Wednesday, 28 January 2009

Process to Randomly Select Respondents in a Survey - Monday, 26.1.2009

Posted on 27 January 2009
The Mirror, Vol. 13, No. 597

“In September 2008, the Human Rights Center, University of California, Berkeley, began a survey called So we will never forget - A Population-based survey on attitudes about social reconstruction - and the Extraordinary Chambers in the Courts of Cambodia [Researchers/authors: Mychelle Balthazard, Sokhom Hean, Phuong Pham, Eric Stover, Patrick Vinck]. The results of this survey were released to the [Cambodian] public on 21 January 2009 at the Sunway Hotel, in a meeting organized by the Cambodian Human Rights Action Committee – CHRAC. the methodology was a systematic random sampling , made according to general technical standard systems to survey people. In the survey, researchers randomly selected 125 communes out of the existing 1,621. This selection was done proportionately to population size at the communes. After that, the researchers randomly selected 250 villages countrywide from these communes. There were four randomly selected families in each village and one member of each family was randomly selected. As a result, the researches had 1,000 selected respondents from all places around Cambodia. Data collected were analyzed using the Statistical Package for Social Sciences.

“Each survey in Cambodia encounteres many obstacles. A major obstacle that should be considered is that it is most difficult to solve what is a clear selection of respondents, to ensure that there is no partiality and that results of the survey reflect the actual reality in the society. There are many factors in the survey of the above Human Rights Center that needed to be solved. The first aspect is that Cambodia has a patriarchal social structure, and there are many big differences between the various groups of people. There are big gaps between the rich and the poor, the highly-educated people and those who are illiterate, between the city residents and the people in rural areas (the differences between these pairs of groups have many consequences, such as the understanding of society, different living standards, differences in education, population density, and different ways of life …). As a result, it is very difficult to clearly conclude how the researchers defined who is a ‘general Khmer citizens,’ because in Cambodia there are many differences between different communities, and there are many ethnic groups. The survey aimed to study the opinions of all Cambodian people, but to explain who all the Cambodian people are is difficult. Nevertheless, the methodology to select respondents by the Human Rights Center was thoroughly conducted by studying previous surveys of the Khmer Institute for Democracy, and it included scientific research methodology. However, we have to suggest that we have to conduct surveys for each group separately rather than for the whole Cambodian people. Doing so allows us to know well to what extent each group of people knows about the Khmer Rouge regime and the Khmer Rouge Tribunal. If we would get such information, we would find it easy to create outreach programs for the villages to educate the people there further.

“The second point relates to the population density in each province. If we conduct surveys by randomly selecting villages and communes without thinking about population density, we will get partial towards provinces with a small density. The above survey of the Human Right Center uses a selection of villages and communes which is proportionate to the population size of the different provinces. This factor is important to note in the above survey.

“The third point is that a survey has to consider the theory of the environment of the people [related to a center] which states that when people live farther away from populated areas, their knowledge regarding the social situation drops. According to this theory, the population density is divided into three main sections: the core section, the middle section, and the outer section. In general, people living in the core section are highly educated, rich, and knowledgeable in many social skills, and they influence people living in the middle and the outer sections. There are many reasons for these results. The first reason is infrastructure: where generally people in the outer sections find it hard to connect to the core sections, because of many reasons, such as damaged roads and lack of travel facilities. The second reason is communication, which affects the mentality of people living in the outer section, because they do not of have access to television, newspapers, and radio. However, at present, the Cambodian economy is growing and people in the core, middle, and outer sections get closer to one another little by little, narrowing the gap of the mentality of the three sections of people.

“The fourth point relates to the selection of respondents so that those chosen obviously represent Cambodian people. Respondents of the survey of the Human Rights Center of the University of California were at the age of 39.8 on average, and the number of men and women were equal. 69% of the respondents lived under the Khmer Rouge regime and the rest of 31% said that they were born after the Khmer Rouge regime. But according to statistic of the National Statistical Institute, 68% of Khmer citizens are 29 or under. Therefore, the other 32% have spent part of their life under the Khmer Rouge regime. This different handling of the statistics might be partial towards those who spent part of their life under the Khmer Rouge regime. This partiality might affect different responses quite a lot, especially related to the status of being a victim, the understanding of the Khmer Rouge regime, and of the Khmer Rouge Tribunal, reconciliation, and compensation.

“Fifth, the Human Rights Center of the University of California recognized that one shortage of this survey is the problem of not meeting respondents that had already been selected through the random selection methodology. The report of the survey said that 147 families were identified and replaced by other families, because nobody was at home (76%), families refused to give responses (5%), and other reasons (19%). Moreover, more than 297 respondents were selected and then exchanged because they were not at home when the researchers went to interview them (85%). Those replacements strongly affected the random selection. One reason is that those who stayed back could not express their interests, social class, and knowledge. Those who went to work outside might be members of families with more strenuous labor and knowledge than members of families staying home. Therefore, researchers could have received the information that Khmer citizens are not much interested in the Khmer Rouge Tribunal or do not know this court.

“The above points are major points of that the Human Rights Center of the University of California studied, and what different organizations that intend to conduct surveys should consider. Furthermore, other obstacles, such as the interpretation of questions from English to Khmer, and different views between researchers and respondents regarding important ideas such as reconciliation, remembrance, what is a victim, who is a perpetrator, what is the understanding of the Khmer Rouge Tribunal and regime. The last point that all researchers should also be aware of, is that people’s opinions are influenced by different events at villages and by outside happenings. This is a reason which creates gaps between previous and future results.
Rasmei Kampuchea, Vol.17, #4804, 25-26.1.2009
Newspapers Appearing on the Newsstand:
Monday, 26 January 2009

No comments: