Document(s) 25 de 27

The fieldwork phase

MODULE 19: FIELDWORK ACTIVITIES

The activities to be carried out between the workshop on proposal development and the workshop on data analysis and report writing consist of field operations and data processing. The facilitator/ resource person for the project should visit the research team at least once but preferably twice, including:

• during the training of research assistants (if required) and pre-test, and
• at the onset of data processing.

(S)he should also be available for consultation by telephone.

The activities that should be completed during the inter-workshop period are discussed below:

1. Briefing of managers, health service personnel and community members concerned

The purpose of the briefing is to obtain support for the project. Such support is necessary in order to obtain resources as well as to obtain permission to collect data. Attention to the following points will increase your chances of obtaining permission to conduct the study and being allocated adequate resources:

• Selecting the relevant audience(s) for the briefings

  It may be necessary to obtain resources and permission for the study at several levels and from different organisations. Briefings should be conducted with:

  — your direct superiors;

  — managers or key persons in the institutions, organisations, and communities that are being studied; and

  — other key persons or organisations that will be involved in the research or utilise the results.

  Note that ‘key persons’ should include the official leaders (e.g., village headmen) as well as unofficial opinion leaders (leader of women’s cultivation group). In a hospital, the nursing sisters in charge of wards are important opinion leaders although they may not be designated as hospital managers.

  Because of the differing interests of these audiences it may be advantageous to brief them separately, using a different emphasis in each presentation.

• Winning support

  Present the project as an institutional project with yourself (team leader) as the advisor or leader of the project. Do NOT present it as a project you have to do as a training or research exercise. The briefing should enable the audience to recognise the benefits the project will bring to their own unit or community. This will encourage them to ‘adopt’ the project and provide support.

  Develop strategies to overcome resistance and generate support. For example, identify a high level officer who is likely to be supportive and invite him or her to the briefing. Evidence of his/her interest will influence others who are lukewarm.

2. Identifying and obtaining project resources

Identify and obtain the resources (manpower, materials etc.) needed to collect data. Refer back to your project document so as to make sure that all the items needed for the study are included.

• Requesting assistance

  Identify specific types of assistance that will be needed and present these requests diplomatically during the briefings. For example, ‘Do you think health centre staff could help fill in two questionnaires per day over a period of 6 weeks?’ is more likely to receive a positive response than ‘I need manpower’ or ‘I need nurses for this study’.

3. Reviewing availability of subjects and information*

It is important to make a personal visit to every site where data will be collected in order to understand the physical and manpower limitations, constraints and special circumstances that could influence data collection. During the visit:

• Discuss with the staff/community members who are on site any routine procedures and patterns of behaviour (e.g., working hours, holidays) which may affect availability of subjects.
• Observe the physical conditions and procedures that are being followed to determine how they will affect your proposed data collection procedures. Remember that data collection will be reliable only if it does not overburden busy staff members or disrupt routine procedures. If researchers familiarise themselves with the actual situation at the site, it is often possible to design data collection procedures that do not interfere with on-going activities.
• Try to use local personnel for data collection, as they are more aware of local customs and problems. They may be less expensive, require less training, and will be less disruptive. Don’t use them, however, in interviews about the quality of their own work with informants who may know them (see Module 10A III (bias)).
• If sources of data include registers, cards, etc., inspect a sample of the data sources so that you are able to modify the data collection tools (compilation sheets) in order for the research team to obtain the data with a minimum waste of time.
• Identify suitable members of staff who can be research assistants and additional supervisors, if the research team itself will not be able to do all the supervision. If data collection has to be done after office hours, remember to devise a system of supervision for those hours as well.

* Sections 3 - 9 are partly based on the following: The Population Council (1970) A manual for Surveys of Fertility and Family Planning: Knowledge, Attitude and Practice. New York, and: Institute Kesihatan Umum (1986) National Health and Morbidity Survey, Supervisors’ Manual. Kuala Lumpur. (Monograph document)

4. Organising logistics for data collection

After finishing an inventory of available resources, the logistics for data collection should be organised. This will involve planning in detail how, where and when data collection will be carried out, elaborating the work plan prepared earlier.

