Development of a platform to reduce apprenticeship dropouts
Ngamboe, Jonathan (2025)
2025
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025051010498
https://urn.fi/URN:NBN:fi:amk-2025051010498
Tiivistelmä
In Switzerland, 24% of young people drop out of their apprenticeship in the first year. This has negative repercussions not only for young people, but also for the economy and society. In many cases, the main problem is their chosen profession. This thesis aimed to tackle this issue by creating a simple online tool to help young people find a fitting career before starting their apprenticeship.
My goal was to build a first version of a platform (MVP) that would guide students through an interest-based test and show the professions best corresponding to their results. I defined the tar-get group as students at the end of mandatory school in Romandy (French-speaking Switzer-land). I chose this region to ensure a focused and manageable scope for the initial development phase, deliberately not including other regions and languages at this stage due to time limitations.
The tool aimed to help students determine their profile using the RIASEC model, a commonly used method to find fitting professions. I built the platform step by step, with modern web tech-nologies to get a simple and user-friendly interface.
I conducted three test phases. The first one showed that the interest test was useful, but that the job suggestions from existing tools were not adapted to the Swiss context. The second phase was based on expert feedback, which led to more relevant suggestions. The third phase led to a functional version of the platform, where users could take the test and access job de-scriptions via external links. The feedback was positive, including the visual aspect and the tool’s simplicity. Nevertheless, users found external websites overwhelming and would have ap-preciated seeing job suggestions more directly on the platform.
The main complexity I met stemmed from the absence of a dedicated Swiss database linking trades directly with RIASEC interest types. This made it challenging to provide fully automatic job suggestions tailored to the Swiss context, and as a result, achieving fully automatic sugges-tions within the project scope was not possible.
Despite this challenge, my project successfully showed a strong need for such a tool and proved that integrating expert knowledge improves job suggestions. Based on positive feedback and the successful test phases, the platform I developed provides a good foundation and should be further developed to contribute to reducing adolescent apprenticeship dropout rates.
My goal was to build a first version of a platform (MVP) that would guide students through an interest-based test and show the professions best corresponding to their results. I defined the tar-get group as students at the end of mandatory school in Romandy (French-speaking Switzer-land). I chose this region to ensure a focused and manageable scope for the initial development phase, deliberately not including other regions and languages at this stage due to time limitations.
The tool aimed to help students determine their profile using the RIASEC model, a commonly used method to find fitting professions. I built the platform step by step, with modern web tech-nologies to get a simple and user-friendly interface.
I conducted three test phases. The first one showed that the interest test was useful, but that the job suggestions from existing tools were not adapted to the Swiss context. The second phase was based on expert feedback, which led to more relevant suggestions. The third phase led to a functional version of the platform, where users could take the test and access job de-scriptions via external links. The feedback was positive, including the visual aspect and the tool’s simplicity. Nevertheless, users found external websites overwhelming and would have ap-preciated seeing job suggestions more directly on the platform.
The main complexity I met stemmed from the absence of a dedicated Swiss database linking trades directly with RIASEC interest types. This made it challenging to provide fully automatic job suggestions tailored to the Swiss context, and as a result, achieving fully automatic sugges-tions within the project scope was not possible.
Despite this challenge, my project successfully showed a strong need for such a tool and proved that integrating expert knowledge improves job suggestions. Based on positive feedback and the successful test phases, the platform I developed provides a good foundation and should be further developed to contribute to reducing adolescent apprenticeship dropout rates.