale of its

popu-larity English is one of the key subjects in juku , which is an integral part

of the high-stakes tests in Japan (Locastro, 1990) In Hong Kong, “English

typically has the greatest demand [in private tutoring] because it is

impor-tant not only as a subject but also as a medium of instruction for other

subjects” (Bray & Kwok, 2003, p 614) English is also one of the most

commonly taught PT subjects in Korea (Lee, Kim, & Yoon, 2004) In

short, English constitutes an essential component of the national

curricu-lum in many countries, and thus it has a major position in the PT market

Poor EFL results in these countries (Baldauf et al., 2008) provide an

addi-tional motivation for PT-E

In the current study we focus on PT-E as a substantial area of English

language teaching and learning We describe the phenomenon of PT-E

in Bangladesh; relate PT-E to scholastic achievement in English;

investi-gate student attitudes and motivations in PT-E; and start to build an

over-all profi le of PT-E in relation to the school system, parent and student

expectations, attitudes and motivations, and outcomes These objectives

are important for TESOL and language education researchers and

prac-titioners worldwide, because they draw attention to the question of how

TESOL pedagogy is implemented, accommodated, and supplemented

by PT-E in less affl uent societies

BANGLADESH

Bangladesh is a medium-sized country in South Asia with a substantial

population of 140 million The annual per capita income is US$411; the

literacy rate of the population aged 7 and over is 45.3% (Bangladesh

Bureau of Educational Information and Statistics [BANBEIS], 2004)

Bangla (Bengali) is spoken by 98% of the population in Bangladesh

English was introduced during British rule and still enjoys a position of

dominance and prestige (Banu & Sussex, 2001; Kachru, 2005) English is

taught as a compulsory subject from the 1st to the 12th grade and also at

the tertiary level

PRIVATE TUTORING IN BANGLADESH

PT is a common phenomenon in Bangladeshi education It is widely

spread in urban as well as rural areas, and shows some complex

charac-teristics which are to some extent different from its implementation in

East Asia, Europe, or North America

Private tutoring in Bangladesh is delivered by mainstream teachers,

teachers from other institutions, or even nonteachers Lessons occur

one-to-one, in small groups (5–10 students), or in large groups (20 students

or more) The venue of the teaching can be the tutor’s place of residence

(to-one or small groups), the tutee’s place of residence (usually

one-to-one), or special teaching centres (large groups) known as coaching

centres These coaching centres are comparable to Japanese juku, or

pri-vate after-school classes Finally, in our study the focus of PT is on based examinations or school-leaving national examinations rather than college or university entrance examinations, although PT is equally prom-inent for these latter goals in higher education in Bangladesh

PT is widely implemented at all stages of pretertiary education in

Bangladesh Education Watch 2006 (Campaign for Popular Education

[CAMPE], 2007) reported that PT participation rate at the primary level averaged 37% However, at 86.5% it was predictably much higher

at the secondary level The PT participation rate at the secondary level

in our study is somewhat lower (75%), consistent with the location of the study in a rural disadvantaged area One crucial factor for the pop-ularity of PT in Bangladesh is inadequate public investment in educa-

tion Education Watch 2006 shows that the public investment per student

per month is US$2.05 at the primary level and US$4.50 at the ary level This investment accounts for 41% and 29% of the total edu-cation expenditure at the two levels, respectively The low investment accounts for poor infrastructure and logistics, which in turn, contrib-ute to the poor quality of teaching and learning (Hamid & Baldauf, 2008)

Poor teacher salaries are another reason why school teachers offer vate after-hour lessons to their own students for extra income Although ethical issues are involved here (Nuland, Khandelwal, Biswal, Dewan, & Bajracharya, 2006), PT is a common practice all over Bangladesh Mainly teachers of English, mathematics, and sciences have the opportunity to earn this extra income through PT There are tensions between these teachers and those whose subjects are not in demand for PT (e.g., Bangla and the social sciences)

The cost of PT is hard to calculate, because it depends on the quency of lessons taken, the duration of the lessons, the subjects taught, teacher quality and reputation, the mode of lessons (one-to-one/groups), and the geographic location (urban/rural) of teachers and students

