INTRODUCTION
This manual is to help you both during the training and while you are in the field.
Your work as a fieldworker will involve the following tasks:
(1) Finding specific sites (plots) in the areas which have been chosen for the study.
(2) Selecting people for the study from the site once you have found it.
(3) Asking these people the questions from the questionnaire and carefully recording all their answers.
NB: Where questions are open-ended (when have to write their answers in words) you must take down their answers in XHOSA and then translate into English afterwards.
This manual is a guide to the methods and rules you will use for all these tasks except that of administering the questionnaire. You will be given a separate manual for this.
EQUIPMENT YOU SHOULD HAVE
1. a 32cm pencil case
2. a sharpener
3. 2 HB pencils
4. an eraser
5. a clipboard
6. questionnaires
7. Maps of areas visiting with starting point plots marked out
8. Incentives (key-rings) for the respondents interviewed
SAMPLING INSTRUCTIONS
Selecting specific sites and people for the study in the correct way is just as important as asking the questions. The process of selecting people for a study is known as sampling.
What is sampling ?
When we do research we want to find out information about a specific group of people. We call this group the study population. In this initial study, the study population is Xhosa women in Cape Town area townships.
We want to find out as much about the study population as possible regarding the subject we are researching. One way of doing this would be to interview every Xhosa woman who lives in Cape Town area townships, but this would be very difficult as it would take up too much time and money and it would be almost impossible to find every single person and interview them.
A more practical way to get information is to select people from a population who best represent that population as a whole. This is called sampling and can be just as good as using the whole population if done correctly. The group of people chosen for the study is called the sample.
Different methods can be used to select a sample from a population. The methods used depend on many things such as the size of the population and the subject being researched.
SAMPLE FOR THIS STUDY
The sample for this study has already been selected. There are 1380 respondents, 30 of which will be used to test the questionnaire, and another 30 to test fieldworker reliability.
The sample has been selected so that the people to be interviewed are spread out evenly throughout the different townships around Cape Town. We call the people that will be interviewed respondents and in this study all the respondents must be female and Xhosa speaking.
We want to interview 440 people from areas with formal housing and services, 440 people from areas with informal housing and services, and 440 people from areas with informal housing and no services or rudimentary services.
We have also set certain quotas for this study. A quota is when we set the number of respondents that have to be selected according to certain fixed criteria.
We have 3 quotas in this study: (1) AGE QUOTA. We must ensure that half the people interviewed in each area are between the ages of 15 and 29 years, while the other half are between the ages of 30 and 64 years.
(2) SMOKERS QUOTA. We must ensure that 13 smokers are interviewed each day. The total number of respondents in the smokers quota is 312. Due to the age quota, half of these must come from each of the age groups.
(3) SNUFF USERS QUOTA. We must ensure that 14 snuff userss are interviewed each day. The total number of respondents in the snuff users quota is 336. Due to the age quota, half of these must come from each of the age groups.
Excluding the pilot study, the total sample is 1320. We will also be interviewing 28 Xhosa women randomly each day. Random means that there is no quota set for these respondents, they do not have to be a smoker or a snuff user. The fieldworker must follow the instructions for choosing the respondent and they might turn out to be a smoker, a snuff user or a non-smoker. The random set must also be equally divided between the two age groups.
For this study, the exact description of the respondent to be interviewed will be clearly marked on the cover page of each questionnaire to avoid confusion. For example, it may say FEMALE, SMOKER, 15-29 YEARS, or FEMALE, SNUFF USER, 30-64 YEARS.
The sampling has been designed in theory but the fieldworkers will make sure it is carried out in practice. The methods and rules in this manual have been worked out to help you to do this. Following these methods and rules may at times seem difficult and unnecessary, but the success of the study depends on it.
SAMPLING METHODS AND RULES
We will now discuss how you will go about sampling.
Sampling will now involve: (1) Finding a marked site on the map of the suburb.
(2) Selecting people from the site.
1. FINDING A MARKED SITE IN THE TOWNSHIP
The sites are plots, which have been chosen for the study and have been marked on housing maps. They are all residential plots and are either formal or informal, and serviced or unserviced. Some of the maps do not show house numbers and only street names and site numbers. When this is the case, the fieldworker must find the road that the site is on and count the houses from the beginning of the road until the marked plot is found.
Please note: Distances may sometimes be quite far between marked plots on the maps so please wear comfortable shoes and warm clothes. Umbrellas will be provided for the fieldworkers in case it is raining.
We have randomly chosen 8 representative areas from each housing context for the main study, and 2 areas from each housing context for the pilot test. The chosen areas are shown on the following pages, according to the phase of the study in which they will be visited.
Phase 1: Checking the Questionnaire
On day 1 in this area, 54 interviews will be conducted - 18 smokers, 18 snuff users and 18 random respondents. Each fieldworker must interview 2 smokers, 2 snuff users and 2 random respondents.
Formal, serviced area - Ilitha Park
Phase 2: Checking Final Questionnaire and Fieldworker Reliability
On day 2, in each of these areas, 27 interviews will be conducted – 9 smokers, 9 snuff users, 9 random respondents. Each fieldworker will be required to interview 1 smoker, 1 snuff user and I random respondent in each of the 2 areas. A total of 54 interviews will be completed in Lusaka.
