What information is available about children in child labour?
To understand from the data how different CLMRS features and modalities affect a system’s outcomes and impacts, ICI requested CLMRS implementers to share anonymized extracts from child labour monitoring data, disaggregated at the child level. Four companies implementing CLMRS shared disaggregated data points from five different projects in Ghana and Côte d’Ivoire. These were compiled into one data set together with data from the ICI CLMRS data base from 7 different projects. Table 1 below provides an overview of the number of child and household interviews available in this compiled data set, by project. This section discusses the information is contained in this compiled data set. Results from a multi- variate analysis of these data are presented in section 2.
Table 1: Overview of system coverage, according to data shared
|Project||Country||# households covered||# child interviews||# children visited|
more than once
As a first observation, it emerges as common CLMRS practice in the sector to collect key additional information about the child and the household alongside information on children’s engagement in child labour, to put the child’s situation in context. All records from CLMRS interviews shared contained basic demographic information on households and children (such as age and sex, level of education, etc.) and the child’s schooling status. Almost all records contained information not only on whether children engage in child labour, but also what types of hazardous tasks they have done, and the number of hours they work per day, or per week.
According to a data mapping exercise carried out prior to the actual data sharing, 53 the majority of CLMRS also keep track of which household or child has participated in awareness raising, and which household or child has received which remediation service. However, the actual data shared by companies did not contain this level of detailed information and could therefore not be included in the analysis of effects of different types of support.
What child labour information do different CLMRS projects collect?
Different CLMRS projects in the cocoa sector collect different types of child labour information. Depending on the design of the data collection tools, a few CLMRS projects identify any case of child labour, while most identify only cases of hazardous child labour (see table 2). In order to identify whether a child is in child labour, data on the number of hours worked, the nature of the work done, and the child’s age need to be collected. In order to identify cases of hazardous child labour, the system only needs to collect information on the types of tasks done by children. Within the compiled data base including granular data from monitoring visits, the number of hours worked by children is available for 9 out of the 12 projects, and the type of hazardous tasks done is available for 11 out of 12 projects.
Table 2 : Child labour information available from different CLMRS projects
|Project||Country||Child labour||Hazardous child labour||Hours worked||Hazardous tasks||At risk of child labour (definition)|
|2||CDI||X||X||-||X||no birth certificate and/or 6-16 year old out of school|
|3||CDI||-||X||-||X||doing light work|
|9||GHA||X||X||-||-||as defined by GCLMS|
What are the most common hazardous tasks reported by children identified by a CLMRS?
For all CLMRS included in this review, data collection tools are aligned to national child labour legislation and the respective Hazardous Activities Frameworks in place in Ghana and Côte d’Ivoire.54 Nevertheless, hazardous activities are aggregated differently between the different data collection tools used, which is why for the analysis of the compiled data set, we group them into 5 broad categories (following the Tulane and NORC survey research reports):
- carrying heavy loads beyond permissible carrying weight
- land clearing (which includes cutting and felling of trees, clearing of forest, bush burning)
- exposure to agro-chemicals (which includes sale, transport, handling, working with agro-chemicals, and in Ghana, being present or working in the vicinity of farm during pesticide spraying)
- working with sharp tools (which includes using machetes/long cutlasses, working with motorized machinery, harvesting overhead cocoa pods with harvesting hook, breaking cocoa pods with breaking knife)
- other hazardous activities include hunting, charcoal production, and managing animal-draught farm equipment.
Figure 2 shows which types of hazardous tasks are reported most frequently by child labourers in the compiled data set of child interviews since 2018.55 In Côte d’Ivoire, the hazardous task most frequently reported by children under CLMRS is carrying heavy loads, reported by 50% of child labourers, followed by the use of sharp tools, reported by 37% of child labourers, and land clearing tasks, reported by 23% of child labourers. Exposure to agro-chemicals is reported less frequently, by 8% of child labourers. CLMRS interviews in Ghana draw a slightly different picture: The use of sharp tools and tasks related to land clearing are reported most frequently by 40% and 39% of child labourers, respectively. Carrying of heavy loads is reported by 16%, and exposure to agro-chemicals by 8% of child labourers. It is interesting to compare these figures with those resulting from the NORC survey research on child labour prevalence in cocoa-growing areas of Côte d’Ivoire and Ghana. According to the NORC study,56 the use of sharp tools is the most prevalent type of hazardous task among children in cocoa farming, done by 31% of children in agricultural households in Côte d’Ivoire and by 43% of children in Ghana.
Children’s exposure to agro-chemicals is also highly prevalent in cocoa production in both countries, with 19% of children agricultural households exposed in Côte d’Ivoire, and 32% in Ghana. Acknowledging that the NORC data collection was done with particular rigor, notably to ensure that the various types of hazardous activities done by children were fully captured, we assume the NORC figures on prevalence of different hazardous tasks to be the best available approximation to reality. When we compare the CLMRS monitoring interview results to the evidence from the NORC surveys, we conjecture that on average, the data collection techniques applied by the CLMRS in this review lack precision and tend to miss out on certain types of hazardous activities done by children; most notably exposure to ago-chemicals. Improvements could be achieved for example through:
- improving the descriptions of hazardous tasks in the data collection tools
- providing additional training of monitoring agents to ensure that they have a sufficient understanding of the different hazardous tasks and how to identify them
- translating hazardous tasks into local languages, jointly with monitoring agents, to ensure that appropriate words and expressions are used when asking children whether they have done any of the activities.