Education Watch 2006 (CAMPE, 2007) shows that the annual average cost

for PT per student was US$49 at the secondary level Importantly, the PT cost constitutes a signifi cant proportion (42% for government schools; 29% for nongovernment schools) of the total education cost borne by families, which is substantially higher than the cost of any other items ( Table 1 ) On average, the proportion (35.5%) is higher than the total contribution of public expenditure per student at the secondary level, which is 29%

The wide popularity of PT in Bangladesh prompts questions about its quality and societal perceptions Although there have been no objective studies exploring these issues, such perceptions appear to be complex and mixed Negative evaluations of PT prevail in the literature, but it is

widely believed that PT participation infl uences student performance in

school and school-leaving examinations For example, Deabnath (2007)

analyzes the results of the Secondary School Certifi cate (SSC) 1

examina-tion in 2007, and observes:

This year’s Secondary School Certifi cate (SSC) exam results warn of

the country’s secondary school education system becoming dependent

on private tuition as they show students of the urban areas who can afford

private tutors fared well while students in the rural areas suffered

dearly (n p.)

However, PT, particularly in its institutional form (i.e., coaching centres),

is questioned on the grounds of poor quality of teaching and its business

orientation (Chakroborty, 2002) One-to-one or small-group PT, on the

other hand, has the potential to offer quality remedial lessons because

the teaching conditions in this case differ clearly from those of formal

schooling In formal Bangladeshi schools, class sizes are commonly

larger than 50, which compromises individualized instruction and even

1 The Secondary School Certifi cate (SSC) examination is the fi rst public examination in the

Bangladeshi education system Students have to sit this examination at the end of Year 10

Successful performance in this examination allows students to study at the higher

second-ary level for 2 years before they can undertake undergraduate studies

TABLE 1 Item-Wise Average Annual Expenditure Per Secondary Student (Rural-Urban and Male-Female

Combined) by Type of Institution (in Taka and in Percent of Total Cost)

Source: Campaign for Popular Education (2007, p 51); reprinted with permission.

Note US$ 1 = 70 Taka approximately; Government secondary schools are fully funded by the

gov-ernment Nongovernment schools are not private schools because they are substantially funded

by the government in the form of teacher salaries and school infrastructure; tiffi n is a snack

taken by students as a substitute for lunch at school

Types of institutions

Government schools

Nongovernment schools

interactions with and feedback from the teacher Resources are poor in terms of books, study aids, and other support, whereas the conditions of individual or even group PT are qualitatively and quantitatively distinctly more favourable

The Current Study

The current study is based on PT-E data from Hamid (2009), a ranging investigation of English language teaching and learning in a rural context in Bangladesh, specifi cally the relationships between sec-ondary school students’ family economic, cultural, social capital, and their scholastic achievement in English The main study was conducted

wider-in a disadvantaged area—a medium-sized subdistrict wider-in Bangladesh Located about 500 kilometres from the capital city of Dhaka, Nadiranga (a pseudonym adopted for ethical reasons and anonymity) is a rural agriculture-based subdistrict with a small town at its centre The total land area of the subdistrict is 224.97 square kilometers, and its popula-tion is 114,350 (56,506 males, 57,844 females; BBS, 2005) It is one of the poorest subdistricts in Bangladesh (Rahman, Asaduzzaman, & Rahman, 2005) The literacy rate of the population aged 7 and over is 33.47% (cf the national rate of 45.3%) There are 12 secondary schools

in the subdistrict, from which around 700 tenth graders participate in the Secondary School Certifi cate examination every year English con-stitutes 2 of the 11 papers which all examinees must sit in this national examination Although English is emphasized in the curriculum, its teaching and learning in rural areas like Nadiranga are supported by only very limited resources English is not required in everyday commu-nication Students do not have access to self-study resources such

as print or electronic materials other than the set textbooks ers and the Internet, which could potentially introduce learners to virtual communities, are yet to be available in schools or homes in Nadiranga

