The Survey will run until Friday, October 2. (GFR Average)
The El Nuevo Día Survey was conducted through face-to-face interviews conducted throughout the island, house to house, with a representative sample of 1,000 registered participants who are 18 years of age or older, one for each household. The in-person interviews were conducted between September 19 and 23, 2020.
According to the Census data, all the results of the survey have been weighted in their correct proportion to adjust them to the variations of age and sex in the population, and by patterns of electoral participation.
In addition to the main sample of 1,000 participants, an oversample was used in the municipality of San Juan. The total sample for the municipality of San Juan is 400 participants, made up of 98 participants from the main sample and 302 participants from the oversample.
This type of survey generates a sampling error no greater than ± 3.1% of what would have been obtained when interviewing the entire population registered to vote. The margin of error is greater when describing smaller subgroups within the sample. For example, the margin of error is ± 4.9% for the complete sample of the municipality of San Juan.
The design of the survey, the interviews, and the analysis of their results were carried out by The Research Office, Inc. in Guaynabo, Puerto Rico.
For more information on The Survey, write to [email protected] . The questions and their results will be available from Saturday, October 3 at elnuevodia.com.
What are the eligibility criteria for this Survey?
There is a single main criterion that all the participants that make up the final sample of The El Nuevo Día Survey had to meet: be 18 years of age or older, and be registered with the State Elections Commission to vote in the general elections on November 3 2020.
At this time, the degree of certainty of voting in the elections is not a criterion for eligibility, although some results are presented by those who are more or less inclined to go out to vote on November 3.
Is the sample of 1,000 participants, as in other surveys?
The main sample of the current El Nuevo Día Poll is 1,000 participants, in this case adults registered to vote. A sample of 1,000 participants has a margin of error of plus or minus 3.1%.
In addition to the main sample of 1,000 participants, an oversample in the municipality of San Juan was used to achieve a base of 400 participants. The participants in this oversample only answered questions related to the San Juan mayoral race.
How much is the base of participants in the contest for mayor of San Juan?
In order to properly measure the race for mayor of the capital, it is necessary to oversample to minimize the margin of error. In the case of the municipality of San Juan, the oversample is 302 participants, so that when adding it to the “natural” sample of 98 participants that comes from the general sample of 1,000 participants throughout Puerto Rico, it would reach 400 participants (with a margin of error of ± 4.9%).
Was the San Juan oversample used to ask about the race for governor or resident commissioner?
The questions related to the race for the governorship and the resident police station, as well as the evaluation of officials, figures and political parties do not include the oversample of San Juan. Therefore, these results are presented based on the total base of the main sample of 1,000 participants. In addition, some results are presented with the bases of affiliates to parties and movements and unaffiliated. These details are presented in the narrative of each story, including the size of these samples and their margins of error.
Plus: Check the stories and columns related to The Survey
Do registered voters who are not fully determined to vote qualify?
In this reading, yes, since the main criterion is registration to vote, but those who have said, at this time, that they “probably” or “definitely” would not vote are not excluded. This group is a minority, represented by a base of 28 participants or just 3% of the sample, which does not have a great impact on the results. In the next installment of The Survey , which will be published close to election day, all participants must have a “definitive” or “probable” intention to vote in the November 3 elections.
Will the results presented in this Survey be the same as those of the election?
The purpose of the current Poll is not to declare who will be the winner in the general election on November 3. On the contrary, its usefulness is mainly to present a portrait of how the population registered to vote thinks at a specific moment in time, between September 19 and 23, 2020, in essence, just over a month from the elections. For this reason, it is specified in the narratives that the results would be if the elections were “today”, in reference to the precise days on which the interviews were conducted.
As it may well happen, it is clear that in the coming weeks there may always be events and situations that arise before the elections that make the most determined to vote change their minds, or that make those who – at the time of conducting these interviews – They were still undecided or thought not to go out to vote that day, change your decision, both to go to vote and who to vote for.
What is the margin of error for results other than on the basis of 1,000 participants?
The margins of error change for subgroups, depending on the sample size of each. As a general rule of thumb, the smaller a subgroup, the greater its margin of error.
In the text of the coverage of The Survey of El Nuevo Día , multiple references are made to the sizes of the margin of error for the main subgroups of interest, which are those affiliated with political parties and those who are not affiliated.
Why are Vieques and Culebra excluded from the sample?
This has always been the case with the El Nuevo Día Survey and is done for practical and logistical reasons. With a sample size of 1,000 participants, the corresponding sample in these two island-towns would be 4 participants, a small size that has little impact on the total when excluded. These 4 participants are redistributed in the rest of the sample, according to the population weight of the municipalities.
When will the complete survey questionnaire with its results be seen?
This questionnaire will be available on endi.com the day after the publication of all the results in the printed edition of El Nuevo Día, in this case, on Saturday, October 3, 2020.
Where can I contact if I have questions, doubts or comments about the results of the Survey?
Send an email to [email protected] and we will answer your questions, concerns or comments.