Recruitment, Training and Organisation of the Field WorkNeil Hawkins
"Palestine as you know is full of uncertainties. The first thing you have to do here in Jerusalem is to find out which particular century anyone else is living in. There are people who still think it's the Middle Ages and claim to have been living in the same house for 1700 years. To us it is 1947, but the Jews are in 5707 and the Arabs have it that it's 1366... On the other hand there are several people living in the next century or two and much ahead of the facts, such as some politicians and press correspondents. So we have given up bothering very much about what year it is, and anyway many of the things that happen in Palestine would be unusual at any time"1.
It was under similar uncommon circumstances that FAFO had to develop a pragmatic structure and organisation for the implementation of the survey, with all its theoretical and technical demands. These demands were not always practical but the emphasis on the quality of the data was uppermost and had to be translated into a workable system. This system also had to take into account the natural limitations of time, money and human resources that exist in any field survey. Finding a compromise that did not sacrifice quality was essential.
The various components of the field work will be dealt with in turn, highlighting the problems faced and the ways in which they were overcome or avoided. The field work proceeded in two stages. Firstly a pilot project was carried out in Gaza in August 1991 to test the questionnaire, the sampling methods, the training course and the control procedures. The valuable experiences gained from this exercise helped FAFO to plan and carry out the second stage, the main survey. The collection of data of the main survey took place from mid-May to mid-July 1992 in both fields. It was important to collect the data in both Gaza and the West Bank at the same time in order to ensure that accurate comparisons could be made. One incident could have altered opinions and answers drastically, thus impairing the desired uniform quality of the data.
As in the pilot project, one of FAFO's concerns was to avoid being linked to and identified with any one organisation or institution. This would have lessened FAFO's effectiveness in the field since it was essential for the success of the enterprise to have as wide a base of support and involvement as possible. An open and pragmatic approach could also be a practical and desirable way to avoid any problems arising out of factionalism.
A further requirement was to hire competent, keen, sensible and, if possible, experienced data collectors. Unfortunately, there are very few trained and experienced data collectors in the occupied territories, and FAFO therefore had to concentrate on hiring people who showed the basic qualities needed.
In November and December 1991 applications were distributed throughout local universities and research institutions. Local leaders were approached and the purpose and implications of the project explained to them. They gave advice on how best to proceed, and their support proved extremely valuable in that it helped pave the way for the acceptance of the project by the local community.
Early on, FAFO considered it important that local institutions should be given an opportunity to use the data gathered for planning and further analysis. Therefore six institutions and universities around the occupied territories were contacted, and in January 1992 FAFO ran an introductory course on SPSS (Statistical Package for Social Science) for two representatives from each institution. FAFO also shared in the purchase of a computer equipped with the SPSS programme for each of the institutions. The aim was not only that these institutions would receive a copy of the data base at the end of the project, but also that they should have the capacity and training to use it. A welcome result of this course was the increased cooperation received from the institutions at all stages of the project, especially the recruiting stage.
FAFO received a total of over six hundred applications. Each application was checked and graded by FAFO and a short-list of 350 names was prepared for interviewing. In December and January each of these applicants was interviewed by two FAFO members in Arabic. The aim of having two interviewers was to avoid accusations of bias.
Each interviewee was graded and FAFO then chose the best in each area, bearing in mind the requirement that at least fifty per cent of each of the teams be female. In Gaza fifty per cent of the field workers were women, while in the West Bank and Arab Jerusalem the female proportion was around seventy five per cent. The candidates who scored highest in the interviews were marked down as potential supervisors and assistant supervisors.
Some candidates who were qualified were found to be working as social workers, and FAFO decided not to employ these. Social workers assess families and recommend aid, and FAFO did not feel it would be in the surveys' interests to employ people who would be known in their areas as social workers. This could easily have led to biased answers if families exaggerated their needs and the severity of their conditions in the belief that the survey might qualify them for aid.
FAFO staff and consultants firstly produced a booklet covering the main aspects of field work, namely: The role and importance of the data collector; the aim of the project; technical aspects and definitions of terms used; confidence-building steps, and an explanation of each section in the questionnaire.
In February 1992, a twenty-hour course was held for the supervisors in the West Bank. The aims were to give the candidates the background to the project, and to provide an in-depth knowledge of the questionnaire, instructions on sampling, group management, confidence building steps in the local community, practical approaches to problem solving and office procedures.
It must be emphasized that all these courses were a two way exchange, as FAFO relied on brief lectures followed by group discussions. This way local knowledge and experience shaped the final outcome of the course and the questionnaire. Contributions and comments from the candidates on all the courses were invaluable, and helped give the participants a feeling of involvement in the project.
Employment was not guaranteed until the candidates had successfully completed the course, which was treated as part of the interview. This led to increased commitment and participation in the course.
Following the supervisors' course FAFO held two courses for data collectors in the West Bank, and one in Gaza. The latter course also acted as a refresher course for the original data collectors from the pilot survey. The field workers were required to complete seven test questionnaires at addresses selected by their supervisors. These addresses were obtained using the maps generated from the supervisors' training course. Only when these questionnaires were accepted by the supervisors and the FAFO-coordinator were the field workers employed. Those who did not achieve the minimum standard were not employed.
The aims of the data collectors' courses were similar to those of the supervisors' course. However, the emphasis was more on the solving of practical problems and learning how to deal with reluctant and suspicious respondents, in addition to gaining an understanding of the ideas behind the questions. The method of role playing possible situations was used to train the data collectors in how to deal with all sorts of respondents and to encourage them to anticipate problems. Time was spent teaching the candidates how to explain questions dealing with opinions without subconsciously influencing an answer, in other words keeping them as neutral and objective as possible at all times.
