Pfeiffer et al. Reproductive Health  (2017) 14:77 
DOI 10.1186/s12978-017-0338-x
RESEARCH Open Access
Understanding resilience of female
adolescents towards teenage pregnancy: a
cross-sectional survey in Dar es Salaam,
Tanzania
Constanze Pfeiffer1,2* , Collins K Ahorlu3, Sandra Alba1,2,5 and Brigit Obrist1,2,4
Abstract
Background: In Tanzania, teenage pregnancy rates are still high despite the efforts being made to reduce them.
Not enough is known about how adolescents experience and cope with sexuality and teenage pregnancy.
Over the past few decades, most studies have focused on vulnerability and risk among youth. The concept
of ‘reproductive resilience’ is a new way of looking at teenage pregnancy. It shifts the perspective from a
deficit-based to a strength-based approach. The study presented here aimed to identify factors that could
contribute to strengthening the reproductive resilience of girls in Dar es Salaam, Tanzania.
Methods: Using a cross-sectional cluster sampling approach, 750 female adolescents aged 15–19 years
were interviewed about how they mobilize resources to avoid or deal with teenage pregnancy. The main
focus of the study was to examine how social capital (relations with significant others), economic capital
(command over economic resources), cultural capital (personal dispositions and habits), and symbolic capital
(recognition and prestige) contribute to the development of adolescent competencies for avoiding or
dealing with teenage pregnancy and childbirth.
Results: A cumulative competence scale was developed to assess reproductive resilience. The cumulative
score was computed based on 10 competence indicators that refer to the re- and pro-active mobilization
of resources. About half of the women who had never been pregnant fell into the category, ‘high
competence’ (50.9%), meaning they could get the information and support needed to avoid pregnancies. Among
pregnant women and young mothers, most were categorized as ‘high competence’ (70.5%) and stated that they know
how to avoid or deal with health problems that might affect them or their babies, and could get the information and
support required to do so. Cultural capital, in particular, contributed to the competence of never-pregnant girls
[OR = 1.80, 95% CI = 1.06 to 3.07, p = 0.029], pregnant adolescents and young mothers [OR = 3.33, 95% CI = 1.15 to
9.60, p = 0.026].
Conclusions: The reproductive resilience framework provides new insights into the reproductive health realities of
adolescent girls from a strength-based perspective. While acknowledging that teenage pregnancy has serious negative
implications for many female adolescents, the findings presented here highlight the importance of considering girls’
capacities to prevent or deal with teenage pregnancy.
Keywords: Resilience, Adolescents, Sexual and reproductive health, Quantitative methods, Tanzania, Urban health
* Correspondence: constanze.pfeiffer@unibas.ch
1Department of Epidemiology & Public Health (EPH), Swiss Tropical and
Public Health Institute (Swiss TPH), Socinstr. 57, P.O. Box 4002, Basel,
Switzerland
2University of Basel, Petersplatz 1, 4003 Basel, Switzerland
Full list of author information is available at the end of the article
© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 2 of 12
Plain English summary known about their health and their sexual and repro-
Despite efforts to reduce teenage pregnancy rates in ductive health, in particular [5, 6].
Tanzania, they are still high. Not enough is known about In the past, adolescents have often been represented
how female adolescents experience and cope with as being very vulnerable to sexual and reproductive
sexuality and teenage pregnancy. Over the past few health risks [1–3, 7–9]. Such thinking is rooted in a
decades, most studies have focused on vulnerability and developmentalist framework that constructs adoles-
risk among youth. The concept of ‘reproductive cence as a separate stage of development during
resilience’ is a new way of looking at teenage pregnancy. which adolescents are no longer children, but not yet
It shifts the perspective from a deficit-based to a adults [10]. This Western notion of adolescence [11],
strength-based approach. The study presented here however, ignores the fact that childhood and adoles-
aimed to identify factors that could contribute to cence mean different things in different contexts [10].
strengthening the reproductive resilience of girls in Dar es In Tanzania, for instance, marital status and mother-
Salaam, Tanzania. hood shape whether adolescents are regarded as
Using a cross-sectional cluster sampling approach, grown-ups or not.
750 female adolescents aged 15–19 years were inter- Nevertheless, the developmentalist discourse is still
viewed about how they mobilize resources (such as reflected in some public health discussions and influ-
social, economic, cultural, and symbolic capital) to ences policy and practice [10]. Recently, however,
avoid or deal with teenage pregnancy. discussions around the post-2015 health agenda for
About half of the women who had never been pregnant women, children, and adolescents highlighted the
fell into the category, ‘high competence’ (50.9%), meaning importance of a paradigm shift from problem-
they could get the information and support needed to avoid oriented approaches towards those that emphasize
pregnancies. Among pregnant women and young mothers, resilience and capacity among young people [12].
most were categorized as ‘high competence’ (70.5%), and While this study does not ignore the various risks re-
stated that they know how to avoid or deal with health lated to teenage pregnancy, it argues that by focusing
problems that might affect them or their babies and could only on adolescents’ problems or weaknesses, their
get the required information and support. Cultural capital, strengths and capacities might be overlooked [13, 14]. In
in particular, contributed to the competence of never- line with Harpham, it contends that, ‘we need to know
pregnant girls, pregnant adolescents and young mothers. what to strengthen among low-income urban populations
The reproductive resilience framework provides new to protect and promote their health, and how to
insights into the reproductive health realities of ado- strengthen it. This requires information about resilience
lescent girls. The findings highlight the importance of rather than vulnerability’ ([5]: p. 115).
considering girls’ capacities to deal with teenage A strength-based approach focusing on resilience has
pregnancy. a long history in child development psychology [15–21].
