P. Oosterhoff
Pauline Oosterhoff is a Research Fellow at IDS. She has over
20 years’ experience in research, advisory services and media
production on sexuality, and sexual and reproductive health rights,
working with international non-governmental organisations (INGOs),
the UN, bilateral donors, private foundations, universities and the
private sector. She is interested in how research and public engagement
can support progressive, inclusive and effective policy and practice.
Pauline is a keen advocate of using mixed and participatory methods
and working in multidisciplinary teams. In addition to her research
and advisory work, she produces documentary films, installations and
immersive interactive events.
Address: Pauline Oosterhoff PhD MPH MA
Research Fellow
Gender and Sexuality Cluster
Institute of Development Studies
Brighton BN1 9RE UK
20 years’ experience in research, advisory services and media
production on sexuality, and sexual and reproductive health rights,
working with international non-governmental organisations (INGOs),
the UN, bilateral donors, private foundations, universities and the
private sector. She is interested in how research and public engagement
can support progressive, inclusive and effective policy and practice.
Pauline is a keen advocate of using mixed and participatory methods
and working in multidisciplinary teams. In addition to her research
and advisory work, she produces documentary films, installations and
immersive interactive events.
Address: Pauline Oosterhoff PhD MPH MA
Research Fellow
Gender and Sexuality Cluster
Institute of Development Studies
Brighton BN1 9RE UK
less
InterestsView All (52)
Uploads
Papers by P. Oosterhoff
action research into alcohol abuse in communities living with
bonded labour in Tamil Nadu and in Bihar. The action research on
alcoholism in these communities is part of a large-scale mixedmethods qualitative and quantitative participatory research programme which was designed to encourage community-based solutions to bonded labour in India. The research was conducted in two
‘slavery hotspots’, both with a high prevalence (over 50%) of households with people in bonded labour, as part of a multi-million dollar
programme of interventions funded by a foundation specialising in
modern-slavery eradication. The programme focuses on workers in
brick kilns and stone quarries and, to some extent, on sex workers in
Bihar and Uttar Pradesh, and on cotton-mill workers in Tamil Nadu.
The participatory research runs in parallel to programme interventions by local NGO’s. This paper examines whether participatory
action research methods can transform violent, stigmatising
approaches to alcohol consumption in marginalised communities
into community-owned, evidence-informed, non-violent harm
reduction approaches. In both locations alcohol consumption is
a highly stigmatised public secret. Bihar has banned alcohol production and consumption.
We used participatory statistics to collect and analyse the data of 3,466 at base-line and 3,177 households at end-line.
We drew a cross-section of households for each NGO and hamlet and randomly sampled households from a list of households based on a social map that complemented the census data. Residents collected the data of the households with technical support of trained researchers. To gain additional insight into the causal relations and local perceptions of what positive change could look like, the research team facilitated focus group discussions on the results at the end of the data collection process in each hamlet or ward at both the base- and endline surveys. These discussions focused on the reasons for the differences in prevalence results and used the tallied-up data to explore how gender, age and caste dynamics shape bonded labour.
We found that the interventions sites have been chosen well as they are in areas where there has been a high prevalence of bonded labour. Bonded labour reduced dramatically in the intervention communities in the period between the base- and endline surveys from 56.2% to 11.6%.
Caste, gender, age, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. Most people in the intervention areas at both baseline and endline belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC). While the prevalence of bondage inside and outside the community has reduced, men and boys are still working more often in conditions of bondage. This reflects the persistence of gendered norms about work that facilitates the acceptance of child labour among boys.
The data do not show a significant linkage between land ownership status and bonded labour. However, the data does show that as the size of the land holding increases, the prevalence of bonded labour decreases.
Caste, gender, age, access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) benefits, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. As reported payment received for the number of days worked through MGNREGA increases, the incidence of bonded labour decreases. This suggests that improved access to economic alternatives, such as MNREGA, may be an important part of an intervention package aiming to reduce bondage.9 While these outcomes are promising, the current reported delay in payment of wages from other sources10 could reduce the benefits of such alternatives.
Health expenses are still the main reason for taking out a loan among all households in the intervention areas. No progress was made in terms of access to health services, and in fact the number of health facilities appears to have decreased.
We found substantial geographic variations within our sample. There are also huge differences in reductions between NGO areas. The causes of these differences are not clear and may be related to the socio-economic characteristics of these areas, effects of natural disasters, the scope and intensity of government, or differences in NGO and community-led interventions.