5. Preparing fieldwork manuals

Manuals or instruction sheets should be prepared for:

• Interviews

  The manual for interviews should have instructions concerning the

  — Purpose of the study

  — Role of the interviewers

  — Way interviewers should introduce themselves to informants

  — Interview techniques

  — Interview guide/questionnaire:
  
  • general format
  • clarification of terms and what the research units are (e.g., household, family, respondent)
  • instructions regarding how to ask complicated questions (e.g., whether to mention pre-categorised answers or not and whether to probe for more than one answer or not)
  • instructions concerning how to fill in answers (e.g., the need to write answers to open-ended questions using the words of the informants)

  — Use of the map (if any)

  — Sampling procedures (and what to do if informant is absent, etc.)

• Other data collection techniques

  Guidelines should be prepared for the implementation of any focus group discussions and interactive or projective research techniques that will be used, so that all members of the research team, including research assistants, will follow the same approach.

  There should be guidelines concerning any measurements that will be made, including instructions on:

  — what to measure and how, and

  — how to properly calibrate measuring instruments

• Supervision

  In addition to all instructions given above, the manual should include a separate section on supervision, with directions, for example, on:

  — maintaining a record of attendance of research team members

  — safe-keeping of data and records

  — recording the number of interviews/FGDs/observations, etc., completed each day

  — ensuring the quality control of field work

  — dealing with non-responses and incomplete interviews, and

  — reporting progress at specific intervals, to superiors and/or funding agencies.

6. Training of research team members, including assistants as well as supervisors

The research team including, in particular, research assistants who join just before the pre-test, must be given explicit training. They should not only be able to collect data properly but also understand other procedures such as the selection of sampling units, map reading and data handling. They may also be involved in the pre-test and in the adjustment of instruction sheets and data collection tools after the pre-test.

The training programme usually consists of:

• discussion on the objectives and methodology of the study,
• reading of manuals or instruction sheets prepared for the study,
• interview training (see Module 10B section V and Annex 12.1),
• field experience (this should include participation in the pre-test described below), and
• discussion on data-collection tools and instruction sheets and how they need to be adjusted (based on field-testing).

The research assistants should be trained together with the whole team, including possible additional supervisors.

7. Conducting the pre-test in the research location, with preliminary data analysis and revision of data collection tools

• The pre-test should assess the validity of the data-collection instruments and procedures, as well as the sampling procedures.
• Reread Module 14 before planning your pre-test.
• Arrange for your facilitator to visit during the training of interviewers and pre-test.
• The study may involve the use of a variety of methods of data collection such as

  — collection of data from recorded sources,

  — face-to-face interviews using interview guides/questionnaires,

  — focus group discussions, and

  — measurements or observations.

• Plan to pre-test all your methods. Analyse the data you collect during the pre-test. Finalise and fill in master sheets, including quantitative as well as qualitative data (using key words). Fill in some of the cross-tables. This process will help you make a realistic assessment of the entire data collection and analysis process and will invariably lead to revisions of some of the tools.
• The pre-test should identify scientific as well as logistical problems and constraints. Discuss these with your facilitator.
• Revise the data-collection and data analysis tools and procedures after the pre-test. Arrange for typing and copying or duplicating of the tools. Check all forms for accuracy before duplicating (see Module 10B). Make sure that sufficient materials and manpower are available for this process. If a computer will be used for analysis, prepare a coding manual.

8. Collecting data

Having obtained permission for the study, and having

• obtained the necessary resources,
• trained the team members,
• organised the logistics, and
• pre-tested and modified the data collection tools and procedures,

the data collection can now be carried out.

9. Processing data

After collecting and sorting the data, all data should be checked for errors. The content may be converted into codes or keywords for processing by computer or using master sheets.

