Survey on Living Conditions Among People with Activity Limitations 2002-2003

Recruitment and training
Recruitment of research assistants are carried out by the responsible body in each country (e.g. Technical Team). Specific requirements for research assistants are set in each country, including minimum formal education. Individuals with disability are particularly recruited for the data collection, but the number can vary according to the strength of the national DPO. The number of research assistants and supervisors will vary according to geographical composition of a country and the size of the population.

Screening and Data collection
Research Teams of approximately 5 - 7 persons with one vehicle and a driver travel together and collect data within pre-determined geographical as for instance a Province. Screening is either carried out as a separate activity or directly linked up to the data collection in one operation. In all the sampled areas, every household is visited and the head of the household responds to the screening question. The data from the listing/screening are entered into a data entry program. Disability prevalence is calculated from this file. Any persons who are presented with at least one "some problems" in one of the WG 6 items below qualifies as being disabled. This threshold is chosen to obtain maximum sensitivity of the screening instrument, and the responses to the 6 questions can later be applied to distinguish between impairment types and severity of disability.

Among households with at least one disabled member, a pre-decided number of households in each EA is randomly sampled. Additional EAs are drawn during the sampling process to be used whenever too few households with disabled member(s) are identified in an EA.

Zimbabwe:
Data collection questionnaires that had previously been used in Namibia (on general living conditions – NPC, 2000) and in South Africa (on disability – Schneider et.al., 1999) were combined and adapted for use in Zimbabwe. In addition, a disability-screening instrument was included as well as a matrix on activities and participation developed specifically for this study and drawing on the concepts of the ICF. The design applied in this study in Zimbabwe is similar to the design applied in the previous study in Namibia (Eide, van Rooy & Loeb, 2003), save some minor differences in formulations of certain questions.

User participation was an important element in the design development. This process comprised:

i) A two-day workshop attended by around 25 professionals, researchers, people with disabilities and civil servants who discussed and tested a draft research instrument
ii) Pilot-testing of the research instrument among 150 households with and 150 households without disabilities in two high-density suburbs on the outskirts of Harare, Mbare and Sunningdale (Eide et. al., 2001b)
iii) Further revisions of the research instrument based on experience from the pilot survey and a second two-day workshop including the same resource persons and stakeholders as previously.

After revision, the questionnaire comprised four key elements;
i) household study on living conditions,
ii) screening for disability,
iii) questions to individuals with disabilities including
iv) the ICF based matrix on activities and participation.
The final version of the questionnaire was developed in English. Simple field tests were carried out during training leading to a few adaptations to local dialects.

From the onset, the target population for sampling was all private households in Zimbabwe excluding institutionalised and homeless people. Due to the circumstances in Zimbabwe at the time of initiating the research, including both security issues and a difficult and fluctuating currency market, it was decided to proceed in a stepwise fashion rather than embarking on a full National survey that, due to these circumstances, may have failed. The research exercise and data collection were thus tackled regionally, yielding population-based studies that covered Matabeleland, Manicaland and Midlands. The map on p55 of the resport (provided under the related Materials tab) indicates the geographical areas that were covered by the study.

Data collection was carried out by 4 – 5 teams in each Region, each team comprising 5 – 7 enumerators. The Principal Investigator in Zimbabwe (Dr. Nhiwatiwa) co-ordinated the exercises, supported by the Assistant Investigator (Ms. Jennifer Muderedzi). Each Team was led by a Supervisor who was responsible for the quality of the work in the field and handed in. A total of 80 – 90 enumerators were involved. Important criteria for being employed as enumerator were fluency in English as well as the relevant local languages, and education level to at least 5 good passes at O Level. Care was also taken to include persons who came from the areas where the data collection took place. Not least, efforts were made to recruit enumerators with disabilities through the participating organisations. Approximately 50 % of the enumerators were disabled.

In order to obtain a control sample of households without disabled members, the household neighbouring each of the identified households with disabled members were systematically selected. The two groups in the sample are thus representative for the population of households with and without disabled members in three Regions in Zimbabwe. The sampled households were visited by one enumerator who carried out the interview with the head of the household. It was the intention that the person identified as having a disability should respond to the disability portion of the questionnaire him/herself. This was the case in 53% of those interviewed. The remaining 47% of the disability questionnaires were completed by a proxy. If the situation arose that no one was present at a selected household, then that household was later revisited. Missing information turned out to be a minor problem, as data collection failed in few households.