We used participatory statistics to collect and analyse the data of 2,970 randomly sampled households at base-line and 2700 households at end-line. We drew a cross-section of households for each NGO and hamlet and randomly sampled households from a list of households based on a social map that complemented the census data. Residents collected the data of the households with technical support of trained researchers. To gain additional insight into the causal relations and local perceptions of what positive change could look like, the research team facilitated focus group discussions on the results at the end of the data collection process in each hamlet or ward at both the base- and endline surveys. These discussions focused on the reasons for the differences in prevalence results and used the tallied-up data to explore how gender, age and caste dynamics shape bonded labour.
We found that the interventions sites have been chosen well as they are in areas where there has been a high prevalence of bonded labour. Bonded labour reduced dramatically in the intervention communities in the period between the base- and endline surveys from 56.1% to 11.1%.
Most people in the intervention areas at both baseline and endline belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC). Most people in the intervention area have a stable lease for the house they live in but are landless. The data do not show a significant linkage between land ownership status and bonded labour.
Caste, gender, age, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. Access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) and a bank account did not reduce bonded labour. Access to safe loans has increased slightly since the baseline, but among households with at least one or all members in bondage, the money lender remains the main source for credit. Health expenses, especially emergency health expenses, which were the main reason why people took out high-risk loans that could trap them in bonded labour are still the main reason for taking out a loan among all households in the intervention areas.
We found substantial geographic variations within our sample. There is variety between the communities, the intervention characteristics and components. These components include community-based groups, access to entitlements, improving performance of schools, activation of government livelihoods programmes, rescue and rehabilitation, awareness raising and legal support, community support groups (CSG), adolescent girls and boys groups, NGO’s interventions at the factories strengthening the Internal Complaint Committees, formation of action research groups (ARGs). Furthermore, collaborations between local groups, local government bodies and the private sector took different forms based on local needs and capacities.
This rich diversity means that we cannot say which intervention is the “magic bullet” that can be scaled up to ensure universal success. What we can say is that a bottom-up approach that stimulates and facilitates collective organisation and local action is effective in reducing the prevalence of bonded labour in the intervention communities living with bonded labour.
environment for children where sexual exploitation is known to occur. The AES is made up of a diverse range of venues that includes restaurants, folk dance bars, erotic dance bars, massage parlours, guest houses, and hotels.
This Working Paper seeks to understand the reasons why the commercial sexual exploitation of children (CSEC) exists in the AES, focusing on how social and cultural norms reinforce and reproduce the material and structural realities of the sector. It also explores how these norms affect the employment trajectories of minors and looks at the role of labour intermediaries – both formal and informal – in intentionally or unintentionally supporting the employment of minors in the AES.
and obstacles that Vietnamese transgender men and women face in a
fast-growing economy mainly composed of small and medium-sized
enterprises. The findings demonstrate that gender norms lead to
economic inequality between trans men and women. They also show
that small businesses and entrepreneurship play an important role in
creating legitimate and satisfying employment for transgender people.
This handbook provides guidelines and training exercises for building communication skills, building capacity to share personal experience within a support group, build social networks, provide emotional support to others inside and outside the group, and solve conflicts within families or with employers or health providers who discriminate against infected persons.
The Freedom Fund, along with 11 partner organisations in Tamil Nadu, is focusing geographically on hotspots/locations where forms of bonded labour are more prevalent. It is working towards reducing trafficking, bonded labour and harmful child labour. As part of this process - an independent evaluation of the hotspot - being carried out by the Institute of Development Studies UK and Praxis India – there is a strong focus on the relevance and effectiveness of partners’ work. The study aims to seek answers to an estimation of prevalence of bonded labour across Freedom Fund areas: on how prevalence differs among populations with different socio-economic characteristics; on the indicators that unpack whether a family is more resilient or prone to bonded labour (such as loans or alcohol consumption); and on whether the method being used for measuring bonded labour is likely to detect the expected change between baseline and endline. While individual NGO partners have monitoring and evaluation systems to measure the progress of their intervention, the aim of our work is to assess the impact of interventions across the Freedom Fund hotspot as a whole.
The Institute of Development Studies (IDS) has been carrying out a programme of research, learning and evaluation in relation to the Freedom Fund ‘hotspot’ in northern India, a project that seeks to reduce bonded labour in the states of Uttar Pradesh and Bihar.
The work for this baseline study builds upon scoping visits comprising interviews with non-governmental organisations (NGOs), focus groups with community members, field observations, the participatory collection and analysis of 353 life stories to identify the most significant indicators of change, and the generation of a baseline of participatory statistics of 3,466 households across 82 hamlets in locations covered by 14 NGOs. This is being followed by the roll-out of a systemic action research programme which combines stakeholders analysing and developing solutions to their problems with follow-up participatory statistical analysis. We will conduct an end-line survey approximately two years after the data collection for this study has been completed.