The data collectors course also provided the supervisors with the opportunity to establish their authority and become the focal point of the group. They were assigned minor administrative tasks and also led the discussions, putting into practice what they had learnt in their course, such as giving everyone a chance to speak, being encouraging and being firm but fair.
FAFO's hope was that an increased understanding of the project and the questionnaire would lead to increased motivation, which in turn would produce dedicated and skilled field workers.
As the areas that were to be covered by the survey are geographically separated it was decided to have two field operations, one in Gaza and one in the West Bank and Jerusalem. FAFO's Middle East Coordinator had overall responsibility for the field operation and was also West Bank and Arab Jerusalem Coordinator, with Gaza having its own Coordinator.
FAFO engaged several local consultants to assist in various stages of the project, in particular recruitment, developing the questionnaire and running the training courses. Due to the difficulty of travel between areas as a result of distance, security restrictions and strikes, it was decided to form six groups of data collectors in the West Bank and four groups in Gaza, each covering a particular area close to where they lived. The problem of travel was somewhat eased by the authorities providing travel permits to data collectors who required them.
The advantage of this structure was that they knew the area, and their proximity not only kept transport costs down but also meant they were able to do more interviews in one day, which increased efficiency.
Each group had between ten and twelve data collectors, and each team had a supervisor and an assistant supervisor. In the West Bank FAFO rented or loaned offices in Jenin, Nablus, Jerusalem, Bethlehem and Hebron to provide venues for each group to meet, in addition to the main office in Ramallah. The supervisors distributed, received and checked the questionnaires in their offices before returning them to Ramallah for entry into the computer. In Gaza, FAFO's centre was based in the YMCA and due to the short distances involved there, was not supplemented by any other offices.
During the sampling stage supervisors went to each survey area and drew maps (see Appendix A). This process served to act as a confidence-building measure in itself because the supervisors, especially in villages and refugee camps, needed to explain the project and its aims before being accepted by the local population.
The initial reaction of many people was that the data collectors and supervisors were from the tax department or were undercover Israeli units. In order to convince people of the data collectors' identity and intentions careful persuasion was needed.
In one instance the local population closed their shops, staged a demonstration and burnt tires as the data collectors approached, believing the data collectors to be tax officials and hoping to scare them off. The data collectors managed to talk to the youths involved and were allowed to continue after delicate negotiations.
Local sensitivities also proved problematic. A particularly religious man refused to let his wife be interviewed so the data collectors approached a person in the local Chamber of Commerce they knew to be a colleague of the man who objected. When the project was explained by the colleague, he consented to his wife being interviewed.
The field workers were assisted by the provision of a FAFO ID card with a photo and with a letter in Arabic clarifying the purpose of the project. The data collectors were carefully trained in how to present themselves, emphasizing FAFO's contact with local institutions and local leaders but being cautious not to promise or imply that any material benefit would follow. This would not only have been unfair, but it would also have encouraged false answers in the misguided hope that by highlighting their condition the respondents may receive aid.
As everyone had to follow the same system using random number tables for any selection, it was possible for each selection stage to be checked. This made it very difficult for anyone to select a respondent out of convenience, not using the designated system, because it could be checked by the supervisors. Any respondents who refused to be interviewed were visited by the supervisor, who would attempt to convince them to answer. This also acted to encourage the data collectors to spend time persuading reluctant respondents.
Supervisors were required to check at least 25% of all interviews. In fact the supervisors controlled almost 30% of all interviews. There were three main types of control. At the beginning, it was emphasized that they should be sitting in with the data collectors during interviews to ensure there were no problems, and to encourage and advise them on how to improve work. The second method involved the supervisors checking the returned questionnaires for mistakes and ambiguities. Finally, supervisors used a short re-interview questionnaire, choosing at random households that had been interviewed and returning to them to check the veracity of the data collectors' observations and the accuracy of the answers provided by the respondent. These measures ensured that interviews were made correctly and that the selection steps were properly adhered to.
A further check was made once the questionnaire had been entered on computer, the details of which can be found in Appendix C of this book.
If an inconsistency was found, the questionnaire was returned to the data collectors for clarification(which may have involved a return visit). Once the questionnaire was returned to the office, any necessary correction were made on the computer. Questionnaires were entered in one centre in Gaza and one in Ramallah almost as soon as they were handed in.
A final check was to re-enter all the questionnaires a second time. The rationale for this was that a value could have been entered wrongly but being "logical" might not have been picked up by the cleaning pass.
Data collectors were restricted to two questionnaires a day unless they
were of a very good quality; then they were given three a day to complete.
This rewarded accuracy and efficiency, for the better the standard, the
more questionnaires they had, and thus the more money they earned.
The motivation of the field workers and supervisors and the system of incentives and controls made an important contribution to the achievement of an extremely low non-response rate and to data of a seemingly high quality.
It is hoped that the data will make a valuable contribution to information on the occupied territories, and not suffer the same fate as statistics in the 1940s:
"Nearly all problems have answers, but no one has yet discovered an answer to the Palestine one, and that is why there are more facts and statistics about Palestine than any other place its size....You can tell from them exactly how many Jewish boys with blue eyes whose parents exported grapefruit to Syria in 1946 now attend Arab schools and vice versa. It seems a pity that from all these figures and graphs and diagrams no one has yet been able to discover what they prove"2