The Great Survey Lie: Why Most Studies in Morocco Measure Nothing

I have led dozens of field surveys in Marrakech, in Safi, in rural villages, in working-class neighbourhoods, and in tourist areas. On every mission, I open with the same question: "Have you ever taken part in a survey, a poll, or a study?". In almost every case, the answer is no.

This is not an anecdote. It is a warning signal. Because in the meantime, dozens of reports, PowerPoint decks, barometers, and studies circulate in ministries, consulting firms, and companies, claiming to represent Moroccan public opinion. If most Moroccans have never been surveyed, whose voice are these numbers really carrying?

Self-selection: the bias that quietly invalidates research

In the scientific literature, this is called self-selection bias (or volunteer bias). It occurs when respondents choose themselves to participate, instead of being randomly drawn from a defined population. Olsen, writing in the Encyclopedia of Survey Research Methods (Sage, 2008), warns that "threats to validity peak with self-selected samples — a category into which far too many Internet polls fall".

Bethlehem, a leading methodologist on web surveys, goes further: without probability sampling, "estimates are often substantially biased", and statistical theory simply ceases to apply. In plain terms: an online survey with 5,000 respondents can be less reliable than a face-to-face survey of 500 people properly sampled.

Morocco's triple trap

The Moroccan context compounds this bias. Three layers stack on top of each other:

1. Uneven coverage — Internet penetration may reach 108.2% in Q1 2025 (DTFE), but 19% of Moroccans remain offline, and home internet penetration in rural areas caps at 47.7% versus 90% in urban areas. The Court of Auditors noted in 2024 that 40% of rural municipalities lack stable connectivity.
2. Digital self-selection — Surveys distributed on Facebook, Instagram, WhatsApp, or by email only reach those who are online, who see the link, and who have the will and time to respond. These profiles systematically differ from the rest: younger, more urban, more educated, more engaged.
3. Motivational self-selection — Among the internet users reached, only those with strong opinions tend to reply. Moderate views, silent classes, and excluded populations vanish from the data.

Result: we are not measuring what Morocco thinks. We are measuring what an urban, young, connected, engaged sub-segment thinks, and then presenting that as the national voice.

The cost for public and private decision-making

The consequences are rarely discussed, but they are serious:

• Misaligned public policies: budgets get allocated based on needs voiced by a visible minority, missing real but silent needs.
• Failed product launches: a brand designs its offer around early-adopter feedback online, only to discover too late that the mass market thinks differently.
• Disconnected political communication: opinion polls end up describing Moroccan Twitter more than Morocco.

What a serious survey should look like

A proper survey is not sending a Google Form to a WhatsApp group. It requires:

• A defined sampling frame (precise target population, explicit sample base).
• Probability sampling or rigorous stratified sampling (by sex, age, residence, territory).
• A physical presence in the field to reach the unconnected and under-represented profiles.
• Enumerator protocols with quality control, supervision, double entry, and random back-checks.
• Methodological transparency: sample size, margin of error, response rate, weighting method.

Without these, we produce decoration. Not knowledge.

The M.PATH bet

At M.PATH, we made a demanding methodological choice: returning to stratified face-to-face, physically walking through neighbourhoods and territories, training our enumerators, controlling quality, triangulating sources. It is longer, costlier, more demanding. It is also the only way to produce data that genuinely resembles the country it claims to describe.

Bad data is always more dangerous than no data: it creates the illusion of knowing. And no one decides worse than someone who thinks they know.
By Ayoub El Asraoui — M.PATH