Our current understanding of resilience comes from a
Background small community of Western-trained psychologists and
Young people hold the key to the future, yet they social workers in the United States and Europe [22].
face multiple and complex challenges. From a public Only in the last few years have psychologists expanded
health perspective, a key concern is the sexual and re- their research to include low- and middle-income
productive health of youth [1]. According to the countries [22–24]. Recently, a group of social scientists
United Nations (UN) the term “young people” is used from Switzerland conceptualized resilience from a social
for the age range 10 to 24. This group can be divided science perspective [25] and proposed a multi-layered
into three subgroups: younger adolescents (10 to social resilience framework. Drawing on theories of
14 years); older adolescents (15 to 19 years); and structuration and social constructivism, they acknow-
youth (15 to 24 years) [1]. Adolescents especially in ledge that resilience is a scientific construct that is influ-
low-and middle-income countries are considered to enced by the values of those who define it [26–30].
be one of the most at-risk groups as far as sexual Moreover, they put human agency in the face of threats
and reproductive health risks are concerned [1–4]. at the centre of their research and acknowledged that
Pregnancy exposes adolescent women to medical, this capacity is shaped by access to various capitals, as
social and economic threats, as they have a high risk defined by Bourdieu [29]. They, thus define social resili-
of dying in childbirth, of being socially excluded and ence as, ‘the capacity of actors to access capitals in order
of living in poverty as single mothers [2]. to - not only cope with and adjust to adverse conditions
Due to rapid urbanization in developing countries, (i.e. reactive capacity) - but search for and create options
many adolescents grow up in urban areas. Despite the (i.e. proactive capacity), and thus develop increased com-
high and growing number of urban youth, not enough is petence (i.e. positive outcomes) in dealing with a threat.
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 3 of 12
Access to economic, social and cultural capitals is to a
large extent structured by power-related symbolic capital’
([25]: p. 289–290).
While the concept of sexual and reproductive health
covers a broad range of issues around pregnancy, this study
focused on teenage pregnancy among adolescent girls aged
15 to 19 years, in two urban and two rural areas in Ghana
(Accra and Begoro) and Tanzania (Dar es Salaam and
Mtwara Town). It is widely acknowledged that reproductive
health research should target women and men [31], how-
ever, this study focused on female adolescents as the indi-
viduals most immediately and directly impacted by teenage
pregnancy. This study examined whether and how actors Fig. 1 Reproductive resilience framework (modified multi-layered social
(family, peers, teachers, etc.) as well as institutions and or- resilience framework by Obrist, Pfeiffer & Henley, 2010)
ganizations (health services, schools, youth development
projects, etc.) influence adolescent women’s competence in
avoiding or dealing with teenage pregnancy. While the short description simplifies Bourdieu’s work, it allows for
study used quantitative and qualitative methods, this article a conceptualization of social resilience [25].
presents the quantitative results from the urban site in Dar According to Bourdieu [29, 30], individual actors are
es Salaam, Tanzania. Findings from the urban setting in constrained by structures but, at the same time, can re-
Ghana have been presented elsewhere [32]. Qualitative shape existing structures. The human capacity to reflect
studies were also conducted in Tanzania and Ghana to and act in relation to a threat is both structured by and
complement the quantitative findings. These results are re- structures economic, social, and cultural capital. The
ported elsewhere [33]. definition of capacities moves beyond the common con-
cept of attributes and personal dispositions used in
Conceptual framework psycho-social resilience scales [34] and allows for the in-
This study started from the dominant public health as- clusion of life skills. Life skills refer to various abilities
sumption that unwanted and unplanned pregnancy may for adaptive behavior [35] such as the capacity to decide
threaten the health, education, and socio-economic well- freely with whom to have sex.
being of adolescent women [3, 9]. Thus, the project focused In line with the social resilience approach, the study
on unwanted and unplanned teenage pregnancy as a poten- focused on competence as an outcome of resilience. In
tial threat to female adolescents. Drawing on the social re- order to highlight the re- and pro-active component of
silience approach [25], the study examines how adolescent social resilience, the competence score was based on the
women build resilience to this threat at the household level, following outcome: being able to competently mobilize
with a particular emphasis on 1) their social, economic, cul- resources in order to avoid teenage pregnancy (for never
tural and symbolic capital; 2) their personal capacities; 3) pregnant adolescents) or to cope with teenage pregnancy
their socio-demographic context; and 4) the outcome: (for pregnant girls and young mothers).
competence (ability to competently deal with the threat of This paper focuses on the role of different types of
teenage pregnancy) (Fig. 1). capital and its impact on the competence score. It is
Building on the work of Pierre Bourdieu [29, 30], cap- hypothesized that:
ital is understood as material and non-material resources
that determine human agency. He distinguishes four (1)Adolescents can develop competencies for dealing
types of capital. Economic capital refers to the command with the threat of teenage pregnancy.
over economic resources such as cash and assets. Social (2)Mobilizing capital (economic, social, cultural, and
capital is defined as the various kinds of valued relations symbolic resources) increases adolescents’ competence
with significant others. Cultural capital is divided into in avoiding or dealing with teenage pregnancy.
