IT 162: Information Technology Research Project 2010
QUALITATIVE METHODOLOGIES
AIM:
The aim of this session is to provide the participants with a general understanding of the various qualitative research methodologies and their application.
OBJECTIVES:
1.0At the end of the session the participant should be able to discuss the advantages and limitations of qualitative and quantitative methodologies.
2.0At the end of the session the participant should be able to discuss at least three (3) qualitative research methodologies that are used alone or to complement conventional surveys.
3.0At the end of the session the participant should be able to discuss methods of selecting participants for - :
3.1In-depth interviewing;
3.2Case studies;
3.3Focus Group Discussions.
4.0At the end of the session the participant should be able to discuss the various types of activities that can be usefully observed by the various qualitative methodologies.
What Is Research For?
Research methodology should be discussed in terms of the fundamental goals of the enquiry.
The contemporary studies are producing a great mass of patterns… the challenge lies in how to hump from the reliable material to an understanding of cause and sequence, e.g., of demographic change. This cannot be done except with an historical perspective and a method for generating and testing hypotheses about change closely related to the particular society being examined.
A good example of the limitations of questionnaires developed in isolation may be the KAP (Knowledge, Attitudes and Practices) studies. Often used to gain qualitative information because the researchers states that financial human resource constraints or “levels of significance” preclude use of other methodologies, seek to gain “sensitive information” through this tightly structured model. Getting answers that are meaningless or facile to put into tables is useless if they are unreliable or unrealistic. Another method of gaining the information needed by the researcher must be employed.
There are a range of in-depth qualitative methods which are better suited to obtaining such information.
WHAT IS QUALITATIVE RESEARCH?It is a type of formative research that offers specialized techniques for obtaining in-depth responses regarding what people think and feels.
It allows you to gain insights into attitudes, beliefs, motives and behaviors of the target group.
It provides answers to “why” questions.
It is interpretive as well as descriptive.
Choices Between Qualitative And Quantitative Data
Qualitative methods permit one to study selected issues, cases and / or events in depth and detail. The fact that data collection is NOT constrained by pre-determined categories of analysis contributes to the depth and detail of the qualitative data
This typically produces a wealth of detailed data about a much smaller number of people and cases. It provides depth and detail through direct quotation and careful description of programme situations, events, people, interactions and observed behaviors. These are collected as open-ended narratives without attempting to fit them into predetermined standardized categories.
Quantitative methods use standardized measures that fit diverse various opinions and experiences into predetermined response categories. It allows one to measure the reactions of a great number of people to a limited set of questions thus facilitating comparison and statistical aggregation of the data. This gives a broad generalization set of findings.
Limits Of The Survey
The survey has one enormous strength, it can cover many people and a large area, so that one can feel satisfied that the results are not unique or aberrant, and confined to a single locality.
Yet the patterns of value are probably largely confined to unambiguous data which are quantifiable by nature, and where reported behavior is not subject to different interpretations in different cultures. Such findings are an essential first step in seeding explanation, but they do not get us very far. An explanation of declining fertility in terms of later age at female marriage does not really get us any closer to understanding why society is beginning to change and what were the earlier movements that began a chain of cause and effect.
The sampling aspect, which ensures representative-ness over a large population, also makes it impossible to study a single person or household in context. The sampling unit, e.g., the husband or wife, may not necessarily be the appropriate unit in all cases, it may ignore the role of other important family members.
There can be problems where closed questions may force respondents to answer in a way that does not express what they feel. Due to time restrictions, there may be no allowance for the respondent to answer the question fully. There can sometimes be problems with language in translation of the questions, and the concepts behind the questions. Surveys may get a narrower range of replies that anthropological studies of the same phenomena.
A fundamental problem is the generalization of hypotheses or assumptions upon which the survey questions are based. Questionnaire construction is rarely based on the concepts of constructing questions that will rigorously test hypotheses put forward as a result of prior research in the culture. Questions are chosen because they have “worked”, and because they may yield clearly quantifiable answers, suited to coding and computers, and this tend towards the over-emphasis of some aspects of behavior relative to others.
Survey analysis is dominated by the research for the statistically significant correlations. This can result in being distracted from the socially most significant point.
