Entrepreneurial Skills

To validate our approach for the informal economy, we conducted a survey among a subset of South African micro-entrepreneurs registered to the Harambee website.

To assess the applicability of the ESCO taxonomy to the informal micro-entrepreneur economy of South Africa, we conducted a series of primary data collection exercises with Harambee users. The core objective of this data collection was to finally address two key questions:

1 - Can all activities in the informal micro-entrepreneur economy in South Africa be associated to at least one ESCO occupation?

2 - Does the ESCO taxonomy effectively encompass both occupation-specific skills and entrepreneurial skills prevalent in the informal micro-entrepreneur economy?

Throughout the data collection process, we collaborated closely with the Harambee team to ensure its relevance to the youth in South Africa. This collaboration encompassed various stages, including conceptualization, survey instrument design, and sampling.

Activity 1: Focused Group Discussions to identify a relevant list of general entrepreneurial skills

Microentrepreneurs constitute a unique segment within the workforce, as they possess two distinct sets of skills. First, they possess occupation-specific skills directly related to their trade; for instance, a hairdresser possesses skills such as hair washing and styling. Second, they exhibit general entrepreneurial skills required for self-employment or own-account work, such as personal initiative and drive, identifying suppliers, resource management, and understanding customer needs.

Harambee facilitated focused group discussions (FGD) with 18-35 year-old, prior or current informal micro-enterprise owners within their network. These FGDs were designed to identify the second set of generic entrepreneurial skills that might be relevant to microentrepreneurs in South Africa

. Ten percent people with disabilities were included in the sample. To encourage participation, a circular was disseminated, inviting eligible individuals to join the FGDs, with selected participants offered R450 for their involvement

Each session featured diverse entrepreneurial activities among participants, ranging from car washing and music production to tutoring, baking, goods selling, and hairdressing. The FGDs were designed as a series of individual exercises combined with group discussions, aiming to gain insights into how young individuals perceive their skills as microentrepreneurs.

Deep-dive - designing the sessions: each group discussion was structured to allow elicit individual and group feedback from microentrepreneurs, while also extracting insights on how they perceived these general skills

The discussions were structured across 5 activities:

Activity 1: Job descriptions and job titles. The objective of this exercise was to elicit how people describe the venture that they run and how they identify themselves in the micro-entrepreneurial space. Participants were asked to write down the different ways they would describe themselves with respect to their business/hustle if they had to put down/look up a job title - one sticky note per title.

Activity 2: Open Entrepreneurial Skills Elicitation. Participants are oriented on how running a business or hustle of their own entails practicing some specific job-related skills but also some general entrepreneurial skills. The objective of this activity was to elicit a collection of general entrepreneurial skills that the participants feel they have needed/need to run their own ventures.

Activity 3: Skills List Resonation. Participants were handed 3 lists of skills, which they are asked to assess individually first. On each list they were to mark out which skills they possessed, skills they do not possess but feel are relevant, skills they feel are unclear/irrelevant to entrepreneurs. Once the three lists were completed, participants were asked to choose which list they resonated with most.

Activity 4: Learned vs. Unlearned skills. This segment aims to get an understanding of which skills are effortful and which ones come more naturally to participants. The exercise is in effect done in two parts: a) group elicitation exercise where the facilitator jots down learned and unlearned skills being called out by participants on the white board + open discussion on the skills being called out; b) as part of activity 5, participants take back the skills they had put up on activity 2 and arrange them on their individual white charts based on their own perception of whether they are learned/unlearned skills.

Activity 5: Individual reflections. Organize individual charts given to reflect inputs. This allowed participants to re-interact with and organize all the pieces they had put up throughout the exercises. At the end, participants were asked to share overall reflections on the exercise in terms of what they learned, the importance of entrepreneurial skills etc.

After the first day, sequencing of Activity 2 and 3 were interchanged. Therefore, Groups 1, 2, and 3 completed the skills list resonation after being primed on entrepreneurial skills through the skills elicitation exercise, and Groups 4, 5 and 6 completed the same in the opposite order i.e. the skills list resonation was completed without any priming discussion on what entrepreneurial skills can include (details on this decision in the next section).

Participants received three skill lists from ESCO and , indicating skills they had, skills they didn't have but deemed relevant, and irrelevant or unclear skills for entrepreneurs. They then chose the lists, and skills within lists, that resonated most with their work.