three forms: embodied (personal dispositions and
habits), objectified (knowledge and tradition stored in Methods
material forms) and institutionalized (educational quali- Design
fication). Bourdieu later added symbolic capital (honor, The study focused on female adolescents aged 15–19 years,
recognition, and prestige) — a power-related resource who form part of the older adolescent age group according
that influences the ways in which actors can access other to the World Health Organization (WHO) [36]. In this art-
forms of capital. Capital is continuously attained, trans- icle, the terms “adolescents” and “girls” are used inter-
ferred, transformed, and repositioned. Although this changeably and refer to the WHO definition of older
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 4 of 12
adolescents. A cross-sectional survey was conducted with A one-stage cluster sampling approach was used to se-
female adolescents from Dar es Salaam, Tanzania. In order lect respondents in Dar es Salaam. The second smallest
to gain insights into female adolescents’ resilience, the girls administrative units in the city, the sub-wards or Mtaa/
were asked a series of questions on how they dealt with Mitaa (Kiswahili for “street/streets”), were used as
sexuality, unwanted/ unplanned teenage pregnancy, and clusters in all three municipalities (Kindondoni, Ilala and
teen motherhood (Table 1). The questionnaire used pre- Temeke). Twelve clusters, four in each of the three
coded multiple responses. municipalities in Dar es Salaam (Map 1), were randomly
selected. Within each cluster, all households with female
adolescents aged 15–19 years were visited and, based on
Study setting and sampling their willingness to participate, they were included in the
Dar es Salaam is the largest city in Tanzania. It is divided study [38, 39]. The research team stopped going to
into three municipalities: Kinondoni, Ilala, and Temeke. households only after all clusters were covered com-
According to the 2012 census, Dar es Salaam Region has pletely. Married girls could be well covered through this
a population of 4,364,541 ([37]: p. 2). Although Dar es approach as most of them either stayed with their family
Salaam is not the official capital city, it is the largest city or their partner’s family. In each cluster, an average of 63
in the country. adolescents were sampled. At the end of the fieldwork,
A total of 750 respondents were included in the study. only six girls could not be included in the study because
To attain the required sample size for the study, taking their caretakers refused to give consent.
into account the female adolescent population (160′266)
and teenage pregnancy rate (15%) in Dar es Salaam, 12
clusters — four in each of the three municipalities in Data generation
Dar es Salaam (Map 1) — were randomly selected. Prior to data collection the questionnaire was pre-tested
Sample size considerations were based on recommenda- and revised accordingly. From November 2010 to Janu-
tions that a logistic regression analysis should have at ary 2011, data collection was conducted by adolescent
least 10 cases for each explanatory variable [28]. It was women, aged 19 to 24 years. A peer-to-peer data collec-
estimated that aproximately 30% of respondents would tion approach was chosen to reduce age-related social
be considered cases (‘high competence’, see Data barriers and to increase trust between interviewees and
Analysis section for details) and that approximately 20 interviewers. The six data collectors were trained during
variables would be included in the final multivariate a three-day training workshop in Dar es Salaam and su-
regression model. In addition, a 10% non-response rate pervised during data collection. Continous monitoring
was allowed. of data collection and weekly exchanges between project
Table 1 Reproductive resilience research designa
Variables Questions (selection of few examples)
1. Socio-demographic background
Socio demographic How old are you? Are you in a relationship?
background
2. Capitals
2.1 Social capital Do you have someone you can turn to in case you have questions related to avoiding/dealing with teenage pregnancy?
Whom do you turn to?
2.2 Cultural capital Do you have access to other information sources in order to learn about how to avoid/deal with teenage pregnancy?
What kind of sources?
2.3 Economic capital Do you have someone you can turn to in case you need money to avoid/deal with teenage pregnancy? Whom do
you turn to?
2.4 Symbolic capital Do you feel accepted within your social environment? Do you strive for a good reputation?
3. Capacities
3.1 Psycho-social dispositions Do you belief that you can successfully manage to avoid/deal with teenage pregnancy? Do you have the ability to
establish and maintain relationships to people?
3.2 Life skills Do you know how to protect yourself from pregnancy? Do you decide freely if, when and with whom you want to
have sex?
4. Competence score
Competence Have you mobilized any social/economic/cultural support in order to actively avoid teenage pregnancy/deal with
teenage pregnancy?
aThis paper focuses on the impact of capitals (highlighted in italics) on competence (highlighted in italics)
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 5 of 12
Map 1 Selected administrative units (Mitaa) in the city of Dar es Salaam
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 6 of 12
leader, field supervisor, and data collectors aimed to score category. All variables with p ≤ 10% in univariate
guarantee quality. analyses were entered in the mulitviariate model.
Analyses were conducted separately by pregnancy status
Data analysis in order to gain insights into the different health realities
A cumulative competence scale was developed as out- of non-pregnant female adolescnts compared to preg-
come to assess reproductive resilience. The cumulative nant girls and young mothers. Only results from the
score was computed based on 10 competence indicators multivariate logistic regression analysis are presented in
that refer to the re- and pro-active mobilization of re- this paper.
