Evaluation Toolkit for Breastfeeding Programs and Projects

June 2012

4.2 Qualitative data collection methods

Page last updated: 04 November 2013

There are a range of methods which can be used to collect qualitative data.

4.2.1 Interviews

Interviews may be structured (with a standard set of questions) or semi-structured (with broad questions or topic headings to prompt discussion) or unstructured (with a single overarching question or topic to start a conversation, but no particular formal questions). Structured interviews don’t allow the interviewer to veer from the set questions, so the responses they get are more standardised. Most interviews, however, are semi-structured, with clear questions identified but with room to probe for more detail or to ask supplementary questions if something interesting is said.

In qualitative interviews, questions can be open-ended, meaning that they are structured to invite people to talk freely. Open-ended questions usually start with words like ‘why’, ‘what’, ‘how, ‘where’, and encourage people to talk about their views or experiences. Closed-ended questions, on the other hand, are structured so that people can answer ‘yes’ or ‘no’, or answer from a limited choice of responses. Closed-ended questions usually start with verbs, for instance: ‘did you think the service was satisfactory?’, or ‘could you remember what the nurse told you?’ Closed-ended questions lend themselves more to quantitative research because you can usually count, for instance, how many people said ‘yes’, or ‘no’. For qualitative research, where your interest is in people’s perceptions or experiences, open-ended questions generally work best.

Interviews sound simple but they do require strong listening skills, empathy, critical thinking (to query or probe some answers), and a warm and friendly persona to build trust.

4.2.2 Focus groups

Focus groups are essentially group interviews, which typically last for about 60-90 minutes and which are led by a facilitator or researcher. A focus group should be no bigger than 8 or 9 people because if it is too big, participants sometimes feel inhibited about speaking out. A focus group is a good way of learning about a group of people and their experiences (for instance, young breastfeeding mothers, Arabic-speaking mothers, grandparents who are assisting new mothers with breastfeeding). It is important that a focus group takes place in a welcoming environment and where people feel safe and secure to discuss the topic freely.

Participants in the focus group may be selected according to identified criteria by an external recruiter (generally if there is a large population involved) or through an in-house process. Focus groups are not intended to be ‘representative’, but are designed as a means of exploring specific topics. The role of the facilitator is to maintain a flow of discussion and cover the desired range of topics through a semi-structured process and to ensure that all participants have the chance to speak up.

4.2.3 Case studies

Case studies can be a useful method of exploring a theme in depth through one person’s experience. They can be very powerful ways of telling a story, or painting a picture of the way in which the service impacts on a person’s life. Information can be collected through a series of extended interviews, or by a person writing down their story and then having it filled out in collaboration with a researcher. There are clearly issues of identification and privacy which need to be considered when using case studies. (This is true even when you are only using isolated quotes from a person, for instance in a larger report; it will still be important that the person can’t be identified without their permission.)

4.2.4 Document analysis

Document analysis uses documents, rather than interaction with individuals, to analyse the topic at hand. For a service evaluation, this could involve analysing minutes of meetings, annual or funding reports, financial records, clinical records, or external sources such as newspapers or government reports. This can be useful, for instance, if you are trying to identify early decisions which influenced the service’s implementation in the past and to trace their impact through the organisation over time.

4.2.5 Observational methods

Observation, including participant observation, may be most useful as a supplement to other qualitative methods, to gain a broader perspective on how a system or service operates. Observation can be undertaken by an external researcher who may, for instance, spend time in a clinic simply watching how clinicians and service users interact, or who may observe interactions in a waiting room. Participant observation means that someone engaged in the system (e.g. a staff member) is doing the observing, rather than an outsider. There are ethical issues associated with observation methods, which will need to be addressed. One of the key factors in the success of observation methods is letting people know that the observer is there (so they are fully informed and give consent), and then having the observer become as unobtrusive as possible so that people stop being self-conscious and relax into their natural patterns. This is a skill which can take time to develop.