Real document
A survey and an editorial disagree: can their numbers share a sentence?
Read a French survey and editorial with different populations and dates, then combine their claims without averaging incompatible percentages.
- C1
- Notice
- Read and answer
- Learn this in the C1 notebook →
In short
Keep each percentage with its population, date and source. Two figures can share a comparison sentence without being averaged. Here, employees at one association and subscribers to another publication are different groups answering different questions.

The dossier below contains a staff survey and an editorial based on a reader poll. The editor argues for remote access to meetings. The survey records one group’s use of remote work. Read the denominator—the group counted—before deciding what can be compared.
Dossier · travail et réunions à distance
Source A · enquête interne · janvier
Population100 salariés de l’association Atelier Nord.
QuestionAvez-vous travaillé à distance au moins un jour la semaine dernière ?
Résultat40 % ont répondu oui.
Source B · éditorial · mars
Population du sondage cité200 abonnés ayant choisi de répondre au questionnaire du journal.
QuestionSouhaitez-vous pouvoir suivre les réunions publiques à distance ?
Résultat70 % ont répondu oui.
PositionL’éditorial défend le maintien d’un accès à distance aux réunions.
Different measures can coexist
Source A measures reported work practice among named employees in January. Source B reports a preference among self-selected responding subscribers in March. The figures differ in population, question and date. A joint sentence should preserve these distinctions rather than invent a single percentage for “people who want remote work”.
Les sources portent sur deux usages distincts de la distance : une pratique de travail dans l’enquête et une préférence d’accès aux réunions dans l’éditorial.
The sources concern two different uses of remote access: a work practice in the survey and a meeting-access preference in the editorial.
A synthesis can compare the broader theme while refusing a false numerical merger. State the relationship at the right level—different uses of distance—then keep the figures bounded by their actual survey conditions.