sources. Depending on pregnancy status (never pregnant
vs. pregnant adolescents and young mothers), a set of Results
questions related to competence were asked (Table 1). Socio-demographic context of respondents
The questions were carefully discussed within the re- Of the 750 sampled teenagers, 16% (n = 112) reported
search team, comprising Tanzanian, Ghanaian, and Swiss that they were pregnant or already mothers. Table 2
social scientists, and statisticians as well as Tanzanian shows respondents’ sociodemographic characteristics by
adolescent data collectors. Particular attention was given
to the local context, including suitability of terms and Table 2 Socio-demographic characteristics of respondents, by
local concepts of teenage pregnancy. The questions were pregnancy status
pre-tested and discussed with Tanzanian adolescents and Categories Never pregnant Pregnant girls and/or
revised thorougly. girls young mothers
Each competence question answered with ‘yes’ (i.e. hav- n % n %
ing actively mobilized resources, continued education, (N = 638) (N = 112)
etc.) contributed ‘1’; questions answered with ‘no’ (i.e. not Age
having actively mobilized resources, did not continue with 15 Years 167 26.2 1 0.9
education, etc.) contributed ‘0’ to the score. Each respond-
16 years 107 16.8 11 9.8
ent among the pregnant girls and young mothers could
score a minimum of 0 (all questions answered with 17 years 114 17.9 8 7.1‘no’)
and a maximum of 10 (100%) (all questions answered with 18 years 108 16.9 20 17.9
‘yes’). For the purpose of this analysis, a score of ≤50% was 19 years 142 22.3 72 64.3
indicative of ‘low competence in mobilizing resources to Education
avoid or deal with pregnancy’, while a score of 51–100% Primary education 277 43.4 85 75.9
was considered as ‘high competence in mobilizing
Secondary education 326 51.3 21 18.8
resouces to avoid or deal with pregnancy’. A 50% cut-off
point was used in order to learn about the broad spectrum Vocational training 13 2.0 0 0
of competencies among respondents. No reliability testing No education 21 3.3 6 5.4
was performed for score development. Relationship status
Statistical analyses were conducted using IBM Single 367 57.5 11 9.8
SPSS Statistics 19 and included descriptive statistics, In a relationship/not married 258 40.4 57 50.9
chi-square tests, as well as univariate and multivari-
Married 13 2.0 38 33.9
ate logistic regressions. Bivariate relationship be-
tween competence score and capital variables, Divorced/separated 0 0 6 5.4
competence score and ability variables as well as Both parents of respondent living together with their children
competence score and demographic variables were Yes 372 58.3 53 47.3
computed. Variables for the logistic models to iden- No 169 26.5 43 38.4
tify determinants of resilience were identified by sug- Others (dead, don’t know) 87 15.2 16 14.3
gestive bivariate relationships. All logistic regression
Respondent’s father has more than one wife
models were controlled for age, since proportionally
more never pregnant girls were in the younger age Yes 213 33.4 37 33.0
than the pregnant/young mothers. The outcome No 327 51.3 58 51.8
variable of the study was the ability to avoid preg- Others (dead, don’t know) 98 15.4 17 15.2
nancy or cope well with it, which was expressed in Religion
the competence scores. Christians 207 32.4 27 24.1
The logistic regressions were fitted to assess the effect
Muslims 428 67.1 85 75.9
of social, cultural, economic, and symbolic capital vari-
ables on the odds of being in a high vs. low competence Others 3 0.5 0 0
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 7 of 12
pregnancy status. Younger girls were more likely to be are increasingly contacted [12, 40, 41]. This transition
categorized as never pregnant and not in a relationship. was confirmed by both groups in the study (Fig. 2).
Most of the married girls reported to be pregnant before However, Table 3 shows that parents/guardians in Dar es
marriage. Adolescents with less education were more Salaam did not contribute significantly to female adoles-
likely than better-educated girls to have started bearing cents’ competence level. Interviewed girls in Dar es
children. At the same time, the data indicates that early Salaam did not talk to their peers (22% never pregnant
childbearing leads to school dropouts. Contrary to girls; 18% pregnant girls and young mothers) about
Ghana, where a similar study was conducted [32], the sexual matters as often as to their parents or relatives
education policy in Tanzania until recently stated that (51% never pregnant girls; 53% pregnant girls and young
girls who became pregnant in school were to be expelled mothers). Although it can be concluded that individual
and not allowed to return following their pregnancy. In actors can contribute to the competence of girls, the
2009, the “Law of the Child” Act, 351 was passed by the quality of information in terms of accuracy and reliabil-
Tanzanian parliament, which amends this policy and al- ity was not covered in this study.
lows girls to return to school. However, it is still unclear
whether it will actually be enforced. Cultural Capital
In a context where family members, partners, and peers
Competence of female adolescents may not offer reliable information about sexual health
Both groups of adolescent women, especially those who and pregnancy, additional sources of information such
were pregnant or already mothers, had high competence as mass media become important for youth [9]. A logis-
scores. About half of the women who had never been tic regression including social (having someone to turn
pregnant fell into the category of ‘high competence’ to for social support), economic (having someone to
(50.9%), meaning they could get the information and turn to for economic support), cultural (having access to
support they needed to avoid pregnancies. Among preg- other sources of information), and symbolic capital (be-
nant women and young mothers, even more respon- ing accepted by others) showed that only cultural capital
dents had ‘high competence’ levels (70.5%) compared to contributed to the competence of never pregnant girls
the female adolescents you had never been pregnant; the [OR = 1.80, 95% CI = 1.06 to 3.07, p = 0.029] as well as
majority stated that they know how to avoid or deal with to pregnant adolescents and young mothers [OR = 3.33,
health problems that might affect them or their babies 95% CI = 1.15 to 9.60, p = 0.026] (Table 3). This finding
and could get the information and support to do so. highlights the importance of cultural capital as a deter-
minant for reproductive resilience. Table 3 provides in-
Social Capital formation about the contribution of different types of
Depending on their pregnancy status, female adolescents media to the competence of young people. Music and
in Dar es Salaam turned to different social actors, mainly the Tanzanian magazines, Fema and Si Mchezo!, pub-
their parents and peers, when in need of information on lished by Femina HIP, contributed significantly to the
how to avoid or deal with teenage pregnancy. In former competence of never pregnant girls. Femina HIP is the
times, other relatives —aunts, in particular — were biggest local multimedia platform and civil society
approached, but due to societal changes and new family organization working with youth, communities, and
