The International Labour Office has become aware of a report circulating on social media, prepared by Kristian Lasslett and Vanessa Gstrein. The report maintains that there are “critical flaws” in the third-party monitoring system relating to forced labour that the ILO is conducting in Uzbekistan.

As in previous years, the ILO will be conducting monitoring for the purpose of eliminating child and forced labour during the cotton harvest in Uzbekistan, which starts next week. At the end of each harvest, monitoring results and the underlying methodology are made publicly available and recommendations are discussed with the Government of Uzbekistan and other stakeholders. Since the ILO started monitoring in 2015, there has been growing political commitment and action on the part of the Uzbek government to fully eliminate child labour and forced labour. The monitoring methodology has been constantly refined and adapted, always with a view to protecting respondents, ensuring independence and supporting the implementation of political commitments.

Actual on-site monitoring of the cotton harvest to eliminate the use of child labour and forced labour is only one aspect of the cooperation between Uzbekistan and the ILO. This cooperation also encompasses large-scale awareness-raising campaigns, capacity building measures, development of complaints’ mechanisms and other remedies and legal and policy reform. Monitoring has been continuously adapted in order to assist in this process.

International ILO monitors as well as Uzbek human rights activists have been trained to engage in the monitoring as soon as the harvest starts. Numerous awareness-raising and educational training sessions have been arranged for local authorities involved in harvest-time activities, including over 7,000 people trained on recruitment and ILO labour standards, 300 labour inspectors, 500 public prosecutors trained on investigation techniques and 250 journalists trained on freedom of speech and labour rights.

All monitoring by ILO experts will respect, as it has in the past, the principles of independence and confidentiality and the need to protect vulnerable persons and groups. A full report, with details on the methodology and lessons learned about its application, will also be published after the 2018 cotton harvest.

Lasslett and Gstrein claim that their analysis is based on a monitoring manual used by the ILO. Actually, what has been reviewed is the report of the monitoring of the 2017 harvest, together with an outdated document from 2015 from the internet. The ILO has a comprehensive and well documented monitoring methodology which includes standards, sampling formulas, data handling policies, ethical safeguards, templates, forms and instructions. These are documents which are intended for the use of the persons carrying out the monitoring under contract with the ILO.

A number of Lasslett’s and Gstrein’s concerns have been addressed in the monitoring of earlier years, and they are addressed in this year’s monitoring as well.

In 2018, particular attention will be paid to further refining the monitoring/research methods and sampling frames. We use a mixture of qualitative and quantitative data.

The monitoring of the cotton fields during the harvest covers observation and interviews to gather qualitative data. The ILO monitors and human rights activists will visit fields to observe working conditions and to conduct interviews with pickers, farmers and local officials.

Interviewees can select the time and venue for the interview. Respondents can use a pseudonym and they are asked not to share any names or personal information of other individuals during the interview. Monitors are strictly prohibited from taking pictures or videos of human subjects. Photographs may only be taken of objects related to working conditions.

Furthermore, monitors are provided with electronic devices (tablets) to record interviews, fill in check lists and work on the write-ups. All collected information will be shared by the monitors on a protected cloud server and deleted from all devices within 24 hours. The device will be protected by a fingerprint or secure code access, so that physical theft of the device will not allow access to the data.

All field interviews are carried out without any government representatives present. In 2018, the monitoring teams consist of an international ILO expert and a human rights activist. Interpreters and drivers are contracted and organized by the ILO.

The 2018 ILO third-party monitoring methodology is being reviewed by an Institutional Review Board (IRB), also known as an Independent Ethics Committee. Appropriate steps are taken to protect the rights and welfare of all who are participating in the monitoring.

Based on the qualitative data collected during the field visits, a set of hypotheses will be developed and tested through nationwide surveys, which are described in more detail below. These surveys are distinct from monitoring and provide crucial complementary information.

Lasslett and Gstrein are concerned that the quantitative survey methodology could have negative consequences on vulnerable participants and raises concerns about the accuracy of data. The ILO study design, (sampling and questionnaires), uses the so-called Item Count Technique, (“ICT” – also known as “unmatched count technique”), when asking sensitive questions from potential victims of forced labour. ICT is an indirect questioning technique which improves, through anonymity, the number of true answers to possibly sensitive, embarrassing or self-incriminating questions. ICT works on the basis that over 3,000 survey respondents are randomly allocated into two groups. One group receives a list of non-sensitive questions, while the other group is shown this same list with an additional question on forced labour added. All respondents are asked to indicate how many, but not which items apply to them. Differences in the means between the two groups are used to accurately estimate the prevalence of forced labour during the harvest.

The concern that questions might be asked and answered in a very short period of time is not valid, as questionnaires used in 2017 were filtered so that the answer to a question determined which questions were asked next. For example, if a respondent answered negatively to the question “Did you pick cotton this year?” there was no need to ask most of the subsequent questions and consequently the time taken for the interview was significantly shorter. Not all questions are asked and answered for all interviews.

Appropriate steps have also been taken to ensure that there is fully informed and voluntary participation by those interviewed for either monitoring or the surveys. This is not by merely issuing a consent form or asking a question but rather carrying out, on the basis of informed consent, a process during which the subject gains an understanding of his or her rights, the purpose of the monitoring or study, the procedures to be undergone, and the potential risks and benefits of participation.

The ILO welcomes constructive criticism of our work and remains committed to conducting monitoring and research in a manner which is systematic, transparent and ethical. We are happy to share further details of our monitoring and research methodology with people who are interested in these matters. Please contact ILO Chief Technical Advisor Jonas Astrup (astrup@ilo.org) if you have any questions.

The ILO Third-Party Monitoring (TPM) Project has been made possible by a Multi-Donor Trust Fund established by the World Bank with support from the European Union, the United States and Switzerland in order to finance activities to support the elimination of child labour and forced labour in the cotton sector in Uzbekistan and to strengthen the capacity of state and non-state institutions to address the sustainability of cotton production and agricultural reform in general.