The steps during this process include:

1. editing/cleaning,
2. categorising and coding,
3. summarising data on data master sheets, or
4. if a computer is used, writing instructions to the computer analyst concerning data input and analysis.

1. Editing

During editing look for:

• Completeness of responses. (Note that a blank space may mean ‘no response’ or ‘don’t know’ unless you’ve made a category for each of these responses.)
• Logical inconsistencies, correcting them whenever possible.
• The possibility of combining responses, if that is more suitable for analysis.(See Module 13 for guidelines on making scores.)

Editing should be done by the research team or under its direction. If several persons are involved in editing, as in the case of large surveys, an editing manual should be compiled beforehand.

2. Categorising and coding

A coding manual or coding instructions on the data collection tools, if required, should have been completed when the questionnaire was finalised after the pre-test. Look at Module 13 and Module 23 for coding instructions and for instructions on how to process data from open-ended questions, and at Module 10C for processing qualitative data from FGDs with key words.

3. Summarising data on master sheets

After the data have been edited and coded they may be summarised on data master sheets.

• Review the master sheets you developed during the proposal development workshop. Have your questionnaires changed since you made your master sheets? Can you categorise the answers to certain questions that you were unable to categorise before?
• Remember that you can use letters to represent the different categories of your variables (e.g., M for male, F for female), and key words.
• Then fill in your data on the master sheets. Do not forget to include information on missing data and non-responses.
• Prepare frequency counts for the variables tabulated in your master sheets and check if they match with the number of respondents in your sample.

4. Computer analysis

If the study is large, or if there are other reasons for the use of a computer, instructions should be written for the computer analyst.

After editing, coding and summarising the data, a preliminary analysis can be made by hand or using the computer. (See Module 20.)

Trainer’s Notes

Module 19: FIELDWORK ACTIVITIES

Presentation and discussion (10 minutes)

This module has been prepared to provide course participants with a succinct guide that covers all the tasks they must complete during the fieldwork period. The module need not to be presented in detail, but participants should be made aware of its content so that they will remember to consult it during appropriate stages in their fieldwork.

Role of facilitators during the fieldwork

Research teams should receive at least one, and preferably two visits from a facilitator during the fieldwork period. If only one visit is possible, this visit should focus on:

1. Checking the progress of the project;
2. If necessary, assisting in obtaining managerial support;
3. Observing the real-life situation in which the project will be implemented, identifying problems and anticipating pitfalls;
4. Evaluating the proposed methodology for data collection with the group (by discussing as well as by pre-testing in the field) and advising any necessary modifications in the research design (sampling, data collection tools);
5. Assisting in the training of research assistants (if required); and
6. Finalising and trying out the procedures for data processing and analysis during the pre-test, and, if a second visit is possible, assisting at the onset of data processing and analysis.

Suggested checklist

1. Determine whether managers and health staff concerned have been adequately briefed.
2. Determine whether permission for data collection has been obtained. (As resource person you can provide support by making courtesy calls.)
3. Review each of the proposed methods of data collection, reassessing:
   • The sampling frame, sampling procedures and sample size
   • The data collection tools that have been developed. Make sure that the tools collect the necessary data for each variable or theme, while not collecting any unnecessary information.
4. Make sure that the research team has visited data collection sites and identified working procedures and conditions, constraints, possible additional resources.
5. Determine whether the pre-test of data collection has been well planned. (Assist in any further planning, if necessary.)
6. Assist with the pre-test and training of research assistants and advise on:
   • revision of data collection tools;
   • preparation and adjustment of the field work manual;
   • supervision of data collection;
   • editing and coding of the data collected; and
   • processing of data (data master sheets).

   NB: The sequence of these activities is arbitrary. Revision of instruments may occur twice, before and after pre-testing. Fieldwork manuals may be prepared before the pre-test but adjusted thereafter.

7. Make arrangements for telephone or written consultation during subsequent stages of the fieldwork, encouraging the participants to contact you when necessary.

Document(s) 25 de 27