structures (moving from extended to nuclear families partners in Tanzania. It publishes Fema and Si Mchezo!,
because of modernity and increasing mobility) parents the top two magazines in Tanzania for 15 to 25 year
Fig. 2 Access to social actors by pregnancy status
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 8 of 12
Table 3 Multivariate logistic regression analysis: Estimated effect of social, cultural and economic capital on the competence score,
by pregnancy status
Never pregnant girls (N = 638) Pregnant girls and/or
young mothers (N = 112)
OR 95% C.I. p-Value OR 95% C.I. p-Value
Social capital 1.40 0.875 2.25 0.159
Peersb 1.71 1.18 2.48 0.004 a
Partnerb 2.07 0.76 5.65 0.157 a
Parents/guardiansb 1.34 0.97 1.86 0.077 a
Other relativesb 0.87 0.62 1.22 0.414 a
Religious leadersb 3.36 0.69 16.36 0.133 a
Teachersb 1.68 0.69 4.08 0.258 a
Nursesb 1.43 0.51 4.04 0.495 a
Cultural capital 1.80 1.06 3.07 0.029 3.33 1.15 9.60 0.026
Booksc 1.63 1.05 2.54 0.031 2.63 0.30 23.23 0.383
Brochurec 1.41 0.75 2.63 0.286 a
Cell Phonesc 1.82 0.17 19.22 0.617 a
Magazinesc 1.96 1.38 2.77 0.001 2.75 0.88 8.60 0.083
Music songsc 3.22 1.88 5.54 0.001 1.67 0.17 16.25 0.660
Radioc 1.70 1.18 2.45 0.004 3.44 1.35 8.76 0.010
Television (TV) 0.91 0.62 1.34 0.633 1.32 0.52 3.35 0.561
Economic capital 0.85 0.39 1.88 0.692
Peers d 3.41 1.73 6.70 0.001 1.24 0.22 7.01 0.809
Partner d 2.08 1.18 3.68 0.012 3.24 3.24 8.61 0.019
Parents/guardians d 1.31 0.90 1.01 0.162 2.98 2.98 7.58 0.022
Other relatives d 1.11 0.80 1.54 0.539 3.23 3.23 8.52 0.018
Religious leaders d a a
Teachers d 10.67 1.35 84.17 0.025 a
Nurses d a a
aThe variable was not entered in the logistic regression model as it was not significant according to the variable selection strategy. All variables significant at the
10% level in univariate analyses were considered candidates for the mulitviariate model
bSpontaneous mention of social actors never pregnant girls and pregnant girls/young mothers turn to for information on sexuality and teenage pregnancy/childrearing
cSpontaneous mention of mass media never pregnant girls and pregnant girls/young mothers turn to for information on sexuality and teenage pregnancy/childrearing
dSpontaneous mention of social actors never pregnant girls and pregnant girls/young mothers turn to for financial support related to avoiding teenage pregnancy
or dealing with teenage pregnancy/childrearing
olds. Catering to secondary school students and written Femina HIP were widely read; 38% of the never pregnant
in English and Swahili, Fema is published quarterly and and 31% of the pregnant girls/young mothers stated that
has a circulation of 180,000. Directed to out-of-school they read one of these magazines (Fig. 3).
youth and written in Swahili, Si Mchezo! is published on While different types of media contributed to the com-
a bi-monthly basis and has a circulation of approxi- petence of never pregnant girls, the same could not be
mately 175,000. The two magazines also seek to reach observed for pregnant girls and young mothers, as mass
young people in the workplace and in organizational media campaigns tend to target girls who are not
settings. pregnant.
Radio contributed to the competence of both groups,
but not everyone has access to cultural capital. The find- Economic Capital
ings show that 57% of never pregnant girls and 54% of Access to economic resources in general did not signifi-
their pregnant peers and young mothers had access to cantly contribute to the competence of female adoles-
television (TV) (Fig. 3). Radio was used widely among cents [never pregnant girls: OR = 0.85, 95% CI = 0.39 to
young mothers or mothers to be, with 60% saying that 1.88, p = 0.692; for pregnant girls and young mothers,
they listen to it, while 52% of the never pregnant respon- the variable “economic capital” could not be entered in
dents mentioned it. Both youth magazines published by the regression model as it was not significant in
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 9 of 12
Fig. 3 Access to mass media by pregnancy status
univariate analysis] (Table 3). However, the analysis of actively to secure their own health and that of their child.
the role of individual actors illustrated that pregnancy The higher competence score of pregnant girls and young
status considerably changed the way social actors matter mothers compared to peers who had never been pregnant
for adolescent girls (Table 3). For never pregnant girls, might reflect a dynamic learning process. Due to preg-
economic capital mobilized through social actors outside nancy and motherhood, female adolescents must develop
the family contributed to their competence, while for support networks and knowledge of resources in response
pregnant girls and young mothers, economic support to their situation. Thus, childbirth might be looked at as a
from family members contributed to their competence. ‘turning point experience’ [17: p. 136] that offers new
opportunities to break away from the past [43]. What
Symbolic Capital might be regarded as a threat in one situation becomes a
Results show that almost all interviewed respondents protective factor in another. This depends on how threats
aimed to establish symbolic capital. Some 99% (n = 634) and protective factors are perceived by the affected actors
of the never pregnant girls and 98% (n = 110) of the [43]. The psycho-social resilience literature highlights that
pregnant girls and young mothers ‘strived for a good major life transitions provide new opportunities for resili-
reputation’. In addition, regardless of their pregnancy ence, not only for young mothers but also for young
status almost all female study participants (99%, n = 634 fathers [44–47]. Findings point to factors that can further
of the never pregnant girls; 98%, n = 110 of the pregnant support adolescents in dealing with teenage pregnancy.
girls/young mothers) reported feeling accepted in their Findings show that parents/guardians in Dar es Salaam
community. This finding highlights the importance of did not contribute significantly to female adolescents’
maintaining symbolic capital. Surprisingly, the symbolic competence level. A literature review [40] of parent-
capital of both groups of female adolescents did not sig- child communication about sexuality in sub-Saharan
nificantly contribute to the competence score. Although Africa argues that discussions on sexual matters tend to
this presentation simplifies Bourdieu’s conceptualization be authoritarian and vague as parents are often over-
of symbolic capital, it helps to provide insights into the whelmed with their new roles and do not know how to
different dimensions of social resilience. While findings provide sex education. Talking to peers significantly con-
indicate that symbolic capital does not contribute to the tributed to the competence of girls who had not been
competence of female adolescents, it is also acknowl- pregnant. Peers are often regarded as crucial in terms of
edged that it is not easy to quantitatively capture shaping adolescent norms that serve as reference points
symbolic capital in all its complexity. for decision making related to health [41]. Peer commu-
nication is characterized by an open atmosphere and less
Discussion restricted by cultural norms and taboos.
