← All projectsStudy · 2026

Le Fossé

Our flagship study: ten Mauritanian statistics and data programmes compared, tool by tool, with 395 real job postings. An audited methodology and a dual score measure how well academic training matches what the labour market expects.

The project

“Le Fossé” (The Gap) grew out of a simple question that had long gone without a quantified answer: does what Mauritanian statistics and data programmes teach really match what the labour market expects? Rather than debate it, we decided to measure it.

We gathered 395 real job postings from the data sector and 430 courses from the curricula of ten programmes, from bachelor’s level to engineering degrees. Every tool requested by recruiters was checked, one by one, against what is actually taught, using a dual score: a strict score (what is explicitly documented) and a broader score (what could reasonably be covered).

The result is a detailed, programme-by-programme diagnosis of the gaps between education and employment, together with concrete recommendations. The methodology is fully audited and traceable: every figure can be challenged line by line.

The study comes in three complementary deliverables: a 94-page report, an interactive dashboard and an application to explore the results by programme and by tool.

Methodology

A method you can check

Every figure in Le Fossé rests on sourced data, written rules and automatic checks. Here is how the study was built, and what it does not claim to measure.

  1. Two documented datasets

    395 job postings collected from 29 platforms between 2 July and 7 August 2026, each with its link, after removing 5 duplicates. 430 courses taken from the curricula of 10 programmes at 7 institutions.

  2. One shared grid

    Postings and curricula are coded with the same four skill types. Variants of the same tool are merged, job families follow published rules, mapped to the ESCO, ISCO, ISCED and ISIC classifications.

  3. A fair comparison

    Each programme is assessed against its own market: every bachelor’s on the same segment of 64 postings and the same tool basket, every master’s on the postings of its specialism.

  4. A transparent index

    The score is the share of demand covered by the segment’s ten most requested tools, each weighted by how often postings ask for it. The formula and the baskets are published.

  5. Two scores, not one

    The strict score only counts tools named in the curriculum. The broader score adds tools a course clearly covers. The gap between them separates a writing problem from a real lack of content.

  6. Everything is traceable

    Every “yes” points to a named course in an audit table. Ten consistency checks run with every calculation, and scores were recomputed with and without duplicates: the difference never exceeds 0.2 points.

What the study does not claim to measure

  • It reads what curricula state, not what is taught in class.
  • Postings are international: national hiring mostly goes through competitive exams that do not list tools.
  • Online postings do not represent the whole labour market.
  • It measures apparent alignment, not graduates’ employability or the skills they actually acquire.

The full compliance grid, point by point, is in Appendix G of the report (in French).