THE MAIN STUDY: PARTICIPANTS AND INSTRUMENTS

The sample for the quantitative phase of the parent study consisted of

228 tenth-grade students (15–16 year olds) who were selected from eight schools in Nadiranga Fourteen of these 228 students were later selected for one-to-one interviews in its qualitative phase There were three instruments used for collecting numerical data: (1) a student survey ques-tionnaire, (2) an English profi ciency test which had been designed by an English Teaching Task Force (BEERI, 1976) commissioned by the Ministry

of Education of the Bangladesh Government, and (3) school records of

students’ grades in English in the school test as well as the SSC

examina-tion in 2007

Survey Questionnaire

The questionnaire (see Hamid, 2009) collected data on students’

family economic, cultural and social capital, and attitudinal and

motiva-tional characteristics and dispositions (habitus) It was administered,

together with the English test, to students in the schools in October–

November 2006 Appointments were made with the head teachers of the

schools On the appointed days the fi rst author met the 10th-grade

stu-dents in their class, briefed them about the study, and requested their

participation They were then given the profi ciency test, followed by the

questionnaire

English Profi ciency Test

This 60-point test is a national test designed in Bangladesh, and

has been widely tested there in different schools for reliability and

valid-ity (Zaman, 1999) Divided into three parts, the test content focuses

on grammar items in scripted dialogues (30 points), vocabulary items

(20 points), and reading comprehension questions (10 points)

Student Interview

After the profi ciency test and the questionnaire survey, 14 students

were interviewed at their homes in December 2006 The interviewees

were selected on the basis of their performance in the English profi ciency

test and in the English papers of the school test Our aim was to interview

both the best performing and the lowest performing students in both

tests in order to determine whether the two types of students differed in

terms of their habitus (i.e., a set of attitudes, motivations, values, career

goals and expectations, and academic self-concepts) and family

back-ground factors (e.g., parental education, occupation, income, and

paren-tal involvement in their education) The interviews were semistructured,

and most of the questions focused on aspects of family social and cultural

capital and the students’ attitudes, motivations, English-learning

experi-ences, future education plans, and career goals, and their perceptions of

2 The school test is the SSC-qualifying test arranged by secondary schools The schools

nomi-nate candidates for the SSC examination based on their performance on this test

the infl uence of their family situations on their academic performance (for interview questions, see Hamid, 2009), with participation in PT-E a prominent topic The interviews were conducted by the fi rst author, and took place in the homes of the students, lasting between 40 and 60 min-utes All interviews were audio-tape recorded with the students’ and their parents’ permission The interviewer also took notes in case there were technical problems in retrieving the recorded interviews All interviews were conducted in Bangla, the L1 of the interviewer and the interviewees

PT-E DATA FOR THE CURRENT STUDY

In the main study (Hamid, 2009), private tutoring was shown to be one source of cultural capital (Buchmann, 2002) that families exploit to enhance their children’s achievements in English Over 75% of the 10th graders who participated in the study took private lessons in English out-side mainstream schooling This high rate of participation in PT-E, par-ticularly in a disadvantaged area, obliges us to investigate the relationship

of PT-E to academic achievement in English and to ask why so many dents are attracted to it In addition, PT-E emerged as a dominant issue

stu-in stu-interviews with students selected from the survey sample The students had clear and patterned views about PT-E, issues which are, so far, unre-searched in the literature We therefore isolated all PT-E data–both quan-titative and qualitative—from the main study in order to carry out a separate, in-depth analysis in this article This analysis has a dual focus:

on the possible links between private tutoring and academic ment, and student perceptions of PT-E, which is designed to reveal whether PT-E does indeed deliver clear benefi ts, and to probe what other factors might explain the large-scale participation in PT-E

Methodology and Data Analysis

The dual focus of the paper involves a mixed-methods approach (Bryman, 2006; Creswell & Plano Clark, 2007; Greene, 2008; Johnson & Onwuegbuzie, 2004; Tashakkori & Teddlie, 1998; Teddlie & Tashakkori, 2003) This methodology calls for the collection, analysis, and mixture of both quantitative and qualitative data at different stages of the research process This research design is based on the “premise that the use of quantitative and qualitative approaches in combination provides a better understanding of research problems than either approach alone” (Creswell & Plano Clark, 2007, p 5) The growing fi eld of mixed method-ology suggests various rationales, objectives, and strategies for mixing quantitative and qualitative data in a single study (Greene, 2008; Greene,

Caracelli, & Graham, 1989; Johnson & Onwuegbuzie, 2004; Creswell &

Plano Clark, 2007) The literature also includes a number of typologies

for mixed-methods research design (Creswell & Plano Clark, 2007;

Johnson & Onwuegbuzie, 2004; Teddlie & Tashakkori, 2006)