While acknowledging that teenage pregnancy often has Cultural capital, defined as the internalization of
serious negative consequences for female adolescents, this cultural values around teenage pregnancy that are repre-
study highlights that common portrayals of adolescent sented in media such as TV, radio or magazines, can
mothers as unprepared ‘children, who have children’ contribute to competence. Nichter [14] points out that
might not hold true for all girls [1–3, 42, 43]. Our study local people respond practically to new information and
shows that, in contrast to public health perceptions of resources, and they do so within the framework of their
teenage motherhood and in spite of the challenges some cultural institutions or knowledge. Using mass media is
youth face, adolescent girls are not mere victims per se one way to make sure that adolescent women get accur-
but also social actors that try to mobilize resources ate information and make informed decisions.
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 10 of 12
This study could not confirm access to economic with qualitative research into local meanings corre-
capital as the most important driving factor [48–50]. sponding with these scientific constructs.
Cultural and social factors, especially for never pregnant To cover pro-active capacities, never pregnant girls were
girls, are equally important. This finding promotes the included in the study. As in other studies, these girls were
value of the social development debate that highlights interviewed in the absence of a threat that had occurred
the role of social and cultural institutions and organisa- already [51, 52], which might lead to overlooking the
tions for development, in general, and for building sex- actual context of coping with adversity ([53]: p. 754).
ual and reproductive health knowledge, in particular. The study presented here used a cumulative competence
Youth-focused magazines or TV programs, for instance, scale to provide a snapshot of teenage pregnancy-related
can build competence. However, one needs to carefully realities of girls in the urban centre of Dar es Salaam. As re-
analyze who benefits from such interventions and who silience is not a stable and durable phenomenon, longitu-
falls through the cracks. dinal studies would be best suited to collecting reliable
information [53: p. 755]. Survey data provide verifiable and
Reflection on the reproductive resilience framework generalizable data [54] but they can only generate a fraction
The application of the reproductive resilience frame- of the information needed to understand the complexity of
work, which builds on the social resilience approach social institutions and relations in which reproductive prac-
[25], proved to be useful to this study in several respects. tices are contextualized [55]. The quality of information ad-
First, the framework provides a clear definition of and olescents received while mobilizing different capitals, in
highlights socio-economic and cultural aspects of repro- terms of accuracy and reliability, was, for instance, not cov-
ductive resilience, thereby allowing us to move beyond ered in this study and would require deeper quantitative as
existing ecological or child-development resilience con- well as qualitative investigation. In addition, factors other
cepts. Focusing on the strengths of people rather than than resilience might account for the differences in the
on their weaknesses alone opens new possibilities for “competence scores” of female adolescents presented here.
dealing with and reducing threats. The framework ac- Quantitative approaches alone are not sufficient to provide
knowledges the learning, self-organization, and creative insight into the full potential of agency and related creativ-
potential of people and institutions that might be able to ity of actors in different contexts. Adolescent competencies
actively deal with threats. in urban areas can only be adequately comprehended by
Second, the framework highlights that resilience is not understanding their everyday life, highlighting the need for
just about psychological traits that individuals are born culturally-adapted, mixed-methods studies.
with but rather that people can develop competencies
that allow them to deal with a given threat. The data Conclusions
presented here illustrate how crucial it is to understand This article examined how various material and non-
how and by whom these capacities are built in order to material resources can contribute to the competence of
identify entry points for future interventions. This study adolescents to avoid teenage pregnancy or to successfully
does not, however, provide insights into whether or how deal with it. The reproductive resilience framework
these capacities are then translated into action. provided new insights into the resources and competen-
Third, the framework provides new insights beyond cies of girls in Dar es Salaam, Tanzania. Although teenage
the social or economic environment. This study makes it pregnancy and motherhood involves various challenges
clear that cultural capital in terms of knowledge passed and risks, many interviewed female adolescents tried to
on by social and cultural institutions and organizations deal with it actively and might turn some of the challenges
can significantly contribute to resilience and therefore into opportunities [33, 56–60]. Strengthening the factors,
needs to be considered in detail in resilience research. that contribute to building competence in female adoles-
While the reproductive resilience framework presented cents, such as providing information in an appealing way
here offers new insights, it also has some shortcomings. through magazines, are important entry points for future
The framework is grounded in the applied realm and interventions.
therefore runs the risk of reducing resilience to norma-
tive and dominant concepts at the expense of the per- Abbreviations
spectives of actors [22, 25]. Although this study focused TV: Television; UN: United Nations; WHO: World Health Organization
on unwanted/unplanned pregnancy, some involved fe-
Acknowledgments
male adolescents might have looked at teenage preg- Without the great support of adolescent girls in Dar es Salaam, this project
nancy as something that could improve their symbolic would not have been possible. We are very grateful to Trudy Harpham, Fred
capital (their status/prestige in their community). There Krüger, Kate Molesworth, Alice Mbelwa, Boniface Kiteme, Stefan Dongus,
Richard Sambaiga, Marcel Tanner, Jakob Zinnstag, Joyce Nyoni, Rose
is clearly a need to complement experts’ views of the key Mwaipopo, Amena Briët and Patricia Schwärzler for comments and
categories (threat, capitals, capacities and competencies) suggestions.
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 11 of 12
Funding 7. National Bureau of Statistics (NBS) and ORC Macro. Tanzania demographic
This research was financed by the Swiss National Center of Competence in and health survey 2004–05. Dar es Salaam: NBS and ORC Macro; 2005.
Research North-South (NCCR) North-South, an international research program 8. National Bureau of Statistics (NBS) and ICF Macro. Tanzania demographic
of the Swiss National Science Foundation (SNSF) co-funded by the SNSF and and health survey 2010. Dar es Salaam: NBS and ICF Macro; 2011.
the Swiss Agency for Development and Cooperation (SDC). 9. UNICEF. Adolescence in Tanzania. Dar es Salaam: United Nations Children’s
Fund Tanzania; 2011.