Creating the skills lists: weighing tradeoffs across needing to combine separate frameworks (ESCO + EntreComp), and attempting to generate a list within ESCO

Tabiya's taxonomy development so far, had been fully contained within the ESCO frameworks, involving repurposing and expanding definitions/titles for pre-existing skills within it. However, when it came to generic microentrepreneurial skills, ESCO did not provide a clear pathway. Some occupations which would most frequently align with these jobs, such as "specialized sellers" listed some of the generic skills we needed but this was incomplete since these were formulated to describe specific occupations, as opposed to a category of employment type.

This raised two questions:

1. Should we use the EntreComp framework for reporting of generic entrepreneurial skills?

2. Would we be able to create a list of generic entrepreneurial skills from existing ESCO skills that functions comparably or better than EntreComp?

For the first, we included the list of 60 threads (competencies) from EntreComp in our exercise as one of the candidates. For the second question, there were two ways we generated the ESCO list: one was to manually survey all skills under most-relevant microentrepreneurial occupations and pick out all generic entrepreneurial skills. The alternate was a text-matching exercise between EntreComp competencies and descriptors with ESCO skills and descriptors may offer a more systematic, algorithmic way of generating the list.

The tradeoffs were that the first offered potentially high local contextualization, and relatively straightforward to do, but it was based on subjective and arbitrary assessment. The second, while more systematic on paper, would require significantly more effort to generate a sensible list. Even then, depending on the success of the text matching, may require human intervention.

The exercise tested all three lists.

The EntreComp list was most preferred, with 16 out of 33 participants (48%) choosing it as the list they resonated with the most. Participants commonly cited the relatability of the EntreComp list as the primary reason for its preference, noting that they could identify with most of the skills on that list. Other positive feedback highlighted its conciseness, clarity, and self-explanatory nature; emphasis on interpersonal skills related to business; and better inclusion of ideas, creativity, and motivation compared to the other lists.

Creating the final list

Based on the FGD findings, EntreComp was a clear starting point for entrepreneurial skills. However, the primary objective of the data collection exercise was to assess the relevance of the ESCO framework in this domain. So a hybrid approach was adopted in formulating the final list: i) each skill in the EntreComp list was manually matched to the closest skill in the ESCO framework; ii) this list was then further enriched by incorporating skills from the ‘retail entrepreneur’ ESCO occupation. The final list included ### skills, collectively serving as the list of potential entrepreneurial skills for the primary data collection.

Activity 2: Collecting descriptions of micro-entrepreneurial activities via an online survey

The second step of our work with Harambee was about deepening our insights into income-generating activities undertaken by young South Africans.

To do so, we developed a short online survey for individuals within the SA Youth network. The primary aim of the survey was to obtain accurate descriptions of the main micro-entrepreneurship activities individuals engaged in to earn income in the past 30 days. To achieve this, we included the following question:

"Tell us about the main way you make money. What things or services do you sell? Write a short sentence or two explaining what you do"

We also collected current contact details, the number of income-generating activities, time allocation across different activities, and a brief title for the respondents' primary activity.

Survey design, sample selection and response rates

The 2-5 minute Typeform online survey was sent out in early November 2023, with a one-week period for respondents to fill out the form and send it back. This form was sent via SMS containing a unique link to the form that helped connect SA Youth registration data to the respondents' survey responses. Participants received R10 (~0.6 USD) in airtime vouchers upon successful completion of the online survey.

The survey was sent out to a total of 35,000 individuals registered on the SA Youth Platform. of which a targeted 15,000 had been identified from Harambee's EJ data as being involved in microentrepreneurial activities, and 20,000 were new users who had joined the platform within 3 months leading up to November 2023.

Total sample: 35,000

Total Survey Responses received: 6,670

Response rate: 19%

Total viable responses: 3,259

Of the 6,670 responses received, 1,583 reported having no income-generating activities in the last 30 days, 1,221 provided incomplete/nonsensical answers or were duplicates, and a further 119 were removed due to being older than 35 years.

5.2.3 Matching descriptions of micro-entrepreneurial activities to ESCO occupations

The next step was to match individual's descriptions of activities from the viable responses with existing ESCO occupations. One option would be to perform this matching exercise manually, similar to the process for the formal economy. However, there were over 3,300 individual descriptions and 3,008 potential ESCO occupations to match these to. Not only would this be significantly labor-intensive, these methods are also highly prone to human-bias and error.