Our approach to integrating quantitative and qualitative data in the

present paper benefi ted from the set of mixed methods designs

pre-sented in Creswell and Plano Clark (2007) They discuss four types of

mixed-methods designs—triangulation, embedded, explanatory, and

exploratory The design types include three dimensions: (a) time of

mix-ing the qualitative and quantitative phases (concurrent or sequential),

(b) weighting (equal or unequal weight of the phases), and (c) type of

mixing (merging, embedding, and connecting) Our methodological

integration was mainly based on their fourth design type (exploratory),

with explanation and complementation as two objectives or principles

for the mixing Our design was guided by the belief that different aspects

of a single phenomenon require different methods of inquiry, and the

fi ndings complement each other to yield a broader and deeper

under-standing of the phenomena under investigation (Greene, 2008) Our

data analysis and results are presented in two distinct phases—a

quantita-tive phase, followed by a qualitaquantita-tive phase, and are “integrated during the

interpretation of the fi ndings” (Johnson & Onwuegbuzie, 2004, p 20)

Thus, among the three models of mixing suggested by Creswell and Plano

Clark (2007)—merging, connecting, and embedding—we follow the

sec-ond method: we connect the two data sets, and the connection is

sequen-tial, because the two phases are chronologically ordered in the research

process In terms of the relative weight of the two phases, we depart from

much customary practice in that our notation is quan→QUAL: We place

more weight on the second, qualitative phase In our view, the qualitative

phase, which investigates students’ attitudes, motivations, and

percep-tions of PT-E, yields more and richer information than the quantitative

phase, which associates students’ participation in PT-E with English

achievement data The two phases of our study are “broadly

complemen-tary, providing different kinds of insights into the different aspects”

(Brannen, 2005, p 180) of the social phenomenon of PT-E

QUANTITATIVE DATA ANALYSIS AND FINDINGS

The students’ participation in PT-E was assessed using a dichotomous

yes or no response in the questionnaire The participation rate for girls

(80%) was higher than that for boys (71%), although the difference was

not statistically signifi cant on a χ 2 test

Table 2 presents percentage distributions of the students’ ( n = 228)

parental occupation, education, and family income levels The monthly

income of almost half of the families (47%) was below Taka 2000, and

only 23% of the families earned Taka 5,000 or more per month The income levels of the families point to the poor socioeconomic conditions

of the subdistrict Nonetheless, over three-quarters of the students ticipated in PT-E

As previously noted, the study includes two measures of English ment: students’ scores in the 60-point English test, and their scores in English in the SSC examination, 2007 The latter data were collected from the schools The SSC English scores were available only in the form

achieve-of letter grades, which were converted into number grades following the grading system used in the SSC examination This grading system was also used to convert the English test scores, which were raw scores, into number grades As can be seen from Figure 1 , student representation in the upper grades (3 or above) was very modest in both sets of grades Therefore, students obtaining these grades were regrouped into one grade (grade 3) for comparison purposes The χ 2 test assessing the rela-tionship between the categories of profi ciency test grades and SSC grades shows a signifi cant statistical association between the two measures of English achievement (χ 2 = 101.61, df = 9, p < 0.0001)

Figure 1 shows that the majority of the students either failed or obtained only lower grades (1 = D, 2 = C) in English Although three stu-dents were able to obtain the highest grade (5 = A+) in the English test, none did so in the SSC examination Furthermore, 37% of the students

in the present sample failed in the SSC examination in 2007, which is higher than the national rate of failure for that year (30%) Thus, these

TABLE 2 Percentage Distribution of Parental Background Characteristics

Parental income per month (Taka)

data confi rm the lower educational standards and English achievement

of students in rural schools (see Hamid & Baldauf, 2008)

Multivariate logistic regression, specifi cally ordered logistic regression, 3

was used to investigate factors that were independently associated with

the SSC English grades and the English test grades The results are

pre-sented in the form of odds ratios (OR) and their confi dence intervals

(CI) OR values greater than 1 indicate a probability of higher grades in

English; OR values less than 1 indicate a probability of lower grades in

English The specifi c predictor variables are PT-E, gender, and mother’s

education

We included gender as a potential predictor in order to investigate

whether gender had any infl uence in the relationship between the

students’ participation in PT-E and their academic achievement in

English We were interested in this variable because other studies have

shown that in Bangladesh the female enrolment rate in secondary

educa-tion is higher than that of males, although females were still behind

3 Since the achievement variables were ordered, ordered logistic regression was used to fi nd

possible associations between the students’ grades in English and the selected predictors