Availability of data and materials 10. Macleod C. Teenage motherhood and the regulation of mothering in the
Not applicable. scientific literature: the South African example. Feminism and Psychology.
2001;11(4):493–511.
Authors’ contributions 11. Harari SE, Vinovskis MA. Adolescent sexuality, pregnancy and childbearing in
CP, CA, SA and BO conceived and designed the study. CP was responsible the past. In: Lawson A, Rhode DL, editors. The politics of pregnancy:
for project implementation, data analysis and writing the manuscript. CP, CA, adolescent sexuality and public policy. New Haven & London: Yale
SA and BO critically reviewed the manuscript and approved of it for University Press; 1993. p. 23–45.
publication. 12. Editorial. Women, children, and adolescents: the post-2015 agenda. Lancet.
2014;384(9949):1159.
Competing interests 13. Tumbo-Masabo Z. Conclusions. In: Tumbo-Masabo Z, Liljeström R, editors.
The authors declare that they have no competing interests. Chelewa, Chelewa. The dilemma of teenage girls. Östersund: Nordiska
Africainstitutet; 1994. p. 211–4.
Consent for publication 14. Nichter M. Global Health: why Cultural perceptions, social representations
The authors confirm that they have obtained consent to publish from the and Biopolitics matter. Tucson: University of Arizona Press; 2008.
participant (or legal parent or guardian for children) to report individual 15. Garmezy N. Resilience in Children’s Adaptation to negative life events and
patient data. stressed environments. Pediatrics. 1991;20:459–66.
16. Garmezy N. Development and Adaptation: the contribution of the
Ethics approval and consent to participate MacArthur Foundation and William Bevan. In: Kessel F, editor. Psychology,
All participants were asked for written informed consent after the purpose of science and human affairs: essays in honor of William Bevan. Boulder:
the study was explained to them. For adolescents aged 18 years and above, Westview Press; 1995. p. 109–24.
informed consent was requested directly. If teenagers were younger than 17. Rutter M. Psychosocial resilience and protective mechanisms. Am J
18 years, written consent was first obtained from their parents or caretakers. Orthopsychiatry. 1987;57:316–31.
In a next steps, the minors were asked whether they assented to participate. 18. Werner E, Smith R. Vulnerable but invincible: a study of resilient Children.
The study was conducted within the framework of the Swiss National Centre New York: McGraw-Hill; 1982.
of Competence in Research (NCCR): North-South funded project “Sexual and 19. Werner E, Smith R. Overcoming the odds: high risk Children from birth to
Reproductive Resilience of Adolescents in Ghana and Tanzania”. The research adulthood. Ithaca: Cornell University Press; 1992.
component in Tanzania was cleared by the National Institution for Medical 20. Schoon M. A Short Historical Overview of the Concepts of Resilience,
Research of Tanzania (NIMR/HQ/R.8a/Vol. IX/935; NIMR/HQ/R.8c/Vol. II/33) Vulnerability, and Adaptation. Bloomington: Workshop in Political Theory
and the Tanzanian Commission for Science and Technology and Policy Analysis, Indiana University, Working Paper W05–4; 2005.
(2010_311_NA_2009_86 dated 19th of October 2010). 21. Coleman J, Haggell A, Editors. Adolescence, risk and resilience. Against the
odds. Chichester: John Wiley and Sons; 2007.
Publisher’s Note 22. Ungar M. Nurturing hidden resilience in troubled youth. Toronto: University
Springer Nature remains neutral with regard to jurisdictional claims in of Toronto Press; 2004.
published maps and institutional affiliations. 23. Ungar M. Handbook for working with Children and youth - pathways to
resilience across cultures and contexts. Thousand Oaks: Sage Publications;
Author details 2005.
1Department of Epidemiology & Public Health (EPH), Swiss Tropical and 24. Ungar M, Liebenberg L. Resilience in action. Toronto: University of Toronto
Public Health Institute (Swiss TPH), Socinstr. 57, P.O. Box 4002, Basel, Press; 2008.
Switzerland. 2University of Basel, Petersplatz 1, 4003 Basel, Switzerland. 25. Obrist B, Pfeiffer C, Henley R. Multi-layered social resilience: a new approach
3Noguchi Memorial Institute for Medical Research, College of Health in mitigation research. Prog Dev Stud. 2010;10(4):283–93.
Sciences, University of Ghana, P.O Box LG581, Legon, Ghana. 4Institute of 26. Berger PL, Luckmann C. The social construction of reality: a treatise in the
Social Anthropology, University of Basel, Münsterplatz 19, 4051 Basel, sociology of knowledge. New York: Anchor Books; 1966.
Switzerland. 5KIT Biomedical Research, Royal Tropical Institute (KIT), 27. Giddens A. Central problems in social Theory. Action, structure and
Meibergdreef 39, Amsterdam 1105, AZ, The Netherlands. contradiction in social analysis. Los Angeles: University of California Press;
1979.
Received: 30 March 2016 Accepted: 9 June 2017 28. Giddens A. The constitution of society: outline of the Theory of
Structuration. Cambridge: Polity Press; 1984.
29. Bourdieu P. Distinction: a social critique of the judgment of taste.
References Cambridge: Massachussetts, Harvard University Press; 1984.
