How credible are opinion polls and what criteria are required ‎to accurately express public opinion?‎

Dr. Samir Abu Rumman in an interview with Al-Yarmouk Satellite Channel

What are the criteria that make any poll accurately reflect public opinion?

– Opinion polls are a multi-tasking and multi-skilled process, and therefore there are many criteria by which we can judge the quality of opinion polls and the extent to which they express public opinion. In terms of quality and accuracy, two types of errors must occur: the sampling error, meaning that the whole community was not taken or asked, but a specific sample was taken, and this process has an error of 2 or 3% according to certain criteria , the other type is errors such as the way the question is asked, the method of data collection, the way the respondent understands the question, and other errors called non-sampling error, these two things are important to realize that when we talk about opinion polls, they do not It expresses public opinion with a degree of accuracy up to 100%. There are other criteria that govern this quality. How was this sample chosen? How did you ask these questions? How was this data interpreted and how was it analyzed? How the questions are built to reflect the different cultures present in the language and others, so we always say that public opinion measurements need a diverse team; In order to mitigate these errors and achieve these standards to the greatest possible degree, in order to be able to judge the quality of this industry and this important scientific tool in expressing the opinions of peoples and people.

What are the mechanisms for selecting samples from those whose opinions are surveyed, or in another way, how can this sample be representative of the segments of society as a whole?

– This is an important question in fact, and the selection of the sample is one of the very important stages in improving the quality of public opinion measurements. From an ideal point of view, there are two basic conditions for saying that this sample is well representative of the community, the condition of randomness and the condition that this randomness be in a systematic manner Therefore, these samples of high and ideal quality are called representative samples of the community; Because there is a possibility that everyone in the community had the same opportunity to be chosen, and this choice was not voluntary, but rather the research team chose them in a random way, and that is why I like the words of the former executive director of the “Gallup” International Polling Organization “if it is done right it is a miracle “if you apply the sample correctly it is a miracle” because you can take a small group to represent the population, but you don’t take it in a biased or unrepresentative way, if you don’t put the population in one basket there is a challenge That the sample really reflects public opinion, and usually gives an example, that if you want to take a community’s opinion about their favorite sports team, you don’t go at the end of a match when the fans leave the stadium, because the result will be that they support one of the teams This or that! You went to the wrong place and chose the wrong sample, and it should be in a place where the diversity in this society is available so that it expresses public opinion, and in the event that the sample represented a group without another, for example, males are higher than females, and in young age groups More than those who are of an older age, above it, the sample is corrected in terms of statistics and its weight, as it is said. With regard to the number, there are several criteria that govern it, such as the percentage of accuracy you want, the financial cost that you can bear to reach a representative sample, the method of data collection, and the percentage of analysis. If you want to know the difference between male and female Jordanians, you need a sample of a small number, but if you want to know the difference between Al-Balqa governorate and the capital governorate or Ma’an governorate, for example, you need a larger sample here; Because the analytical level of the category has expanded further.

Sometimes we follow some opinion polls, local, regional and international, which come out with results that may be related to issues that may be considered taboo or that no one can comment on. How can we ensure that this person who was asked in the street answered the question accurately or without any qualms? The issue is related to the nature of the respondents’ opinions. How can I ask “X” of people in an issue that is considered taboo or forbidden in society? How can scientific studies be sure that this opinion taken from this person can originally accurately express his position?

– In fact, this is a beautiful question, and it is one of the questions that is always asked, especially in some environments that have some limits, whether political, religious or social. How are sensitive questions asked and how do you ensure that the answer is correct? The truth is that there are several recommendations that can be made in general, first in terms of research, not to dare to ask questions that the respondent is unlikely to answer in an accurate manner, and this is sometimes seen in the answer options, as the respondent is not expected to answer about the level of trust in the institutions The security that frightens him is like his answer about other parties that there is no sensitivity in talking about, such as an economic or media institution or others, so even in the options there will be a discrepancy in the respondent’s response, accuracy and frankness among the different options, and therefore it is said I know in which country you are and ask appropriate questions; So that there are no misleading answers, even if the answer is courtesy, as is the case in the example of the level of trust in the security services, sovereign ministries or political authorities in the country, and there is another matter, which is some of the techniques that opinion poll institutions use, to do what It can be called infiltrating the respondent’s mentality. For example, if we want to know to what extent he is afraid of a particular event in the future, he may feel embarrassed to declare that he is afraid, and this can be replaced by the question: Do you expect this to happen in the future? Perhaps The expectation is indicative of fear or other, as well, it is possible to ask about political symbols in a certain way, such as determining a degree on the thermometer or choosing between preference and not, and other things. Questions that do not arouse fear in the respondent at the time they are asked and do not provoke ridicule at the time of the results, and the pollsters must know their self-worth, positively and negatively, in the way of dealing with these questions.