Target: Eligible voting population
Clients: Specialized Agencies & Companies
Who will come out winning in the next election? What do people really think about political parties and their candidates?
Currently, the electoral timetable in Spain has municipal and presidential elections scheduled in all of the Regions before the end of 2015.
Cosmos has had an intense year developing voter opinion telephone polls for the communities of Andalusia, Madrid, Valencia & Catalonia in order to find out what people think in light of the upcoming elections.
In general, the polls for the different projects and clients were looking to attain these basic objectives:
• Identify people’s main concerns and issues.
• Assess how municipal services are valued.
• Determine how current politicians, candidates and political parties are viewed.
• Learn which political parties are favored as well as which would never be voted for.
• Ask people who they voted for in the last election and who they will vote for in the next.
In line with these objectives, a survey, consisting of 10 to 25 questions, was designed that featured both open and closed-ended questions.
Subsequently a sample target population for the surveys was determined. The size and characteristics of which were established through the use of statistical analysis in order to minimize the margin of error.
Given that it was impossible to reach the entire population due to a lack of time and resources, a representative sample had to be defined.
The size of this representative sample had to be large enough for us to gather a sufficient number of opinions so as to guarantee that it represented the views of the original target population. In order to do so, we looked at an array of variables that were of interest to our clients. Relying on our extensive Database we carried out the surveys, appropriately applying the following factors:
• Sex: A defined number of both men and women
• Age: 18-65 year olds
• City of residence
• Marital status
• Education level
• Employment status
The surveys could be conducted using a variety of methods: telephone, CATI, online & face-to-face. Each and every survey was reported on in detail and recorded in our telephony system for subsequent verification.
In addition to controlling the possible sampling error by adjusting the sample size to the previously mentioned variables, Cosmos trained its agents to know how to avoid interviewer errors that might bias the collected data.
For this to happen, the agents were first given the survey script for review and to clarify any doubts that the agents may have had. Then the training department had to instruct the agents in how to conduct the polls without offering any leading or confusing information or causing any errors on their part.
Furthermore, the teams were overseen by team leaders and supervisors whose purpose was to correct any errors that might have come up in the process and mitigate their effect on data.
The following are a few examples of the political polls Cosmos carried out:
Given that time is of the essence during an election year, the services were usually required to be performed within very short timeframes. Upon completion of the services the results were submitted to a notary for authentication and registration purposes.
Many clients asked whether people were willing to give their voter opinion. The answer to this question was a resounding yes. They were without a doubt open to sharing their thoughts and ideas with us and in many instances did so with true enthusiasm.
Another common question was if the surveys honestly matched the voters’ opinions or if they showed bias. The truth of the matter is that if the statistical analysis for the market research is correctly defined, the questions on the survey are well-formulated (avoiding repetition & partiality, taking into account the most significant variables etc.) and the interviewers do their job with professionalism, then the political opinion polls will achieve their objective of transparency.
Mediante la presente queremos certificar la correcta ejecucion ... ver todo,