Quantifying the influence of Vocational Education and Training with text embedding and similarity-based networks

TL;DR

It is found that VET courses associated with Singapore’s 4th Industrial Revolution economy demonstrate higher influence than those related to other future economies, highlighting a disproportionate distribution of education supply for the labor market.

Abstract

Assessing the potential influence of Vocational Education and Training (VET) courses on creating job opportunities and nurturing work skills has been considered challenging due to the ambiguity in defining their complex relationships and connections with the local economy. Here, we quantify the potential influence of VET courses and explain it with future economy and specialization by constructing a network of more than 17,000 courses, jobs, and skills in Singapore's SkillsFuture data based on their text similarities captured by a text embedding technique, Sentence Transformer. We find that VET courses associated with Singapore's 4th Industrial Revolution economy demonstrate higher influence than those related to other future economies. The course influence varies greatly across different sectors, attributed to the level of specificity of the skills covered. Lastly, we show a notable concentration of VET supply in certain occupation sectors requiring general skills, underscoring a disproportionate distribution of education supply for the labor market.

Figures (7)

• Figure 1: Schematic of text embedding. a. The diagram shows the embedding process, where textual descriptions of skills and VET courses are transformed into vector representations using a pre-trained Sentence Transformer model. b. Schematic of constructing a course network from the course-skill bipartite network. The course network is the co-skill network of VET courses. c. Illustration of the course network. Each node represents a VET course that has more than one neighbor from the course-skill bipartite network. The node color signifies the VET course category, while the node size demonstrates the degree centrality. Links are omitted in the visualization for simplicity. The colored nodes represent the top 8 categories with the highest number of VET courses, highlighting the most dominant sectors in SkillsFuture Singapore.
• Figure 2: The $k-$core subgraph with $k=100$. The color of the nodes indicates different VET course sectors.
• Figure 3: Course influence and VET specificity. a. Comparison of course influence across different VET sectors. The box plot includes the median, quartiles, and 1.5 interquartile range (IQR) of course influence values for each VET sector. The dashed line indicates the overall median value of course influence over all sectors. b. The relationship between the specificity and course influence at the VET sector level. The horizontal axis represents the median specificity for each VET sector, while the vertical axis represents its median course influence. The text labels and the shade denote the name of VET sectors and the confidence interval of the simple linear regression, respectively.
• Figure 4: Occupational course supply and comparison with transferability. a. Construction of the course-occupation network from the tripartite network of courses, skills, and occupations. Links are connected for entities with embedding similarity higher than 0.6. b. OCS index for each occupation sector in Singapore. The average for each sector is denoted by red triangle, and the blue dashed line indicates the average of the OCS index over all occupations. The box plot denotes the median, quartiles, and 1.5 IQR. c-e. Comparison of the average OCS index and average transferability for care economies (c), digital economies (d), and industrial 4.0 economies (e) across occupation sectors. The colored line and shade represent the simple regression plot and its confidence interval, respectively.
• Figure S1: Average path length between VET sectors within the course network. The heatmap demonstrates the average path length between the top 15 VET sectors. The lighter the plot, the farther the distance between the sectors.