[do we have anything that discusses the risks of carrying over small errors/biases in framework or taxonomy development that can have large consequences in the long run?]

To streamline the process and mitigate the risk of bias, we opted for a two-part, sequential matching procedure leveraging artificial intelligence (Ai). In the first phase, we designed an algorithm capable of identifying up to three relevant ESCO occupations for each provided description. In the second phase, we reviewed these matches manually to evaluate the accuracy of these matches. Adjustments were made if a match was deemed inadequate or if we believed that an apparent occupation match had been overlooked.

NLP Techniques for Occupation Categorization: We considered various approaches - including Word2Vec, Doc2Vec, and BERT - but, ultimately, the easiest and most applicable approach for this first exercise was to make a simple prompt to OpenAI’s GPT models. The algorithm used OpenAI's GPT model to analyze all ESCO job descriptions and sort them under the following parameters:

• We restricted the potential ESCO matches to the 4.1-digit level of the ESCO taxonomy. Limited scope ensures details while reducing risks of narrow, non-generalizable data on specialized jobs and skills.

• The algorithm could offer multiple categories if there were several potential matches, but it was instructed not to provide more than three matches.

• Zero matches were also allowed.

In total, 3,236 individuals (99%) were successfully matched to at least one potential ESCO occupation.

Distribution of occupation matched to respondents' answers

Activity 3: Assessing relevance of ESCO occupations and skills for informal entrepreneurs via a phone survey

Finally, we evaluated the suitability of the ESCO taxonomy for informal microentrepreneurs. We developed a phone survey with three main objectives:

1. Objective 1: Verify if the ESCO occupations accurately represent individuals' work and if they would willingly identify themselves with these labels.

   • Evaluation: Participants were presented with each matched ESCO label and its description. They were then asked to consider whether these applied to their main way of making money.

2. Objective 2: Evaluate the relevance of the ESCO skills associated with each matched occupation

   • Evaluation: W.

3. Objective 3: Assess the relevance of the identified list of general entrepreneurial skills.

   • Evaluation: Five general entrepreneurial skills identified during the previous process were randomly selected for each participant.

The complete questionnaire can be found here

For each skill or knowledge tag, we designed several questions in order to determine its relevance to individuals’ work. These included whether individuals considered the skill as essential or optional, their perceived proficiency in the skill, their perception of how well other entrepreneurs in similar roles perform the skill, the importance of the skill to their own work, its importance to other entrepreneurs engaged in similar activities, and the perceived importance of the skill in a formal sector job akin to their work. We also incorporated a section aimed at identifying potential skill omissions from the ESCO taxonomy for informal micro-entrepreneurs. This section prompted participants to indicate if there were any other skills that were essential for their work that we had not inquired about.

Survey design, sample selection and response rates

Data collection took place between November 15 and December 8, 2023. The Southern African Labour and Development Research Unit (SALDRU) was contracted to carry out the primary data collection for the phone survey, with Dr. Jacqueline Mosomi as the PI. Data collection was conducted via telephone, assisted by computer-assisted telephonic interview (CATI) software, and the research study received approval from the UCT Commerce Ethics Committee (COM/00513/2023).

. The objective was to achieve 1,500 successfully completed interviews from the specified sample. Respondents were offered a R50 incentive to encourage survey completion, and the survey was designed to take approximately 20 minutes.

Total sample: 3,236

Total Survey Responses received: 1,515

Response rate: 46.8%

The primary reason for non-response (45.8%) was the participant being uncontactable despite multiple attempts. Another 4.1% requested a call back but did not answer, and only 2% refused to participate in the survey.

Results of the Phone Survey

The results of the micro-entrepreneurship survey seem to corroborate our approach, especially by highlighting the absence of systematic bias regarding own skill evaluation between different subgroups in the data. The full results of the survey can be found here.

Importantly, young South African on average consider that they are more skilled than the rest of job-seekers. This is positive, as it means that they are confident in their own skills.

Contrary to what one may expect, the data does not highlight any statistically significant difference in the estimation of respondents' own skills between males and females. If anything, it shows that female respondents tend to deem their own skills as more advanced, compared to their male counterparts.

Finally, there is no statistically significant difference between respondents' own skill evaluation based on whether or not they matriculated (finished high school).