Principles of Qualitative Methods: Section 4. Conducting Qualitative Research

Sampling

A key feature of all research is to design and select a sample for study. In social research (including public health research), a sample is a group of people who have been selected by the researcher for particular reasons relating to the research question. However, a sample could also be a selection of documents to be analysed, a method we will return to later on (see document research). In contrast to quantitative studies, research studies using a qualitative approach appear to have small sample sizes. However, even where a study involves a very small population the design and selection decisions made by the researcher have to follow a clear sampling strategy, allowing a third party to understand how the research participants (i.e. the sample) were selected and how they reflect the characteristics of the population under study.

In quantitative research, elements of the population are chosen at random and will have a known probability of selection; this strategy is known as probability sampling. Qualitative research follows a different logic and uses non-probability samples for selecting a population for study. Here the individual participants are selected deliberately for their specific characteristics that are of importance to the study. For example, a study investigating the experiences of dyslexic students in higher education will select students who are enrolled at a university, have completed at least one academic year and have been assessed as dyslexic. This strategy uses three sampling characteristics from a more general population of dyslexic students in higher education and is also known as purposive sampling. This strategy has two principle aims: (a) to ensure that all the key characteristics of relevance to the subject under study are covered, and (b) to ensure that within each key criterion some diversity is included so that the impact of the characteristic concerned can be explored. In the dyslexic student example, the diversity needs to be secured through the possibility of recruiting male and female students from different age groups and ethnic backgrounds. This might be achieved by sampling at a number of universities that vary in the kind of student populations they have. On the other hand, sampling methods might be used to minimize the diversity or even to find the extreme cases among those fulfilling the key criteria.

As discussed above, qualitative samples are usually small in size, following Ritchie, Lewis and Elam (2003), there are three main reasons for this:

' First if the data are properly analysed, there will come a point where very little new evidence is obtained from each additional fieldwork unit. This is because a phenomenon need only to appear once to be part of the analytical map. There is therefore a point of diminishing return where increasing the sample size no longer contributes new evidence. Second, statements about incidence or prevalence are not the concern of qualitative research. There is therefore, no requirement to ensure that the sample is of sufficient scale to provide estimates, or to determine statistically significant discriminatory variables. Third, the type of information that qualitative studies yield is rich in detail. There will therefore be many hundred of 'bites' of information from each unit of data collection. In order to do justice to these, sample sizes need to be kept to a reasonably small scale' (p. 83-84)

In more practical terms, it is simply unmanageable to conduct and analyse hundreds of interview transcripts, notes from observations and general field notes in the average time-scale of studies even when more than one researcher is involved. Although anticipating the possible disappointment of the reader we will not suggest a bench mark figure for a qualitative research size as every study will require an individual decision. However, there are plenty of references made to sample sizes in the literature on qualitative research methods.

Much more could be said about the intricacies and practicalities of the sampling process. The selection of references given at the end will help those who are interested in conducted a piece of qualitative research and those who have embarked on a project.

Access

Public health research often involves sampling participants through the National Health Service, Social Services, charitable organisations or other non-governmental organisations, who have a responsibility for protecting the privacy of those people under their auspice. Consequently, such organisations and their employees, with whom the researcher might need to negotiate access with, can act as gatekeepers. Given the multitude of empirical studies that are conducted in the UK, organisations can show resistance towards allowing researchers to access their patients/members/clients which they may see as being an already over-researched population. Even in commissioned studies where the funder has power to grant access, the way in which access is negotiated on the ground can be critical to the success of a study. Under these circumstances, it is pivotal to secure access to a research site and/or a population within that site before embarking on the study, i.e. when designing the study. This is of particular importance for researchers bidding for funding and those studies requiring ethical approval. See Lewis (2003) for ideas on how to engage effectively with organisations and their representatives.

Ethical considerations

Informed consent
Participants are required to give informed consent. This means that they should be informed about the purpose of the study, who is funding the study, who the researcher is, how the data will be used, what is required of the participants, how long their involvement will last and that their participation is voluntary (i.e. that they may withdraw from the study without giving a reason). All this information should fit onto one page and allow enough space for the participant and the researcher to sign and date the form. It is good practice to give a copy of the consent form to the participant.
Studies requiring ethical approval will be advised on the wording and content of the consent form; examples are usually available from the committees' websites.

