Right-wing extremist accounts and the Chega newspaper claim that a survey reveals “André Ventura is already leading the presidential race with 25% of vote intentions”, but this result is from a voluntary inquiry, not a survey.
Claim: “Intrapolls survey” puts André Ventura “ahead in the presidential election vote intentions”
On Monday, September 22, Folha Nacional, Chega’s party newspaper, stated that “according to an Intrapolls survey, André Ventura leads the vote intentions for the presidential elections, garnering 25% of preferences.” (https://archive.ph/FU90h).
The news article mentions that “the CHEGA leader is ahead of Henrique Gouveia e Melo and António José Seguro, both with 21%, and Luís Marques Mendes, who registers 20%.”
In the third and final paragraph, Folha Nacional also wrote that these data come from a “survey released by Intrapolls through its official Instagram account,” with data “positioning Ventura as the most likely candidate to win the first round of the presidential elections.”
On the same date, the far-right account ‘Resistência Lusitana’ claimed on social media platform X that “André Ventura leads the presidential polls.” According to this post, with over 128,000 views and hundreds of shares, “the CHEGA leader has 25% of vote intentions” and “the system is in shock”: https://archive.ph/6RPle.
The alleged Intrapolls data was also used by Chega’s communication, which shared a chart with the results and the assertion that “Ventura is already ahead in the first round of the presidential elections.” Here, the party acknowledges it is an “inquiry” by Intrapolls, without any further explanation: https://archive.ph/UTp3U.
Facts: Intrapolls does not conduct surveys, only voluntary inquiries
An analysis of the various Intrapolls accounts quickly reveals it is not a survey company. On X, for example, the project clarifies that “it does not conduct surveys but does electoral analysis, which may be supported by non-representative free participation questionnaires.”
On Facebook, there is also an “important clarification” emphasizing “the page does not and never has conducted opinion surveys representative of the Portuguese universe, in their proper demographic proportions, nor does it have that goal. What it does are electoral analyses partially based on ‘feedback’ from a non-representative sample of the country.” (https://archive.ph/gmun9)
This publication, also shared on X (https://archive.ph/9dICm), explains that “the purpose is to build a large-scale national electoral scenario, with quality superior to other available electoral maps” and admits that “there is often confusion between what is and isn’t a poll.”
“Locally, surveys with random collection and rigorous methodology tend to be more accurate. However, these surveys do not offer the macro scenario we aim to present. It is reiterated that local surveys conducted in accordance with the legal procedures of the Media Regulatory Authority (ERC) are ‘a priori’ more reliable for local analyses than the studies carried out by this page, which have a cross-sectional and large-scale focus.”
The numbers cited by Folha Nacional and Chega can be found in this Instagram post, where Intrapolls presents data and charts with percentage vote support estimates for seven candidates: https://archive.ph/ZJJds.
The accompanying text also presents estimates for the first round, with André Ventura in first place with 25%, Henrique Gouveia e Melo and António José Seguro in second, both with 21%, and Luís Marques Mendes in third with 20%.
On Tuesday, Intrapolls also published “estimated scenarios for a second round with André Ventura,” where the Chega leader “is thoroughly defeated” in “3 potential scenarios,” namely against António José Seguro, Luís Marques Mendes, and Gouveia e Melo: https://archive.ph/LAmls.
This publication also makes it clear that this “study is not and does not intend to be seen as a survey” and that “the sample is not representative of the universe of Portuguese voters, consisting only of voluntary participants” whose responses were processed and adjusted “based on reported sociodemographic characteristics and voting intention.”
This is not the first time Chega or accounts linked to the party have released opinion polls as surveys, or even distorted surveys. A consultation of the most recent ERC deliberations reveals that six proceedings have recently been initiated against Chega precisely for violating the Survey Law: https://archive.ph/rfiUQ.
The proceedings mainly result from “violating rules on the public disclosure of opinion polls,” namely “due to the absence in publications of the express warning that such results do not allow, scientifically, generalizations, representing only the opinion of the respondents,” but there is also a case of “distortion of results” from a Catholic University/CESOP poll for RTP.
In one case, the ERC even initiated an official procedure against the company Sondelcerto, whose opinion poll allegedly conducted among the Portuguese community in Canada was widely disseminated by Chega in early May.
In this process, ERC detected “strong indications” of “document forgery and computer forgery” by Sondelcerto, “with the intent to influence electoral perception,” which is why it referred the case to the Public Prosecutor’s Office at the end of July.
Contradictory
Lusa Verifica contacted the director of Folha Nacional but received no response to the question about “why the article violates the Survey Law by omitting mandatory publication elements, namely the technical data, if it really is a survey, or the express warning that those results are not extrapolable to a different universe than that of the respondents, if it is merely an inquiry, also mentioned in the text.”
However, after this contact, Folha Nacional modified the news to eliminate the reference to a survey, replacing it with “Intrapolls inquiry,” but without adding the mandatory warning about the scope of those results (https://archive.ph/fo2W8).
Tiago Campos, founder of Intrapolls, categorically stated, “there is no survey, because Intrapolls does not conduct surveys, only promotes voluntary inquiries” (https://archive.ph/QHD96) and that “no post-announcement survey by André Ventura has been released.”
According to this analyst, the data used by Folha Nacional and Chega is based “on two inquiries, the most recent with 471 responses and the older one with 1873 responses, totaling 2344 responses,” but the results “are not solely intended for voting but mainly for electoral expectations.”
Regarding Chega’s use, Intrapolls “would prefer they not publish the scenarios” in that way. “I don’t decide what they publish. There have been issues in the past; I, along with the team at the time, stated what needed to be said, clarifying what the scenario referred to. From there, I ignored it because I need a completely neutral position regarding all parties and because I feel there’s not much I can do,” says Tiago Campos.
On quality, this analyst believes he has a proven model with some protection against manipulation attempts. “One or two responses manage to elude the control I conduct, but generally, it’s not possible to distort the results,” he says.
ERC, for its part, confirms that “Intrapolls is not accredited to conduct surveys,” but has not yet clarified the specific outcomes of the violation proceedings already initiated against Chega for Survey Law violation.
In all elections, the regulatory authority has reminded the ‘media’ and other stakeholders of the rules applicable to disclosing surveys and opinion polls, as it did concerning recent legislative elections : https://archive.ph/eg5w9.
Lusa Verifica Assessment: False
So far, no survey has been submitted to ERC placing André Ventura at the forefront of the presidential race, namely in the first round of the presidential elections scheduled for January 18, nor any survey conducted after his Presidential candidacy announcement.
As expressed by the organizers, the opinion surveys disseminated by the Intrapolls project are not surveys. More than voting intentions, they mainly reveal electoral expectations and do not allow generalizations or extrapolations to the national electorate, representing only the opinion of the respondents.
