Interviewing as a data collection method

AORN Journal, August, 2001 by Michelle Byrne

Interviewing is one of the most common methods for collecting data in qualitative research. Interviews allow participants to provide rich, contextual descriptions of events. The process of interviewing is time-consuming, and the quality of data often is dependent on the aptitude of the interviewer. This article will address preparing and implementing a research interview, as well as strategies for transcribing and analyzing the data collected.

PREPARING FOR INTERVIEWS

Before selecting interviewing as a data collection method, the researcher must determine whether the research question can be answered appropriately by interviewing people who have experienced the phenomenon of interest. A hypothetical study will be used to illustrate one process the researcher could use to facilitate interviewing. One focus for a research study could be to explore the lived experiences of fathers during cesarean section procedures. For this study, it has been proposed to interview fathers within 48 hours after delivery and then again between one and two months postdelivery.

You may wonder why the researcher would want to interview the same person more than once. There are two major reasons. First is the perspective of temporal knowledge. The term temporal is a philosophical term that defines how a person's current situation and knowledge have been influenced by his or her experiences and how the current situation will partially determine his or her future.(1) This explanation of temporality suggests that a person's perceptions of events may change with time.

The second reason for conducting more than one interview is to meet methodological rigor criteria for prolonged engagement. A second interview allows the participant to confirm or clarify information given in the first interview. A second interview also may ensure heightened trust between the participant and the researcher, thus enabling further disclosure of private thoughts and feelings.

IMPLEMENTING INTERVIEWS

The first step in all research is to conduct a pilot study. The researcher should practice how he or she will gain access to a sample, conduct interviews, and transcribe and analyze the data. Different approaches may exist depending on the methodology used.(2) A consistent approach should be used with both the pilot and actual study.

Accessing the sample. Initially, the researcher must decide whether to interview fathers whose significant others experienced emergent cesarean sections or fathers whose significant others experienced scheduled cesarean sections. Another sampling issue may be whether to include fathers of all ages. All of these decisions should be documented in a research journal. A literature review and access to participants may assist the researcher in deciding on sample criteria. In this hypothetical study, the sample comprises fathers between the ages of 20 and 45 who speak English and whose significant others underwent a planned cesarean section to deliver first-born children.

The next step is finding fathers to participate in the study. The researcher could talk to obstetricians for referrals. Contacting childbirth educators in the area also may elicit referrals. Another option may be to access the information from a surgical schedule. Keep in mind, however, the researcher will need appropriate permission from an institutional review board before contacting potential participants. After receiving names of potential participants for the study, the researcher will need to make preinterview contact.

Preinterview contact or introduction usually is made via telephone. This discussion should provide the potential participant information regarding the study, ensure protection of privacy, and explain what will and will not be done with the data. If the potential participant agrees to participate in the study, a mutually agreed time and place for the interview should be decided.

Conducting the interview. The interview should take place in a quiet environment with no distractions. Ideally, the researcher should sit across from the participant with a tape recorder between the two to capture both voices equally well. This positioning also enables the researcher to note nonverbal cues, such as eye contact, crying, laughing, or hand gestures.

The type of interview will be determined by the research question, methodology, and literature insight. Interviews can either be structured or unstructured.(3) In a structured interview, the researcher asks explicit questions consistently of all participants. In an unstructured interview, the researcher asks open-ended questions. The researcher's probes then are related directly to the participant's answers.

An interview generally consists of three stages.(4) The first stage includes introductions, provides an overview of the process, and builds trust. The middle of the interview usually is where the bulk of useful data is derived. Questions and probes should be directed to the phenomenon of interest. It is important at this stage for the researcher to keep the discussion focused on the topic. Participants who ramble or discuss irrelevant issues will increase the cost and time for transcription of audiotapes. If a participant spends 10 minutes discussing a new sports car, that discussion will not generate any new knowledge about the experience of cesarean section for fathers. The final portion of the interview should be a summary of the participant's responses and allow for confirmation or additional information to be given.