Anonymity and confidentiality
Whilst it is clearly important to ensure anonymity and confidentiality for the participant, it is difficult to specify the degree to which this should take place. For instance should the identifying details of a person (name, place, age, occupation) be changed to a degree that people who are not his/her friend/family member do not recognise him/her or to a degree that family and friend could not identify the interviewee or to such an extent that the participant him-/herself could not recognise themselves? Such decisions depend not only on the nature of the research study, but also on whether the participants themselves are public figures. Confidentiality refers to the information that is provided by the participant but that should not be disclosed. Of course, it would not make sense to interview someone who would not wish that anything they said to be used in the research. Indeed, most participants know that the information they provide will be used. The researcher needs to decide carefully what information is pertinent to the research and how it will be presented. It is the researcher's responsibility to explain to the participant how the data will be used (e.g. quotations from transcripts) and how the identity of the participant will be protected.

Protecting participants from harm
It is important to assess how participation in the study might be harmful for some people in the sample. This is particularly so in studies that touch upon sensitive topics which might uncover painful experiences and lead people to disclose information which they have rarely or never previously shared. Less dramatic, though equally important; 'Interviews can have a certain seductive quality: participants may appear comfortable and may disclose information apparently willingly during an interview, but may later regret having been so open. They may also be left with feelings and thoughts stirred up by the interview long after the researcher has moved on' (Lewis, 2003: 68).

For these reasons, it is important to confer an understanding of the study and what it involves and to allow the participant to ask questions. The information provided to the participants however needs to be balanced by the consideration that they should not be too pre-informed about the issues of the study as this will impact on the results. It is also advisable to provide the participant with a contact number so that he/she can contact the researcher after the interview. Some ethics committees require contact details of the organisation employing the researcher so to allow participants to file a complaint.

Although researchers should be alert to the discomfort of an interviewee, and even offer to pause the interview, the researcher should not take on the role of an advisor or counsellor. However, it is appropriate to provide the interviewee with information of services and organisations that might be helpful. Cases where the information disclosed by the participants indicates that they might be in risk of being harmed pose serious ethical problems (see for example, Lewis:2003:69-70).Researchers are advised to discuss such possibilities and how to respond to them with their employers before embarking on the data collection.

Protecting researchers from harm
Although most ethical issues are concerned with the safety and benefit of the participants, the researcher, too, needs to be protected from harm. Most often data collection takes place in a variety of places that are neither fearsome nor frightening. However, the more private settings in which face-to-face interviews take place can be less safe, e.g. when they take place at a participant's home. When interviewing people who are, at least initially, strangers, it is important to strike a balance between sensible caution and paranoia (Arksey & Knight:1999). The level of safety procedures might vary depending on the research. A study that involves recruiting participants in a public area (e.g. a hospital ward) allows the researcher and the potential participant to get to know each other (at least briefly), exchange contact details and build a rapport, so that both sides have already conceded some information to the other side. Where recruitment involves Internet resources, researchers should be more cautious and double check information provided by potential participants. Ultimately, it is important to plan ahead and calculate time and other necessary resources into the research budget. Here is a list of helpful tips:

• Gather information about the area you will be researching in (e.g. public transport, demographic details; http://www.upmystreet.com).
• Leave an itinerary of your movements with a colleague or friend and contact them when you have returned.
• Assess the layout of the interview location and make a mental note of how to leave the building in an emergency.
• Strike a balance between developing a rapport with the interviewee and how this might be misinterpreted.
• Actively take decisions about the situation.

For more ideas and advice see 'stay safe' by the Social Research Association:
http://www.the-sra.org.uk/stay_safe.htm
where the Code of Practice for social researchers can be downloaded.

Ethical approval

It is important to establish at an early stage of the research whether ethical approval is required, from which committee, and what procedures and requirements are involved. Most ethics committees or application bodies have Internet websites that can help with some initial questions at the same time as appearing overwhelming. If ethical approval is required, researchers need to be aware that the process can take several months, so that it is important to allow for this time in the overall time-table. Many ethics committees are more used to dealing with quantitative research, randomised control trials or medical studies (e.g. involving human tissue). Researchers may therefore find themselves having to provide an explication of qualitative research methods, and dealing with questions or requirements that do not easily transfer to qualitative research. Patience, flexibility and tenacity are key factors in successful applications.

© ICrinson & MLeontowitsch 2006