CONDUCTING RESEARCH
interview with Marija Dragovic, Project Assistant, Association of Students with Disabilities (ADS) Belgrade

How did the implementation go?
There were 14 people who administered the “survey“ throughout the municipalities in Serbia and Montenegro. We wanted to include all the actors involved in problem of discrimination; line ministries, institutions and city services, but we ended up conducting it in cooperation with centers for social welfare, the Ministry of Human and Minority Rights and the People’s Office. The common reply from the other stakeholders was that people with disabilities were not under their jurisdiction.
In terms of timing, it took us a long time to formulate the questionnaire, more than a month in total. The time it took to get answers from the participants varied but the topic of discrimination can be very personal and people have lots of bottled up emotions so they often have a need to speak at some length.
Once we had the responses from all survey takers, we had to begin the data analysis process. A psychology student assisted us in the analysis of the data as he had experience in conducting surveys. He also helped to teach us how to use SPSS, a statistics computer program for analyzing data.

Which results would you stress?
The understanding of discrimination varies widely amongst people. Some people with disabilities even thought that discrimination meant not receiving the benefits they were entitled to; others understood it as being insulted on the grounds of their disability by a person without one.
It is also striking how a personal experience of discrimination can differ even if the situation and the personal background are practically the same; some people just do not recognize it as such. However, it took us by surprise that the feeling of discrimination was the highest in the most unexpected of places: in hospitals and at physical rehabilitation facilities. At the same time, we expected to find a much higher level of respondents who felt discrimination. Instead we have some 60 percent who answered affirmatively that they have not ever experienced discrimination.
Another important finding: the readiness to go to court with an individual discrimination case was significantly higher among those who felt that they had not been discriminated against. That says a lot.
What are you going to do with the findings of the survey? Will you present it to the general public through the media? Or do you submit the results directly to state institutions and bodies?
For this “survey“ we will publish a brochure in which we present the results. We are also planning round tables across the country, gathering several municipalities together; putting the local NGO’s together with representatives of local authorities.

What were the specific problems?
The length of the questionnaire is definitely something to seriously consider. Looking at it now I see that in some places we have repeated things or have asked the questions in such a way that the responses are unusable. This is where the role of an expert is crucial: they act as a guide, putting boundaries on the questionnaire. In addition, they can make sure that the questionnaire matches the general purpose you are trying to achieve with your research.
Yet, the more resounding problem is how to get a sample from a wider population. There is no database you can use, with names and contacts of people with disabilities of all ages and educational and geographic backgrounds. So, a wider sample was not really possible even though we tried our best to achieve it, even using the snowball method2.
Not having a large source of data, we were forced to get our sample from the members of other organizations of people with disabilities that were willing to cooperate. Anyway, having mainly DPO members as the sample also produced biased results. Members of the disability movement are far more aware of human rights issues and discrimination than other people with disabilities and the results show that. For example, 26% of the respondents had higher education, whereas according to data shown in the Poverty Reduction Strategy Paper for Serbia (PRSP), only 7 percent of people with disabilities completed university - a striking discrepancy. The results of the “survey“ also showed that in answering questions pertaining to household income, 42% of the respondents were below the official poverty line, whereas according to the PRSP, it is over 70%.

What would be your specific recommendations, by different stages in conducting a survey: the concept design, the implementation, results analysis, and the dissemination and use of the findings?
Only people with specific experience and training can administer surveys, there is no question about it. Also, in “surveys“ like this, next time we will include the definition of the key concept; in this case, discrimination. If the participants had heard the definition of discrimination, their position might have been more clear and therefore of more use to us.
Also, the team of people administering the survey needs to be very well informed on the philosophy and the ideas behind disability issues. For example, they should be able to explain the social model, plainly and with ease. Also, whoever is planning to do a survey should definitely work with an expert to design the questionnaire. By expert, I mean psychologists and sociologists who have had experience carrying out this kind of research.

1 The research conducted is more of a census of DPO members than a survey as the sample was not determined randomly.
2 The snowball sampling method is a special non-probability method used when the desired sample characteristic is rare. It may be extremely difficult or cost prohibitive to locate respondents in these situations. Snowball sampling relies on referrals from initial subjects to generate additional subjects. While this technique can dramatically lower search costs, it comes at the expense of introducing bias because the technique itself reduces the likelihood that the sample will represent a good cross section from the population. More information available at:
http://www.statpac.com/surveys/sampling.htm