We’re building a new UNICEF.org.
As we swap out old for new, pages will be in transition. Thanks for your patience – please keep coming back to see the improvements.

Evaluation database

Evaluation report

2016 India: Summative Evaluation of UNICEF India’s Cotton Corridors Project: ‘Preventing Exploitation and Protecting Children’s rights in Andhra Pradesh and Karnataka.’

Author: Oxford Policy Management

Executive summary

With the aim to continuously improve transparency and use of evaluation, UNICEF Evaluation Office manages the "Global Evaluation Reports Oversight System (GEROS)". Within this system, an external independent company reviews and rates all evaluation reports. The quality rating scale for evaluation reports is as follows: “Highly Satisfactory”, “Satisfactory”, “Fair” or “Unsatisfactory”. You will find the link to the quality rating below, labelled as ‘Part 2’ of the report, and the executive feedback summary labelled as 'Part 3'.


This report presents findings from an independent summative evaluation conducted in 2015–2016 by OPM and Glocal of the "Cotton Corridors Project: Preventing Exploitation and Protecting Child Rights in AP and Karnataka" (henceforth Cotton Corridors project). The Cotton Corridors project ran between June 2008 and December 2014, with a budget allocated by IKEA Foundation of EUR 4,794,135, and was implemented by UNICEF India and the Governments of AP and Karnataka.


The objective of the Cotton Corridors project was to create improved systems and structures for preventing exploitation and protecting children in Kurnool and Raichur districts, building on the work of an earlier phase of project activity that started in 2006. These districts were selected because they had a particularly large estimated number of children working in cottonseed and cotton production and were out of school. The project aimed to improve child protection outcomes in these districts, specifically to:

  • return 18,750 long-term and short-term drop outs aged 6-14 to school; 
  • improve the quality of child friendly infrastructure in 350 schools; 
  • train 1,500 adolescent girls in relevant skills; 
  • form 591 collectives for young people to discuss and have a dialogue with policy-makers and 
  • equip 4,500 women as leaders on child protection issues.

The Cotton Corridors project took an innovative approach to achieving these objectives. The key characteristics of this approach were:

  • working closely with state, district and block-level government officials and community structures to create an enabling environment for improved child protection through engagement, technical assistance, and information, education and communication (IEC); 
  • forming, training and supporting three types of community groups in each village (a Child Protection Committee [CPC], a Child Club [CC], and a Girls’ Collective) to work with community volunteers (CVs) to identify and address child protection issues in their villages; 
  • collecting, monitoring and agreeing data on children who were not in school; and 
  • conducting with government and communities a set of activities designed to improve child protection outcomes, including:
    • conducting enforcement drives on farms or businesses employing children, rescuing children and returning them to formal or bridge education, and fining employers; 
    • conducting enforcement drives on under-age marriages and fining parents and priests; 
    • funding children at risk of dropping out to attend open school from 10th class and sit for exams; 
    • organising and funding Non-Residential Special Training Centres (NRSTCs) for children who have recently dropped out of school and need to catch up before re-entering; 
    • organising and funding Residential STCs (RSTCs) for long-term dropouts; 
    • organising and funding Child Migration Prevention Centres (CMPC) to allow children to remain in school while their parents migrate for work; and 
    • organising and funding skills training for adolescents.

The purpose of the evaluation is to provide an independent assessment of whether the project’s strategies and approaches contributed to the achievement of objectives, and to identify lessons that can be learned for future attempts to improve child protection. The evaluation answers five key questions organised around five evaluation criteria:

  • relevance: To what extent were the outputs of Cotton Corridors project consistent with recipients’ requirements, country needs, global priorities and UNICEF’s policies? 
  • effectiveness: To what extent were outcomes of the various strands of the Cotton Corridors project successfully achieved? 
  • efficiency: How economically were resources/inputs converted into results? 
  • sustainability and replicability: To what extent have benefits from Cotton Corridors project continued after UNICEF support has been completed and why, and to what extent are the successes of the project replicable elsewhere? 
  • equity and gender: To what extent have activities affected marginalised communities, addressed differences in gender, and empowered both girls and boys?

