3.4 Procedures
3.4.1 Refinement and Field Testing of the Questionnaire
Phase 1 of the project was the refinement of the questionnaire that was carried out in Alberta as a basis for the Generations Alberta project (Keating & Munro, 1991). Several drafts of the questionnaire were written in consultation with the North American research team, Alberta Agriculture staff and our collaborator, Dr R. Marotz-Baden, who conducted comparable research in the United States. Demographic variab~s, farm history variable, stress, family cohesion, family expressiveness and the stake scales were all part of the final questionnaire as has been discussed in the previous section. The questionnaire was reviewed and revised many times by a panel of experts to maximise face validity. This questionnaire was then taken for field testing by two researchers in two different parts of Alberta (central Alberta and southern Alberta). Each interviewer administered the questionnaire to four members of a two-generation farm family (male and female of the older generation and male and female of the younger generation). Feedback from these eight individuals was given to the researchers by the interviewers. Respondents reported that the most sensitive questions were those about stress, income, farm debt and total farm assets. However, all of the respondents still chose to answer these questions. Thus it was decided that these questions would remain in the questionnaire but that interviewers would be given special instructions not to press respondents to answer questions that they found invasive.
In its transfer to New Zealand, some changes were made to the questionnaire, specifically to farming terminology, so that the agricultural terms were more appropriate to New Zealand. Four questions were added to the New Zealand study following consultation with MAF policy advisers. These questions were related to the farm families perception of the farm as a possible pension fund.
3.4.2 Interviewer Recruitment and Training
Interviewers were recruited by the project manager and on the recommendation of district MAF consultants. A two day training seminar was held in Christchurch in April 1992. The seminar was conducted by Heather McCrostie Little, the project manager. The four interviewers, each of whom was either resident in the research community or in a major neighbouring town (i.e. Timaru and Christchurch), accepted the invitation to take part in the research project.
Each of the interviewers attended the training seminar. Pre-meeting preparation included familiarisation with the earlier New Zealand study (Keating & Little, 1991). Each of the interviewers was also sent a copy of the initial research proposal so that they were also familiar in advance with the aims and objectives of the research. During the training seminar the interviewees were provided with:
- an overview of the research project the research objectives
- methods for contacting families, explaining the project, and making appointments
- methods for conducting the interviews and managing information
- procedures to ensure accuracy of data collection and delivery of information back to the project manager
- information on how to ensure confidentiality both within the farm family and the community
- mailing instructions for the completed questionnaires via fast post to the project manager
A review of the questionnaire was completed section by section to familiarise interviewers with the questionnaire and the type of questions involved. Role playing was also conducted as a further familiarisation tool. Before the completion of the training seminar each interviewer had the opportunity to take part in a 'trial interview' with a member of a farm family. Following these trial interviews, a dc-briefing session was carried out. A timeline was identified for the interviewers during which interviews were to be completed. Interviewers were also informed that there would be a follow-up meeting at the end of the project at which time they would receive a copy of the final report.
3.4.3 Sample Selection and Interview Procedure
The families who took part in the study were selected by MAF Policy and Agriculture NZ district consultants. Eligibility consisted of the father of the family being between 50 and 70 years old and the children being 18 years or older. The ideal family group included parents, their most involved child and the child's spouse. A maximum of 4 members in each family was interviewed. In all of the families selected the most involved child was male.
The sample is not a random sample of farm families. Selection was purposefully made to include two-generation families who were farming together. It is likely that families who agreed to take part in the study were among those who were most successful at managing generational issues in their business. We presume that families in which there were acrimonious relationships would not be farming together or would be less likely to consent to interviews of all members. Thus the finding from this study cannot be generalised to all farmers but are most likely representative of two generation families with relatively harmonious working relationships.
The initial contact with the prospective farm family was made by telephone by the district consultant who was required to confirm participation of each family member separately. A letter on MAF letterhead but composed by the researchers, was then mailed to each generation, to confirm their participation (Appendix 1).
The interviewers telephoned respondents to set up interviews several days in advance. Each family member was interviewed separately and privately. Interviews were approximately 1 hour in length. Interviews within the same generation and the same families were scheduled back to back if possible. This was done both to deter communication about the interviews until all family members had been interviewed and to ensure effective use of time and energy resources given the considerable distances faced by all the interviewers as they completed their interview schedules.
The interviews were conducted in the homes of the families, usually in an area with table space for writing. The interviewer spent the first few minutes talking with the family member to help participants to become more relaxed. Respondents were asked to fill out the scale portions of the questionnaire (i.e. sections E and F) themselves. However the interviewer was available to explain questions and keep the interview on track. Once the interview was completed, the interviewer could then engage in informal conversation about the study and respond to other questions. The interviewers were encouraged to keep a log recording qualitative information about the interviews relating to ideas not covered by the questionnaire and the general feeling of the interview on the whole.
3.4.4 Data Management
Completed interviews and their interview numbers were stored in a secure area. A list of the families involved in the study was stored in a separate area. Immediately upon completion of the first set of interviews, the interviewer sent the questionnaires via Fast Post to the project manager. It was important for the interviewer to code her initials and the family group on each completed questionnaire to ensure that family groups were kept together. Contact with the interviewer was made when the project manager received the interviews and when the interviews had been checked for accuracy and completeness. This system of interview and Fast Post was carried Out for the first three individual family sets to ensure continued accuracy. Once the project manager was assured that the interviewer was carrying out the interview correctly, interviews were completed and stored securely before being sent by Fast Post in groups of three family sets. In every instance the family sets were mailed immediately after completion.
Data collection began within two weeks of the April 1992 training seminar and should have been completed by August of that year. However several factors prevented that completion date from being maintained. Illness of one interviewer required a delay of approximately three months in interviewing in one area of the project. Two heavy snowfalls in July and August 1992, necessitated a further delay in appreciation of the practical consequences of the snow and its' emotional residue. The interviews were completed in January 1993.
The questionnaire was designed so that coding could be done directly into a computer. Coding and data entry required approximately 2 hours per interview. Coding and data entry began in Christchurch in October and was carried out by Alison Nan who had been involved in coding and entry of the Alberta study. Data entry was completed in January 1993 with data cleaning and initial analysis continuing through April 1993.
Contact for Enquiries
Rural Affairs Coordinator
Sector Performance Policy
MAF Policy
Ministry of Agriculture and Forestry
PO Box 2526
Wellington
NEW ZEALAND
Phone: +64 4 894 0675
Fax: +64 4 4 894 0745
Contact this person
