Acapulco: Morena at 43.8% in a poll fielded entirely through social-media ads
Dato Duro Inteligencia measured Acapulco's 2027 municipal election with 800 self-administered digital questionnaires, recruited through targeted ads on Facebook, Instagram and TikTok. Morena 43.8%, Movimiento Ciudadano 11.5%, PRI 7.8%. The methodology note is as interesting as the result.

Eighteen months out from the 2027 municipal election, Dato Duro Inteligencia's study in Acapulco, Guerrero — 800 valid questionnaires, fieldwork November 1 to 4, published November 5 and reported by Infobae — puts Morena on 43.8% of vote intention, followed by Movimiento Ciudadano with 11.5%, PRI 7.8%, PVEM 5.6%, PAN 4.5% and PT 3.1%. Some 11.2% say they would vote for none of them and 9.1% have not decided.
With names on the ballot the picture barely moves and the fragmentation sorts itself out: the Morena-PT-PVEM alliance with Ricardo Salinas Méndez reaches 44.2%, MC with Ricardo Astudillo Calvo rises to 15.2%, the PRI with María del Pilar Vadillo Ruiz stands at 7% and the PAN with Rocío Morales Morales at 5.5%. The undecided share grows to 17.4%, almost double the party question: putting names in does not reduce indecision, it reveals it.
The methodology note deserves close reading, because it is the format becoming standard in Mexican local polling: a self-administered digital survey, recruited through targeted ads on Facebook, Instagram and TikTok, non-probability quota sampling by sex and age, weighted against the INE voter roll and INEGI data, a theoretical margin of error of ±3.5 points at 95% confidence, a 38.6% refusal rate and 14.3% weighted non-response on the main question, funded by the pollster's own resources.
That design is defensible and it is cheap, but it changes what the number means. In a non-probability sample the margin of error is «theoretical»: it describes arithmetic dispersion, not the probability that the sample represents the universe. And recruiting through ads on three platforms introduces a known bias: it over-represents people who use those networks and answer surveys, which in a city like Acapulco is not the average voter. Weighting against the voter roll corrects age and sex; it does not correct the propensity to be on TikTok.
If you field online, publish the full methodology note and call the margin by its name: theoretical. Demographic weighting does not turn a non-probability sample into a probability one; it corrects who answered, not who could have answered. Report refusal and non-response rates, and benchmark your result against a phone or face-to-face measurement before building a strategy on that number.
Summary written by the FOCUS editors based on the original story. Rights to the original content belong to its publisher.
FOCUS courtesy space