1. World Health Organization. Building a better future for youth: learning from 30. Bourdieu P. The forms of Capital. In: Richardson JE, editor. Handbook of
experiences and evidence. Geneva: World Health Organization; 2006. Theory and research for the sociology of education. New York: Greenwood
2. Chandra-Mouli V, Svanemyr J, Amin A, Fogstad H, Say L, Girard F, et al. Press; 1986. p. 241–58.
Twenty years after international conference on population and 31. Dudgeon MR, Inhorn M. Men’s influences on Women’s reproductive health:
development: where are we with adolescent sexual and reproductive medical anthropological perspectives. Soc Sci Med. 2004;59(7):1379–95.
health and rights? J Adolesc Health. 2015;56:S1eS6. 32. Ahorlu CK, Pfeiffer C, Obrist B. Socio-cultural and economic factors
3. Bearinger L, Sieving RE, Ferguson J, Sharma V. Global perspective on the influencing adolescents’ resilience against the threat of teenage pregnancy:
sexual and reproductive health of adolescents: patterns, prevention, and a cross-sectional survey in Accra, Ghana. Reprod Health. 2015;12:117.
potential. Lancet. 2007;369:1220–321. 33. Sambaiga R. Sexual inter-subjectivity and the quest for social well-being: an
4. Patton GC, Sawyer SM, Santelli JS, et al. Our future: a lancet commission on ethnographic inquiry of adolescent sexuality and reproduction in urban
adolescent health and wellbeing. Lancet. 2016;387:2423–78. southern Tanzania. PhD diss. Basel: University of Basel; 2013.
5. Harpham T. Urban health in developing countries: what do we know and 34. Ahern N, Kiehl E, Sole M, Byers J. A review of instruments measuring
where do we go? Health & Place. 2008;15:107–16. resilience. Issues Compr Pediatr Nurs. 2006;29:103–25.
6. Montgomery M, Stren R, Cohen B, Reed H. Cities transformed: demographic 35. World Health Organization. The sexual and reproductive health of younger
change and its implications in the developing World. National Academies adolescents: research issues in developing countries. Geneva: World Health
Press, Washington, DC; 2004. Organization; 2011.
Pfeiffer et al. Reproductive Health  (2017) 14:77 Page 12 of 12
36. World Health Organization. The second decade: improving adolescent
health and development. Geneva: World Health Organization; 2001.
37. United Republic of Tanzania. Population and housing census. Dar es Salaam:
United Republic of Tanzania; 2012. p. 2012.
38. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study
of the number of events per variable in logistic regression analysis. J Clin
Epidemiol. 1996;49:1373–9.
39. Kish L. Survey sampling. New York: John Wiley & Sons; 1965.
40. Bastien S, Kajula LJ, Muhwezi WW. A review of studies of parent-Child
communication about sexuality and HIV/AIDS in sub-Saharan Africa. Reprod
Health. 2011;8:25.
41. Coleman J. Free riders and zealots: the role of social networks. Sociol
Theory. 1988;6:52–7.
42. Breheny M, Stephens C. Irreconcilable differences: health professionals’
constructions of adolescence and motherhood. Soc Sci Med. 2007;64:112–24.
43. Collins B. Resilience in teenage mothers: a follow-up study. Wellington:
Ministry of Social Development; 2010.
44. Werner EE. Protective factors and individual resilience. In: Shonkoff JP,
Meisels SJ, editors. Handbook of early childhood intervention. New York:
Russell Sage; 2000. p. 297–314.
45. Easterbrooks MA, Chaudhuri J, Bartlett JD, Copeman A. Resilience in
parenting among young mothers: family and ecological risks and
opportunities. Child Youth Serv Rev. 2010;33:42–50.
46. Marsiglio W. Studying fathering trajectories: in-depth interviewing and
sensitizing concepts. In: Day RD, Lamb ME, editors. Conceptualizing and
measuring father involvement. Mahwah: Erlbaum; 2004. p. 61-82.
47. Arnett JJ. Emerging adulthood: a Theory of development from the late
teens through the twenties. Am Psychol. 2000;55:469–80.
48. Cleveland H, Jacobs G. Human choice: the genetic code for social
development. Minneapolis: World Academy of Art & Science, University of
Minnesota; 1999.
49. Ranis G, Stewart F, Ramirez A. Economic growth and human development.
World Dev. 2000;28(2):197–219.
50. Ranis G. Human development and economic growth. Center discussion
paper no. 887. New Haven: Yale University; 2004.
51. Friborg O, Hjemdal O, Rosenvinge JH, Martinussen M. A new rating scale for
adult resilience: what are the central protective resources behind healthy
adjustment? Int J Methods Psychiatr Res. 2003;12(2):65–76.
52. Montross LP, Depp C, Daly J, Reichstadt J, Golshan S, Moore D, et al.
Correlates of self-rated successful aging among community-dwelling older
adults. Am J Geriatr Psychiatr. 2006;14(1):43–51.
53. Bonnano GA. Uses and abuses of the resilience construct: loss, trauma, and
health-related adversities. Soc Sci Med. 2012;74:753–6.
54. Price N, Hawkins K. Researching sexual and reproductive behavior: a peer
ethnographic approach. Soc Sci Med. 2002;55:1325–36.
55. Lockwood M. Structure and behavior in the social demography of Africa.
Popul Dev Rev. 1995;21(1):1–32.
56. Vance CS. Anthropology rediscovers sexuality: a theoretical comment. Soc
Sci Med. 1991;33(8):875–84.
57. Johnson-Hanks J. Uncertain honour: modern motherhood in an African
crisis. Chicago: University of Chicago Press; 2006.
58. Silberschmidt M, Rasch V. Adolescent girls, illegal abortions and ‘sugar-
daddies’ in Dar es Salaam: vulnerable victims and active social agents.
Soc Sci Med. 2001;52:1815–26.
59. Van Reeuwijk M. Because of temptations: Children, sex, and HIV/AIDS in
Tanzania. Diemen: AMB Publishers; 2010.
60. Halley MC. Negotiating sexuality: adolescent initiation rituals and Cultural
change in rural southern Tanzania: Case Western Reserve University; 2012. Submit your next manuscript to BioMed Central 
and we will help you at every step:
•  We accept pre-submission inquiries 
•  Our selector tool helps you to find the most relevant journal
•  We provide round the clock customer support 
•  Convenient online submission
•  Thorough peer review
•  Inclusion in PubMed and all major indexing services 
•  Maximum visibility for your research
Submit your manuscript at
www.biomedcentral.com/submit