The key intended audiences are: i) departments in the Governments of AP, Karnataka and India relevant to child labour (Labour, Education, Women and Child Development [WCD], and Planning), and ii) UNICEF India staff, specifically, the Hyderabad Chief of Office, India Chief of Child Protection, and the Policy, Planning and Evaluation (PPE) and Child Protection networks. In order to ensure that the evaluation delivers on its objectives to its key audiences, UNICEF constituted an Evaluation Steering Committee of the Cotton Corridors project and evaluation staff, and an Evaluation Reference Group also including government officials and independent experts.


The methodology used for the evaluation was limited by its timing and scope: the evaluation was commissioned nine months after the project ended and there was no possibility of a counterfactual. This meant that the evaluation relied on (non-representative) primary data collected from 64 focus group discussions (FGDs) with group members and parents, 144 scale questionnaires with group chairs, 192 in-depth interviews (IDIs) with key stakeholders in 16 villages, and 19 IDIs at district and state level, all conducted a year after project completion, and on project and secondary documents. This introduces limitations to the rigour of the assessment, in particular further down the ‘results chain’ to child protection outcomes, compelling the evaluation team to rely more heavily on triangulation, induction and judgement. However, despite these limitations, the timeline of the evaluation provided an opportunity to assess the sustainability of the project.

Conclusions and Recommendations:

The Cotton Corridors project is an innovative example of a holistic, community-based and government-led approach to addressing the complex issues of child protection in two districts with large numbers of working children. The project rightly identified the needs of, and major problems of, children in Raichur and Kurnool; adopted a gender-sensitive approach to programme design and implementation with its focus on young girls; introduced innovative models for dealing with child work and schooling; was very effective at changing the attitudes and approaches of government officials; and leverage government expenditure for child protection. However, the sustainability of these positive changes is not assured either at community or government-level, and the changes in child protection outcomes remain fairly small — reflecting the enormity and difficulty of the challenge.

UNICEF and Government have much to learn from this project. In particular, first the challenges of improving child protection are substantial, and this project underlines the importance of taking a holistic approach. Enforcement drives alone, innovative education approaches alone, IEC and community engagement alone will not be effective. Yet even in combination, these approaches are still limited by income poverty and employment options for most households. Second, the model of joint implementation led by a government official was appropriate for getting government involvement and funding, but closer involvement from UNICEF staff might have strengthened aspects of the implementation and funding. Finally, community- and NGO-led approaches can be very effective, but the effectiveness depends substantially on the strength of the community and NGO (and individuals within them), and on the working relationship with government.

Since the project has ended, we do not make recommendations that are directly applied to this project, but instead recommendations for further attempts to improve child protection in AP and Karnataka. We prioritise five recommendations, the first three for UNICEF and the last two for the government:

  • UNICEF should continue to work with government and NGOs on child protection issues in AP and Karnataka, because they have not fully been addressed and will take much longer to do so. The model used in this project is largely appropriate, but more direct project management from UNICEF staff (in terms of capitalising on UNICEF’s technical expertise on CP, being more rigorous with regard to M&E etc.) and a longer period of engagement will increase the probability of sustainable change;
  • for future UNICEF projects, a clear exit and sustainability plan must be designed and carefully implemented, supervised by UNICEF staff; 
  • it becomes imperative that a child protection strategy (targeted at eliminating child labour) acknowledges and incorporates the angle of financial constraints on households that might disable them from adopting favourable child protection behaviours; 
  • government officials in both states should attempt to replicate, and institutionalise, the integrative practices that were developed around child protection during the project. This is particularly important in tackling government level coordination that may arise during the project duration; 
  • inter-district and inter-state linkages will be important in tackling the impacts of child migration. Since migration occurs across districts and states, this will require the development of comprehensive policies that link social protection scheme across these boundaries.

Full report in PDF

PDF files require Acrobat Reader.



Report information




Child Protection - Child Labor



Sequence #:

New enhanced search