ICATUS X ESCO Mapping
The ICATUS Taxonomy provides a complete picture of how people use their time. Based on the assumption that all activities build human capital, we attempt to connect ICATUS and ESCO.
Last updated
The ICATUS Taxonomy provides a complete picture of how people use their time. Based on the assumption that all activities build human capital, we attempt to connect ICATUS and ESCO.
Last updated
The International Classification of Activities for Time-Use Statistics (ICA-TUS) classifies .
ICATUS activities are meant to partition the universe of conceivable daily activities in an exclusive manner: each activity can only fall into only one ICATUS category. Therefore, the taxonomy provides a solid starting point and strong conceptual structure into our work on the unseen economy. Additionally, as the universally-recognized and accepted standard framework for time-use statistics, it allows for nation-wide and cross-country comparisons of how people use their time.
However, while ICATUS is designed to be consistent in definitions with economic frameworks like the SNA 2008 and ISIC, it serves a broader purpose than defining economic or productive activities. Thus, it is not directly mappable to the ISCO, and by extension, ESCO skills - the core framework we use.
Hence, the first objective of Tabiya and Harambee's work on ICATUS was to match its activities to ESCO skills. In the long run, in concept, this could allow SAYouth users, from the unseen economy or with experience in the unseen economy, to select their main time-uses and be presented with candidate skills to highlight on their CVs and inform a job-matching algorithm.
The teams had two levels at which these matches could be made:
Time-use activity to skill: Matching ICATUS activities directly to ESCO skills using natural language processing (NLP). In this case, skills matched may be spread out across various occupations and not cohesive. Matching outcomes and precision would be systematic but highly reliant on the richness of skills and activity descriptions.
Time-use activity to occupation: Matching ICATUS activities to ESCO occupations. In this case, we would be tentatively attributing all skills under the matched occupation as a potential candidate skill. Given the shorter list of occupations to match to, this could be done manually by team members.
Given the scarcity of the readily available data on platform users' description of their daily tasks, the second option was selected. ICATUS activities were thus each matched to up to 4 ESCO occupations manually by a team of 3 researchers. These matches then allowed the team to derive a candidate list of skills for each ICATUS activity by the ESCO taxonomy.
Assigning ESCO occupations to ICATUS activities was based on similarity between their definitions. For instance, "preparing meals and snacks" in ICATUS involves - according to its definition - tasks that are those of "cooks" and "kitchen assistants" in ESCO. Therefore, the complete list of ESCO skills associated with "cooks" and "kitchen assistants" was assigned to "preparing meals and snacks" in ICATUS.
The skills list obtained from the first-stage mapping was hardly useable. Each ICATUS activity ended up being associated with a large number of skills. In the example pictured above, "Preparing meals and snacks" generated a first-stage list of 59 skills. These skills fall along a broad spectrum of signaling power and relevance for a jobseeker from the unseen economy: some are too elementary to be worth mentioning such as "washing and cleaning vegetables" while others being too formal-sector specific for most home-cooks to have relevant knowledge of such as "designing indicators for food waste reduction".
In order to provide a manageable list of skills to job seekers on the Harambee Platform 0.1, the second-and third stage of this exercise attempts to trim this tentative list of skills. First, the Tabiya team removes skills deemed obviously not consistent with the definitions and description of the respective ICATUS activities. This was done to provide a relatively leaner and relevant list to experts at the Harambee panel of South African labor market experts and intermediators.
For this second stage skills selection, Tabiya followed 4 rules:
As a first step, we isolated and mainly focus on skills, to the extent that certain knowledges listed in ESCO tend to be redundant with skills i.e. their presence in the list does not bring new information content. For example, if a young job seeker "uses cooking techniques", this implies that they to know "cooking techniques".
However, certain knowledges were kept in for panelists' discretion when they brought important new information. For instance, "common children's diseases" may be deemed an essential knowledge for child care, whether or not one has firsthand experience of having dealt with them.
#2 Delete irrelevant skills:
We exclude skills that had been unduly associated to ICATUS activities on account of the two frameworks not being completely aligned in definition, concept and function. In ICATUS, each activity is clearly defined, including a detailed list of tasks that are included or excluded, along with examples. ESCO follows a similar structure, but ICATUS activities are usually broader and conceptually different from ESCO occupations.
"Budgeting, planning, and organizing duties and activities in the household", an ICATUS activity, is not an explicit occupation and thus absent in ESCO. To align ESCO skills with this ICATUS activity, it has been matched to the ESCO occupations "Office Clerk," "Accountant," and "Bed and Breakfast Operator," which cover tasks related to household management. However, each occupation extends beyond the ICATUS activity. For example, "Bed and Breakfast Operator" includes the skill "serve beverages," a task that is completely irrelevant to the ICATUS activity in question.
#3 Delete skills that are deemed too formal
Some skills inside ESCO occupations directly refer to situations or tasks only possible to do in the seen economy, whether it is formal or informal. For example, "maintain customer service" cannot be appropriately associated with "serving meals and snacks", as a home-cook would not have "customers" and would only be serving meals and snacks to one's own family members or at best house guests. The underlying assumption here is that this process requires using only literal meanings of skills; in this case, unpaid service to family and guests is fundamentally different from having to serve paying customers.
#4 De-duplication
Duplicated skills, both verbatim and in content, were removed from the aggregate list of skills originating from stitching all potential ESCO occupation matches. For example, "Outdoor cleaning" was associated with both "prune plants" and "prune hedges and trees", it is debatable how much marginal information retaining both provides.
However, content de-duplication is not straightforward within ESCO. In the above example, "prune plants" does not contain the subskill "prune hedges and trees", even though hedges and trees are obviously plants. Instead, it contains the subskill "perform hand pruning". Basically, the organization of ESCO itself is not straightforward or intuitive, complicating content-based de-duplication.