A central aim of the study was to estimate the prevalence of bonded labour in the selected intervention communities of the Freedom Fund hotspot in northern India. The study does not extrapolate from this estimate to make an estimate of the prevalence beyond this intervention area. Prevalence data help mainly to understand the profile of families in bonded labour and any correlations with different variables. The analysis of life stories provided a better insight into the life situations of families in bonded labour and explored questions of why and how. A range of other indicators could therefore be generated from the causal factors emerging from the life story analysis. The team facilitated a discussion on the results at the end of the data collection process in each site. These discussions focused on the reasons for the differences in prevalence results using the tallied-up data to explore how gender, age and caste dynamics shape bonded labour, with most adults in bonded labour working inside the village; most boys in bondage working outside the village; and a group of families with all members in bonded labour working outside the village.
The estimates from this participatory statistical analysis show the correlations of bonded labour with various factors. Where possible, conclusions have been drawn about whether this quantitative analysis corroborates certain widely held assumptions with regard to forms of bonded labour in India.
Key findings are as follows:
• Within the sample of 3,466 households, most households have a member in bonded labour. Among the 51% of the families that had people in bonded labour, 29% had all of the working family members in bonded labour and 22% had at least one enslaved family member. The interventions are clearly in the right spot.
• There are huge geographic variations within our sample: in some intervention areas, the vast majority of households had some form of bonded labour (>95%), while in others the rate was less than 10%.
• Within the intervention communities, the prevalence rate of households with at least one member in bonded labour was 53.0%. With a standard deviation of 0.4991032, and a desired confidence level of 90%, the corresponding confidence interval is ± 0.014, meaning that we can be 90% confident that the true population mean falls within the range of 51.72 to 54.51%.
• Among the total number of 3,366 bonded labourers in 3,466 households, 568 were bonded labourers aged below 18 years and 467 of these were boys. Most of the boys who were involved in bonded labour worked outside the village. Adults in bonded labour worked more often inside the village.
• Caste, gender, age, access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) benefits, and loan-taking are the key factors at the individual and household level related to bonded labour in this hotspot. Within the hotspot as a whole there are different economic activities but there are currently few economic opportunities available that do not involve some form of bondage.
• With regard to social status, most people in the intervention areas belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC).
• The data show a link between land ownership status and bonded labour. While 61.9% of landless households have at least one person in bonded labour, as many as 75.9% of landless households have every working member of the household in bonded labour. As the size of the land holding increases, the prevalence of bonded labour in those households decreases. Within the intervention communities in both states, most people have a stable lease for the house they live in.
• Health expenses are the main reason for taking out a loan among all households in the intervention areas.
• With regard to MGNREGA, as payment received for the number of days worked increases, the incidence of bonded labour decreases slightly.
• Access to a bank account does not have any significant impact on the status of bonded labour.
action research into alcohol abuse in communities living with
bonded labour in Tamil Nadu and in Bihar. The action research on
alcoholism in these communities is part of a large-scale mixedmethods qualitative and quantitative participatory research programme which was designed to encourage community-based solutions to bonded labour in India. The research was conducted in two
‘slavery hotspots’, both with a high prevalence (over 50%) of households with people in bonded labour, as part of a multi-million dollar
programme of interventions funded by a foundation specialising in
modern-slavery eradication. The programme focuses on workers in
brick kilns and stone quarries and, to some extent, on sex workers in
Bihar and Uttar Pradesh, and on cotton-mill workers in Tamil Nadu.
The participatory research runs in parallel to programme interventions by local NGO’s. This paper examines whether participatory
action research methods can transform violent, stigmatising
approaches to alcohol consumption in marginalised communities
into community-owned, evidence-informed, non-violent harm
reduction approaches. In both locations alcohol consumption is
a highly stigmatised public secret. Bihar has banned alcohol production and consumption.
We used participatory statistics to collect and analyse the data of 3,466 at base-line and 3,177 households at end-line.
We drew a cross-section of households for each NGO and hamlet and randomly sampled households from a list of households based on a social map that complemented the census data. Residents collected the data of the households with technical support of trained researchers. To gain additional insight into the causal relations and local perceptions of what positive change could look like, the research team facilitated focus group discussions on the results at the end of the data collection process in each hamlet or ward at both the base- and endline surveys. These discussions focused on the reasons for the differences in prevalence results and used the tallied-up data to explore how gender, age and caste dynamics shape bonded labour.
