Introduction: enriching facts and figures.

How do we improve reproductive health? The answer partially lies in enhancing our ability to transform facts and figures into effective programs for individuals, families, communities, and nations. Often, establishing a foundation for this transformation requires a complementary touch: research that is qualitative as well as quantitative.

To understand something "in truth" we need to know not only the facts but the human experience of them. Qualitative research is concerned with meaning: how people interpret their experience and how they use those interpretations to guide the way they live. Qualitative methods have been developed largely within the social sciences, where they are key elements for field-based, observational, descriptive, and explanatory research. For example, in-depth interviews are used to explore topics that are defined by the researcher, while the content of the topic is defined by the person being interviewed. In semi-structured interviews, both the topic and its content are defined by the researcher, but the wording and order of the questions may vary from one interview to another. Structured interviews, in turn, maintain a consistent wording and ordering of the questions. All three types of interviews use a one-on-one conversational approach that gives participants freedom to raise issues that the researcher did not anticipate.

Focus group discussions are similar to semi-structured interviews but involve a small group of people. The group dynamics are often helpful for understanding social norms. Participant observation is a method where researchers place themselves in the social context of the people they are studying, engage in informal conversation, and make systematic observations.

Often, two or more qualitative methods will be combined to collect data on the same phenomenon from multiple perspectives, or qualitative methods will be combined with quantitative approaches. Social science research uses a design strategy called triangulation to combine multiple methods in a way that compensates for potential bias or error in the use of any single method.

Because it brings researchers and the people they study together in conversation, qualitative research is often collaborative and participatory. The research questions and study results may be presented to the participants and their communities for comment and discussion. Community members may help design the research or seek the help of researchers to answer questions of their own. Such partnerships can enrich the research design and also facilitate the translation of research results into community action for change.

All observation is subjective, and we address this fact through careful research design. Quantitative research tends to address subjectivity by using tools that give replicable, reliable answers to specific questions about some aspect of the observable world. It could be said that quantitative research shines a single, bright spotlight in one place. For example, the design of a clinical trial to test the efficacy of a vaginal microbicide for preventing HIV acquisition is strongest when it is narrowly focused on the single question of whether women using the microbicide are less likely to become infected than women not using the microbicide. Qualitative research, in contrast, tends to deal with subjectivity by shining several spotlights from different directions and assessing the amount of agreement and disagreement among the different views. The design of a microbicide program for preventing HIV acquisition, for example, would require information about the way women and their partners make sexual decisions. Collection of that kind of information is well suited to systematic qualitative inquiry; for example, in-depth interviews with both partners that allow for a comparison of their responses, combined with focus group discussions to understand normative expectations about how couples make decisions.

Qualitative research often requires more targeted sampling strategies than does quantitative research, with a focus on determining the range of variability. For example, the strategies may seek to identify the range of variability in the way women negotiate contraceptive use with their partners or the kinds of options that men and women consider culturally acceptable for resolving domestic disputes. Sample sizes reflect the extent to which people differ from each other with regard to the research topic. If a high degree of consensus about a particular topic exists in a community, very few people need to be interviewed. Conversely, when there is little consensus, more people need to be interviewed.

Neither a quantitative nor a qualitative research approach is inherently better. The type of information needed from a particular study should determine which approach to use. In many situations, combining quantitative and qualitative research methods enhances understanding on multiple levels The key issue is whether the conclusions reached are ultimately derived from systematic, scientifically sound data.

Note: Dr. MacQueen, an anthropologist, has conducted both qualitative and quantitative research in support of HIV-prevention clinical trials.

Kathleen M. MacQueen, PhD, MPH

FHI Senior Scientist, Behavioral and Social Sciences