1. Open ended responses
This is the most elementary form of qualitative data – i.e. open-ended responses on questionnaires. There are severe limitations to open ended data collected in writing on questionnaires. These limitations are related to writing skills of the persons completing the questionnaire. It is important to keep in mind that the purposes of qualitative and quantitative data on questionnaires are different, yet complementary. The statistics from the quantitative part, the standardized items make summaries, comparisons and generalizations quite easy and precise. The narrative comments from open-ended questions are typically meant to provide a forum for elaboration, explanation, meanings and new ideas.
2. Participant Observation
Malinowski (1967; “A Diary in the Strict Sense of the Term” Harcourt, Brace and Jovanovich; NY) is often quoted as an example of how to conduct participant observation. While participation is a matter of degree, Malinowski stressed that the anthropological field worker should totally immerse him or herself in the lives of the people; and that can only be done through months of residence in the local community. Whenever possible the field worker should master the language of the people, though much to the behavior available for observation is non-verbal. Residence in the research community ensures, as Malinowski suggests, that the field worker observes details of daily life and activity enacted by people who have become relatively indifferent to the presence of the foreigner.
WHEN TO DO PARTICIPANT OBSERVATIONS?- When little is known about the phenomenon;
- When there are important differences between the views of insiders as opposed to outsiders;
- When the phenomena are somewhat obscured from the view of outsiders;
- When the phenomena are hidden from public view.
Preliminary data from participant observations provide the field worker with insights and clues necessary for developing interview questions, psychological tests, or other more specialized research tools. The chronicle of a field project usually consists of the interplay between participant observation and the other modes of data collection.
Fieldwork requires much more than simply being there and passively watching what people are about. A host of other materials must be considered in order to make individual observations useful. By structuring observations and systematically exploring relationships among different events – through interviewing, meticulous eye witnessing, and perhaps administering “test” – participant observation can be converted to scientific use.
Gold (1969), however, argues that the observer may participate in the society under study to different degrees, but that the person should still live in that society for a reasonable period of time, speak the local language, etc., although they may choose not to become personally involved in local politics, decision-making, rituals, etc.
Another important difference between fieldwork and casual observation is in the recording of what is seen and heard. Memory cannot be trusted. Every individual has areas of special interest and expertise that affect habits of observation. One should find out how extensively note taking must be in order to ensure accuracy of recall. One must learn to direct attention to features that tend to be neglected. The field worker must examine with great care the nature of the classifications and concepts employed in field observation.
Primary reporting of concrete events and things in field work should proceed at as low a level of abstraction as possible. A vague note such as “A showed hostility toward B” is overly general, and will be difficult to interpret in later months. Better to write “A scowled and spoke harshly to B, saying a number of negative things, including ‘Get the hell out of here, Mr. B’ He then shook his fist in B’s face and walked out of the room.” Thus the field worker should describe the observations themselves rather than the low-level interferences derived from the observations.
Participant observation is not the whole answer, but it is part of the answer, and there are certain areas where it has great strengths. Caldwell quotes the example of the difficulty of obtaining correct information on the distribution of the quantity and type of food within the family except by direct observation. For example, if one is interested in differential feeding of boy and girl children (e.g., if son preference) parents may reply to a direct question that they feed both sons and daughters equal amounts of food, but observation may show a different, unequal distribution pattern. The reason being that the population suspect that behind questions on the subject there lie beliefs in equity and accordingly they provide incorrect and vague testimony. The Caldwell’s also used it for measuring such things as labour inputs of different members of the family.
Caldwell uses the long, semi-structured conversation which may go on for hours. It is flexible and follows what the respondent considers important, though also being directed by the researcher. They found that other people, relatives, neighbors, usually joined in, and maybe replaced the initial respondents. This was regarded as an advantage, rather than a problem where respondents might be tempted to change their story to fit the expectations of others.
Choosing five families and spending a few days with each may tell more about the health beliefs and practices in a community that a survey covering 500 families. Observations may be made on diet, access to food, cash flow, relations within the family, water storage and use, the disposal of human and household wastes, and beliefs about health and illness. It is well known that people behave differently that they say they do.Participant observation allows the researcher to quantify observed activities but, more importantly, to see the context in which they occur and the pattern they represent.