We found that the interventions sites have been chosen well as they are in areas where there has been a high prevalence of bonded labour. Bonded labour reduced dramatically in the intervention communities in the period between the base- and endline surveys from 56.2% to 11.6%.
Caste, gender, age, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. Most people in the intervention areas at both baseline and endline belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC). While the prevalence of bondage inside and outside the community has reduced, men and boys are still working more often in conditions of bondage. This reflects the persistence of gendered norms about work that facilitates the acceptance of child labour among boys.
The data do not show a significant linkage between land ownership status and bonded labour. However, the data does show that as the size of the land holding increases, the prevalence of bonded labour decreases.
Caste, gender, age, access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) benefits, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. As reported payment received for the number of days worked through MGNREGA increases, the incidence of bonded labour decreases. This suggests that improved access to economic alternatives, such as MNREGA, may be an important part of an intervention package aiming to reduce bondage.9 While these outcomes are promising, the current reported delay in payment of wages from other sources10 could reduce the benefits of such alternatives.
Health expenses are still the main reason for taking out a loan among all households in the intervention areas. No progress was made in terms of access to health services, and in fact the number of health facilities appears to have decreased.
We found substantial geographic variations within our sample. There are also huge differences in reductions between NGO areas. The causes of these differences are not clear and may be related to the socio-economic characteristics of these areas, effects of natural disasters, the scope and intensity of government, or differences in NGO and community-led interventions.
We used participatory statistics to collect and analyse the data of 2,970 randomly sampled households at base-line and 2700 households at end-line. We drew a cross-section of households for each NGO and hamlet and randomly sampled households from a list of households based on a social map that complemented the census data. Residents collected the data of the households with technical support of trained researchers. To gain additional insight into the causal relations and local perceptions of what positive change could look like, the research team facilitated focus group discussions on the results at the end of the data collection process in each hamlet or ward at both the base- and endline surveys. These discussions focused on the reasons for the differences in prevalence results and used the tallied-up data to explore how gender, age and caste dynamics shape bonded labour.
We found that the interventions sites have been chosen well as they are in areas where there has been a high prevalence of bonded labour. Bonded labour reduced dramatically in the intervention communities in the period between the base- and endline surveys from 56.1% to 11.1%.
Most people in the intervention areas at both baseline and endline belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC). Most people in the intervention area have a stable lease for the house they live in but are landless. The data do not show a significant linkage between land ownership status and bonded labour.
Caste, gender, age, and loan-taking are key factors at the individual and household level related to bonded labour in this hotspot. Access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) and a bank account did not reduce bonded labour. Access to safe loans has increased slightly since the baseline, but among households with at least one or all members in bondage, the money lender remains the main source for credit. Health expenses, especially emergency health expenses, which were the main reason why people took out high-risk loans that could trap them in bonded labour are still the main reason for taking out a loan among all households in the intervention areas.
We found substantial geographic variations within our sample. There is variety between the communities, the intervention characteristics and components. These components include community-based groups, access to entitlements, improving performance of schools, activation of government livelihoods programmes, rescue and rehabilitation, awareness raising and legal support, community support groups (CSG), adolescent girls and boys groups, NGO’s interventions at the factories strengthening the Internal Complaint Committees, formation of action research groups (ARGs). Furthermore, collaborations between local groups, local government bodies and the private sector took different forms based on local needs and capacities.
This rich diversity means that we cannot say which intervention is the “magic bullet” that can be scaled up to ensure universal success. What we can say is that a bottom-up approach that stimulates and facilitates collective organisation and local action is effective in reducing the prevalence of bonded labour in the intervention communities living with bonded labour.
environment for children where sexual exploitation is known to occur. The AES is made up of a diverse range of venues that includes restaurants, folk dance bars, erotic dance bars, massage parlours, guest houses, and hotels.
This Working Paper seeks to understand the reasons why the commercial sexual exploitation of children (CSEC) exists in the AES, focusing on how social and cultural norms reinforce and reproduce the material and structural realities of the sector. It also explores how these norms affect the employment trajectories of minors and looks at the role of labour intermediaries – both formal and informal – in intentionally or unintentionally supporting the employment of minors in the AES.
and obstacles that Vietnamese transgender men and women face in a
fast-growing economy mainly composed of small and medium-sized
enterprises. The findings demonstrate that gender norms lead to
economic inequality between trans men and women. They also show
that small businesses and entrepreneurship play an important role in
creating legitimate and satisfying employment for transgender people.
This handbook provides guidelines and training exercises for building communication skills, building capacity to share personal experience within a support group, build social networks, provide emotional support to others inside and outside the group, and solve conflicts within families or with employers or health providers who discriminate against infected persons.