The implementation of ordered logistic regression was based on the proportional odds

model, which for each independent variable estimates a single parameter to describe the

relationship between each pair of ordered categories Father’s occupation, education, and

parental income (see Table 2) were dropped from the analysis because of their substantial

association with mother’s education

FIGURE 1 Comparison of Students’ SSC English Grades and English Test Grades

Source : Hamid and Baldauf (2008, p.19), reprinted with permission.

males in terms of achievement, particularly in rural areas (CAMPE, 2007; Raynor, 2005)

As displayed in Table 3 , multivariate logistic regression analysis revealed that

students who took private lessons had double the chance of attaining

•

the SSC examination than their female counterparts (OR = 1.6, p =

0.087), although this result was not statistically signifi cant;

neither gender nor mother’s education was found to be signifi cantly

•

associated with grades in English in the SSC examination;

for student achievement in the English test, the other measure of the

•

dependent variable, only PT-E (among gender, mother’s education, and private tutoring) was found to be signifi cantly associated Students who had access to PT-E were 2.8 times more likely to secure higher grades in the English test than those who did not have such access (95% CI = 1.72–5.03)

The regression analysis thus shows a positive association between PT-E participation and English achievement, after adjustments for mother’s education and gender However, each model represented approximately

TABLE 3 Ordered Logistic Regression Analysis of the Association of English Grades and Specifi c

Predictors

Note Value of 1.0 = reference category * 0.05 ** 0.001 *** 0.0001

Predictors

Outcome variables Secondary School Certifi cate

English English profi ciency test

Odds ratio

95% Confi dence interval Odds ratio

95% Confi dence interval Private tutoring

Yes 2.13* 1.12–4.04 2.82*** 1.72–5.03 Gender

10% of the total variability, which limits the validity of the fi ndings We

therefore analyzed the qualitative data, in order to explore possible

link-ages between factors operating in PT-E in a more fi ne-grained manner

from the students’ subjective perspective

QUALITATIVE DATA ANALYSIS AND FINDINGS

The interpretive phase of the study aimed to construct, describe, and

analyze the interviewees’ perceptions of PT-E We drew on Potter’s (1996)

template for qualitative data analysis, supplemented by the data

manage-ment and analysis methods outlined in Huberman and Miles (1998),

Miles and Huberman (1994), and interpretive or qualitative content

analysis (Ahuvia, 2001; Hsieh & Shannon, 2005) In particular, we

fol-lowed Hsieh and Shannon’s (2005) “conventional approach” (p 1279)

to qualitative content analysis Instead of applying predetermined

catego-ries, this approach generates categories from the data and is thus

described as inductive category development

We fi rst present a summary of the interviewees’ participation in PT-E

and the different measures of their English achievement in Table 4

The table serves as a quick reference to individual interviewees, their

family capital profi les, 4 the extent to which they took private lessons in

English, and the different measures of their performance in English

The information about PT-E presented in a descriptive form both verifi es

and extends the yes/no responses in the quantitative phase of the study

Eleven out of 14 interviewees participated in PT-E to varying degrees

This participation rate (78%) is close to that (75%) for the survey

sample

Table 4 shows that male and female students were from a range of

fam-ily capital backgrounds, from high to low; both groups of students

partici-pated in PT-E to varying degrees, and were high as well as low achievers

In other words, none of these categories showed gendered patterns

Thus, although we had anticipated that gender might infl uence the

stu-dents’ participation in PT-E and consequently their academic

achieve-ment, this was not actually the case The student interview data also

showed that gender was not an issue to this sample of 10th graders

The process of our data analysis was divided into three subprocesses,

as described by Huberman and Miles (1998): data reduction, data

dis-play, and drawing conclusions First, all 14 interviews were transcribed

verbatim, following the transcription conventions suggested by Richards

4 The family capital profi les (see Table 4) were based on information from student

inter-views, student responses to the survey questionnaire, and the fi rst author’s notes of the

students’ home visits See Hamid (2009) for the constituents of the family capital profi le

categories