3. Life Histories/Stories
Allowing people, especially women, to tell their stories gives considerable insight into what they consider important. By studying the life histories of only 20 women, one can determine patterns of reproduction, uncover women’s feelings about marriage, childbirth, contraception, and about their image of themselves and their roles in society. Furthermore, few women have ever been asked to tell their story and are pleased to have someone take an interest in them.
4. In-depth Interviews (IDI)
This method is characterized by extensive probing and open ended questions conducted on a one-to-one basis.
WHEN TO USE IN-DEPTH INTERVIEWS.- One has complex subject matter and knowledgeable respondents
- Highly sensitive subject matter
- Geographically dispersed respondents
- Peer Pressure
This is the equivalent of the life history approach widely used by anthropologists. It involves discussions with a small number of respondents selected on the basis of certain criteria such as having had a particular experience, in the case of this study for example, having experienced a child death.
The idea is that by carrying out an extended discussion (up to several hours) around the topic, information may arise which would not ordinarily come out in a formal interview. The time invested in the interview is not simply a matter of how much time is necessary to cover the topic, but more importantly, the researcher must try to establish a bond of trust with the respondent in order that his or her true attitudes and feelings will be verbalized. A further advantage of IDI is that the researcher can follow up leads given by the respondent even though the respondent may not realized their significance. This certainly proved to be the case with the interviews about events preceding the child deaths, where details of symptoms, various treatments and related illness-management experiences could be explored in depth.
An example of how useful information is obtained using this approach came from the 1988 ‘Health of the Urban Poor’ study in Jakarta, where an interview was conducted with the mother of a three year old child who had recently died of measles. In describing the steps taken to manage the illness the mother explained that at stage she had taken the child to a ‘wise’ old woman who sprinkled the child with ‘holy’ water. In elucidating available options in the health service, this old woman would not normally be defined as part of either the traditional or modern health systems, but in fact she acted as part of the traditional system and was seen as such by some people in the community.
The in-depth interviews in these studies concentrated on families where a child had died, where respondents had reported having a child sick recently with: diarrhoeal disease, or parasites (worms), or respiratory infection, or measles, or high fever, or the respondent had a malnourished child. There was also questioning of respondent about their experiences while using various health services.
With in-depth interviews it is necessary to select respondents according to certain criteria, thus it was necessary to have access to information about them before selection. This meant that, unlike the PO and FGD, it was necessary to carry out the in-depth interviews after the household survey.
Interviews, more or less structured, are and easy way to get information on many specific topics, such as family planning and folk potions/ideas about contraception. Interviews with local doctors, herbalists, midwives or community leaders may provide data, not only on commonly occurring illnesses, but also on people’s understanding of their causes and treatment.
Interviews and participant observation are complementary methods of testing the completeness and accuracy of data obtained from one or other methods.
5. Focus Group Research
The FGD is a technique being increasingly widely used. The idea is that a group of informants are gathered together and the researcher guides a discussion through various points of interest. It is important that there not be any single dominant member of the group but rather that all members feel free to present their points of view, even if they differ with other members. Therefore the selection of any group must try to avoid mixing people from different status (power, wealth, etc.) or education categories. The researcher should not try to impose a strict structure on the discussion but rather let it flow around the points under investigation. If a topic such as the provision of health services is under investigation it is important not to include both providers and recipients of the service in the same group, as each have different underlying interests which may result in conflict and suppression of the discussants’ true feelings.
In essence, this methodology capitalizes on group dynamics, allowing a small group of respondents to be guided by a skilled moderator into discussion at increasing levels of depth and varying focuses.
WHEN TO USE FOCUS GROUP METHODOLOGY?- Group interaction is important or felt to allow discussion to more readily occur
- One wishes to include an observational element in the study
- Cost and time factors are important
- Need to generate ideas
- Evaluation of message concepts
- For problem identification and definition
A focus group consists of a small number of participants (usually 6 to 12) from a target population who, under the guidance of a moderator (the researcher), discuss predesignated topics of importance to the particular research study. Although the moderator covers topics according to predetermined guide-lines, the discussion is essentially open-ended. There is flexibility in the order in which topics are covered and leeway to follow up unanticipated lines of discussion central to the theme of the session as they arise.