The Freedom Fund, along with 11 partner organisations in Tamil Nadu, is focusing geographically on hotspots/locations where forms of bonded labour are more prevalent. It is working towards reducing trafficking, bonded labour and harmful child labour. As part of this process - an independent evaluation of the hotspot - being carried out by the Institute of Development Studies UK and Praxis India – there is a strong focus on the relevance and effectiveness of partners’ work. The study aims to seek answers to an estimation of prevalence of bonded labour across Freedom Fund areas: on how prevalence differs among populations with different socio-economic characteristics; on the indicators that unpack whether a family is more resilient or prone to bonded labour (such as loans or alcohol consumption); and on whether the method being used for measuring bonded labour is likely to detect the expected change between baseline and endline. While individual NGO partners have monitoring and evaluation systems to measure the progress of their intervention, the aim of our work is to assess the impact of interventions across the Freedom Fund hotspot as a whole.
The Institute of Development Studies (IDS) has been carrying out a programme of research, learning and evaluation in relation to the Freedom Fund ‘hotspot’ in northern India, a project that seeks to reduce bonded labour in the states of Uttar Pradesh and Bihar.
The work for this baseline study builds upon scoping visits comprising interviews with non-governmental organisations (NGOs), focus groups with community members, field observations, the participatory collection and analysis of 353 life stories to identify the most significant indicators of change, and the generation of a baseline of participatory statistics of 3,466 households across 82 hamlets in locations covered by 14 NGOs. This is being followed by the roll-out of a systemic action research programme which combines stakeholders analysing and developing solutions to their problems with follow-up participatory statistical analysis. We will conduct an end-line survey approximately two years after the data collection for this study has been completed.
A central aim of the study was to estimate the prevalence of bonded labour in the selected intervention communities of the Freedom Fund hotspot in northern India. The study does not extrapolate from this estimate to make an estimate of the prevalence beyond this intervention area. Prevalence data help mainly to understand the profile of families in bonded labour and any correlations with different variables. The analysis of life stories provided a better insight into the life situations of families in bonded labour and explored questions of why and how. A range of other indicators could therefore be generated from the causal factors emerging from the life story analysis. The team facilitated a discussion on the results at the end of the data collection process in each site. These discussions focused on the reasons for the differences in prevalence results using the tallied-up data to explore how gender, age and caste dynamics shape bonded labour, with most adults in bonded labour working inside the village; most boys in bondage working outside the village; and a group of families with all members in bonded labour working outside the village.
The estimates from this participatory statistical analysis show the correlations of bonded labour with various factors. Where possible, conclusions have been drawn about whether this quantitative analysis corroborates certain widely held assumptions with regard to forms of bonded labour in India.
Key findings are as follows:
• Within the sample of 3,466 households, most households have a member in bonded labour. Among the 51% of the families that had people in bonded labour, 29% had all of the working family members in bonded labour and 22% had at least one enslaved family member. The interventions are clearly in the right spot.
• There are huge geographic variations within our sample: in some intervention areas, the vast majority of households had some form of bonded labour (>95%), while in others the rate was less than 10%.
• Within the intervention communities, the prevalence rate of households with at least one member in bonded labour was 53.0%. With a standard deviation of 0.4991032, and a desired confidence level of 90%, the corresponding confidence interval is ± 0.014, meaning that we can be 90% confident that the true population mean falls within the range of 51.72 to 54.51%.
• Among the total number of 3,366 bonded labourers in 3,466 households, 568 were bonded labourers aged below 18 years and 467 of these were boys. Most of the boys who were involved in bonded labour worked outside the village. Adults in bonded labour worked more often inside the village.
• Caste, gender, age, access to Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) benefits, and loan-taking are the key factors at the individual and household level related to bonded labour in this hotspot. Within the hotspot as a whole there are different economic activities but there are currently few economic opportunities available that do not involve some form of bondage.
• With regard to social status, most people in the intervention areas belonged to the Dalit (or Scheduled Caste) social category, followed closely by Other Backward Classes (OBC).
• The data show a link between land ownership status and bonded labour. While 61.9% of landless households have at least one person in bonded labour, as many as 75.9% of landless households have every working member of the household in bonded labour. As the size of the land holding increases, the prevalence of bonded labour in those households decreases. Within the intervention communities in both states, most people have a stable lease for the house they live in.
• Health expenses are the main reason for taking out a loan among all households in the intervention areas.
• With regard to MGNREGA, as payment received for the number of days worked increases, the incidence of bonded labour decreases slightly.
• Access to a bank account does not have any significant impact on the status of bonded labour.