-
Voting intentions in Colombia in March 2026
Ahead of the presidential elections in Colombia, a poll conducted between 16 and 18 March by GAD3 for RCN Televisión indicates that 35% of respondents would vote for Iván Cepeda, the candidate for the Pacto Histórico. He is followed by Abelardo de la Espriella, with 21% support, and Paloma Valencia, the Centro Democrático candidate, with 16%.
-
Looking at possible run-off scenarios
We asked respondents about their voting intentions in various presidential run-off scenarios.
In a potential run-off between Cepeda and Abelardo de la Espriella, Cepeda would secure 45% of the vote, compared to 36% for De la Espriella.
If the second round were between Cepeda and Fajardo, the results would be 44% and 32%, respectively.
In a hypothetical scenario between Cepeda and Paloma Valencia, Cepeda would reach 43%, whilst Paloma Valencia would secure 40%.
On the other hand, if the contest were between Abelardo de la Espriella and Fajardo, De la Espriella would secure 33% compared to Fajardo’s 25%.
Finally, in a potential second round between Paloma Valencia and Abelardo de la Espriella, the percentages would be 27% and 27%, respectively.
Technical details:
Population: Residents of Colombia aged 18 and over who are eligible to vote in the 2026 elections and have access to a mobile phone.
Geographical scope: Nationwide coverage of the mainland (31 departments and the Capital District of Bogotá). The sample design ensures territorial representativeness and the inclusion of municipalities with the largest populations, in compliance with Article 4 of Law 2494 of 2025.
Data collection procedure: Telephone surveys (CATI) conducted by trained interviewers, using software-assisted applications.
Sample size: 1,200 nominal interviews. Effective n after weighting (Kish): 1,076 interviews
Sampling error: Maximum margin of error ≈±3.0% for the total sample at a 95% confidence level and p=q=0.5, calculated based on the effective number of responses after weighting (Kish: n_eff=1,076). The standard error and margin of error are reported (to the CNE) by indicator in “Annex X. Indicator Quality” (includes valid n per indicator, SE, MoE and DEFF).
Sample weighting: Adjustment of representativeness using two-stage post-stratified calibration (raking). First stage of national adjustment for socio-demographic variables (gender, age and socio-economic stratum) according to DANE 2025 projections. Second stage of national adjustment for electoral behaviour (2022 Vote Recall) based on official data from the Registrar’s Office. The process includes reinforcement of political likelihood ratios and variance control through the limitation of extreme weights to maximise sample efficiency (effective n)
Confidence level: 95%, under the worst-case scenario of p = q = 0.5 and simple random sampling
Sampling frame: Territorial selection is carried out in proportion to population size, based on the National Population and Housing Census conducted by DANE.
In compliance with the provisions of Law 2494 of 2025, the sampling design ensures the inclusion of municipalities and districts with the largest populations, as well as the coverage of those municipalities of regional significance where there are no municipalities or districts with a population of 800,000 or more.
This criterion applies exclusively for the purpose of ensuring the study’s adequate territorial representativeness.
Fieldwork dates: 16–18 March 2026.
Legal entity that carried out the study: GAD3 Colombia S.A.S.
Survey professional: Dr Narciso Michavila Núñez, Javier Sánchez García and Mario Martín Fernández.
