A study carried out by independent researchers found that the report on alleged fraud in Bolivia's Oct. 2019 presidential election by the Organization of American States (OAS) was "deficient," based on "incorrect data" and "inappropriate statistical techniques."
RELATED:
A Washington Post Survey Rules out Electoral Fraud in Bolivia
In the document, entitled "Late vote changes indicate fraud? Evidence from Bolivia," where researchers Nicolas Idrobo, Dorothy Kronick and Francisco Rodriguez participated, the authors state that they reviewed "the qualitative patterns that the OAS and other researchers presented as "inexplicable" and that they concluded that:
- Patterns can be explained, "without invoking fraud."
- Results alone "do not question the credibility of the process."
- The "integrity of the process" was not evaluated," but "only the quantitative evidence" that "played an important role in the evolution of the political crisis in Bolivia."
In early Dec. 2019, following a post-electoral crisis in the Andean country that left two massacres (Senkata and Huayllani), the OAS published its final report on the Oct. 20 elections and concluded that the then-president, Evo Morales, had won irregularly to avoid a second round, not getting the 10 percent difference needed with his closest contender.
According to the international organization, Morales' victory "was statistically improbable," and his proclamation would have been "for a massive and inexplicable increase in the votes of the Movement Toward Socialism (MAS) party in the final 5 percent of the count."
However, according to one of the experts who participated in the study, Francisco Rodriguez, said that "the statistical evidence of the OAS was carefully examined" and "problems with its methods" were found." Once we correct those problems, the OAS results disappear, leaving no statistical evidence of fraud," he said in an interview with The New York Times.
According to this analysis by the Social Sciences Research Network, the "discontinuous leap" found by the OAS after the 95 percent vote count, qualified by the organization as "extremely unusual" and which "questioned the credibility of the process" was the result of two statistical errors which, if correct, "eliminate the appearance of a jump," and it is concluded that:
- An apparent "jump" in the participation of the holder's vote was due to analyst error.
- An inappropriate statistical method was used that artificially created the appearance of a break in the voting trend.
- The analysis of the variation of votes mistakenly ignored a strong trend.
- In the 2019 elections, almost identical patterns appear in the data of the 2016 election, when a constitutional referendum was held, which was not contested.
- The "surprising" tendencies in the votes counted late can provoke conflicts and, in the case of Bolivia, "dramatic political consequences."
This is not the first study to question the OAS report. In Nov. last year, two international studies were released, one carried out by the Center for Economic and Policy Research (CEPR) and the other by Walter Mebane, professor at the Department of Political Science and Statistics at the University of Michigan and one of the world's experts in electoral fraud.
Meanwhile, the OAS has defended its report, which resulted in demonstrations that generated a climate of violence characterized by the burning of electoral courts, blockading of roads, calling for a national strike and persecution of Indigenous people, peasants and members of social organizations that supported Morales, which resulted in a coup.