On day 3, the same respondents that were interviewed on day 2 must be found and interviewed by a second fieldworker. Only half the questionnaire will be questioned.
Informal, serviced area - Lusaka, Nyanga
Informal, unserviced area - Lusaka, Nyanga
Phase 3: Main Study
From day 4 to 28, in each of the following areas, 55 interviews will be conducted – 13 smokers, 14 snuff users and 28 random respondents. The fieldworkers will work in teams. The way the teams will work is explained on page 10:
Formal, serviced area - 1. Section C & D, Khayelitsha, Town 1, Village 1
2. Section I & J, Khayelitsha, Town 1, Village 2
3. Makhaya, Khayelitsha
4. Nyanga, formal housing areas
5. New Crossroads
6. Old Crossroads
7. Mandalay
8. Gugulethu, formal housing areas
Informal, serviced area - 1. Trevor Vilakazi, Khayelitsha (60%)
2. Harare, Khayelitsha
3. Macassar, Khayelitsha
4. Site B, Khayelitsha (50%)
5. Site C, Khayelitsha (85%)
6. Griffiths Mxenge, Khayelitsha
7. Lower Crossroads (50%)
8. Langa, informal serviced suburbs
Informal, unserviced area - 1. Greenpoint, Khayelitsha
2. Site B, (Site B North) Khayelitsha (50%)
3. Trevor Vilakazi, Khayelitsha (40%)
4. Mkhonto Square & Mpeta Square, Nyanga
5. Browns Farm
6. New Rest , Guguletu
7. Thambo Square & Phola Park, Guguletu
8. Langa, informal, unserviced areas
2. WHAT TO DO WHEN YOU HAVE REACHED THE SITE
You must now choose people for the study from those living on the site. If there are many people living on the site, this may require a number of steps, which are listed below. However, in many cases there will only be one household living on the site - if so you can go straight to step 3.
STEP 1. CHOOSING A MAIN DWELLING
When you arrive at the site, it is important that you first determine the number of dwellings on it. If there is more than one dwelling, then they must be numbered by you from left to right and you must use the random tables provided to choose which dwelling to approach first. (You will learn how to use these tables in the training)
Please Note: If you are in a formal, serviced area then you must not interview anyone living in an informal dwelling in the area. Likewise, if you are in an area marked for our study as an informal area, then you must not interview anyone in that area who lives in a formal house.
STEP 2. CHOOSING A HOUSEHOLD
Enter the dwelling and introduce yourself. Speak to a responsible person and find out if there is more than one household living in the dwelling. A household is defined as a group of people who eat together. Also ask who is at home. If there is more than one household with people at home, pick a household from these using the random number table provided (as you did in Step 1).
STEP 3. CHOOSING PEOPLE FROM THE HOUSEHOLD
For this step you will need to look at the description of the respondent required for the interview. You will find this on the information sheet attached to each questionnaire. A Xhosa speaking female must be chosen in the age group described for each interview. There may also be a requirement for her to be a smoker or a snuff user, but you will know this beforehand by looking at the questionnaire cover sheet. It is very important that the correctly aged respondents are interviewed so please ensure that the right respondent is chosen for each interview.
If there is more than one person in the chosen household who fits the respondent description, then write down all the names of the eligible people in the spaces provided on the information cover sheet on the questionnaire. Remember to write their names down in order, from eldest to youngest, and then using the random number table provided for this in the cover sheets, choose a respondent to be interviewed.
Remember to ask about all members of the household including those who are not home at the time of your first visit. If there is a person who fits the respondent description and they are not home or available, two call-backs at different times of the day or week must be made before a substitute can be found. If a substitute must be found, the same strict instructions for moving to an adjacent site must be followed.
RULE FOR MOVING TO AN ADJACENT SITE
The rule for moving to an adjacent site is that you will go to the house that is on your left as you leave the first site (i.e. if you are facing the road). Remember that will only move to an adjacent site if there is no one suitable for an interview on the first site.
RULE FOR CALL-BACKS
The rule for calling back to find a respondent who is not home on your first visit is as follows:
1. Find out what time the respondent is expected back home.
2. Return at the suggested time and if the respondent is not there yet, you must call-back a second time a little later before a replacement is found.
3. If you are required to find a replacement, a woman must be found in the area who is the same age and the same smoking / snuff-using status.
INFORMAL, UNSERVICED SQUATTER AREAS
You will visit certain shack and squatter areas which have been marked on the map, but for which we have no formal site plans / maps. You will have to select a certain number of dwellings from each area. You must do this as randomly as possible, making sure that you choose the dwellings from all over the area. You must choose dwellings from central parts of the area as well as from the edges.
You will then go on to each dwelling that you have selected and proceed as you would at sites in the mapped areas.
THIS IS HOW YOU WILL WORK
You will work in three teams. One team of 2 fieldworkers will be doing the smokers quota, another team of 3 fieldworkers will be doing the snuff users quota, and another team of 4 fieldworkers will be doing the random respondent interviewing. Every week the teams will be changed so that each fieldworker will get to work on each of the 3 teams.