The Strengths and Limitations of Interviews as
A Research Technique for Studying Television Viewers

Alison Oatey

In order to conduct investigations today, researchers use a variety of techniques. These fall into two categories: qualitative and quantitative methodology. Qualitative methodology is the type we will focus most on in this essay. This uses smaller samples than in quantitative methods, seeing each individual as a unique being. This research is more interested in the depth of the data rather than breadth and requires the researcher to play an active role in the data collection. (Wimmer and Dominick 1997:84). It is often difficult to draw definitive conclusions from the findings or at least generalise them to larger groups because of the small scale of the method and the often quite unrepresentative samples that are often used. "Qualitative research is a useful mass media tool only when its limitations are recognised" (Wimmer and Dominick 1997:85). Qualitative research relates mainly to interviews done on a small scale. These are the focus for this essay, as much of the media research uses this (sometimes in combination with observational diaries or participant observation) to gain data. (Hammond, Breakwell and Fife-Schaw 1995:238). It is a technique used especially in the area of ethnographic research (Jensen and Jankowski 1991:33).

Firstly we should consider what an interview is. Frey and Oishi (1995:01) define it as "a purposeful conversation in which one person asks prepared questions (interviewer) and another answers them (respondent)" This is done to gain information on a particular topic or a particular area to be researched. Interviews are a useful tool which can lead to further research using other methodologies such as observation and experiments (Jensen and Jankowski 1991:101). Interviews can have one of two basic structures. They can be either structured (closed interview style) or unstructured (open interview style). Open-ended or unstructured interviews are defined by Nichols (1991:131) as "an informal interview , not structured by a standard list of questions. Fieldworkers are free to deal with the topics of interest in any order and to phrase their questions as they think best." This type of structure uses a broad range of questions asking them in any order according to how the interview develops (Breakwell, Hammond and Fife-Schaw 1995:231). Open-ended questions allow the interviewer, if they wish, to probe deeper into the initial responses of the respondent to gain a more detailed answer to the question (Wimmer and Dominick 1997:156). The richness of the data is therefore entirely dependant on the interviewer. They themselve, must judge how much or how little they should probe or say themselves.

There are of course both advantages and disadvantages to this type of structure. It is particularly useful as a pilot study, to test out what peoples responses would be to a particular issue. It may throw a completely different light on an issue that the interviewer had previously never considered (Wimmer and Dominick 1997:139). Freedom for the respondent to answer how they wish to is important in giving them a feeling of control in the interview situation. This version also has its disadvantages, namely in terms of the amount of time needed to collect and analyse the responses (Wimmer and Dominick 1997:139). Due to the varied nature of the responses, it is necessary to use the content analysis technique to analyse it. This is what takes the time. Open questions used in this unstructured interview approach can cause confusion either because of the lack of understanding of the question by the informant or by the lack of understanding of the respondent's answer by the interviewer (Wimmer and Dominick 1997:140). Despite some of these disadvantages, open-ended questions are very important. Gray (1987) showed this when she studied women’s relations to video technology. It was found that women wanted to tell their stories therefore needing open-ended questions to enable them to talk freely (Jensen and Jankowski 1991:155)

Closed or structured interviews are defined by Nichols (1991:131) as a social survey where "the range of possible answers to each question is known in advance. Often, possible answers are listed on the form so that the interviewer simply marks the appropriate reply in each case. This approach is much more standardised using a prearranged list of answers for the respondent to choose from. There is little freedom for flexibility, due to the fixed question order. Each person is given the same questions therefore being uniform (Wimmer and Dominick 1997:139 and 156). This has its advantages in that the information is easily quantifiable and allows the responses to be compared. Due to the lack of flexibility in this approach, it means that there is "little room for unanticipated discoveries" (Breakwell, Hammond and Fife-Schaw 1995:231). People may feel that their response does not fit any of the designated answers.

Using these two structures, there are 2 basic types of interview used in everyday research. The first of these is known as one to one interviews, personal interviews or intensive interviews. This type of interview uses a small sample averaging 30 people (Nichols 1991:13). The interview usually last up to several hours. It focuses on the use of open-ended questions allowing the respondent to answer freely. Questions that follow are then entirely based on how the respondent’s answer leads the interview. The questions are not therefore standardised (Wimmer and Dominick 1997:100). It is a suitable way to deal with sensitive or taboo topics yet as a whole requires a very good rapport to be established between interviewer and respondent (Nichols 1991:13 and Wimmer and Dominick 1997:100). This interview approach is flexible, providing a large amount of detail. It is clear that the answers are solely those of the person being questioned. The intensive situation that the interview is conducted in, may in itself allow information to be gained without directly asking for it (Wimmer and Dominick 1997:157). This approach also has its disadvantages essentially because it is time consuming and very costly. Interviewer bias is also quite a problem. The non-standardisation of the questions in this method means that it is difficult to generalise it on a larger scale (Wimmer and Dominick 1997:100 and 158).

The second type of interview technique is a group interview or focus group study. This was defined by Wimmer and Dominick (1997:97) as "a research strategy for understanding audience/ consumer attitudes and behaviour" The members of a focus group should feel very much at ease with each other before conducting the interview, ideally they should perhaps know each other already (see Buckingham’s study later). The members of the group should be of the same sex and share similar backgrounds in order to rule out any confounding variables (Nichols 1991:14). Conversation in a focus group can be either structured or unstructured (often somewhere in between) and can last up to two hours. Discussion is guided constantly by the interviewer whilst the respondents (usually 6-12 of them) discuss and express opinions with each other (Wimmer and Dominick 1997:97 and 455).

Focus groups, like one to one interviews, "allow for the collection of preliminary information about a topic. They may be used in pilot studies to detect ideas that will be further investigated using another research method" (Wimmer and Dominick 1997:97). Because of this, each interview approach can be used in studying television in combination with a further research methodology. Focus group interviewing is much cheaper and quicker to run, than intensive one to one interviews and responses tend to be more complete and less inhibited (Wimmer and Dominick 1997:97). Interviewing in a group can also provide disadvantages. One person may consistently undermine the others, dominating the conversation.

Some researchers claim that the focus groups are not a good research methodology because of the potential influence of one or two respondents on the remaining members of the group. These critics say that a dominant respondent can negatively affect the outcome of the group and that group pressures may influence the comments made by individuals (Wimmer and Dominick 1997:461).

Although this could occur, it may also be that other respondents ideas spark new ideas with others, creating a snowball effect. The use of what is known as an extended focus group method, whereby each respondent fills out a questionnaire prior to the focus group discussion expressing their own personal views. This limits the problem of them being unwilling to express their opinion infront of the rest of the group (Wimmer and Dominick 1997:97). Focus group interviews frequently use unrepresentative samples as participants must often volunteer themselves to do it. It is only a certain kind of person that will do this. The quality of the data acquired from a focus group interview may also not be as good as with one to one interviews (Wimmer and Dominick 1997:107).

In general there are obviously advantages and disadvantages for using any interview method. It allows questioning to be guided as you want it and you can clarify points that need to be made clearer much more easily than in something like a mailed questionnaire (Frey and Oishi 1995:03). The technique does however rely on the respondent being willing to give accurate and complete answers (Breakwell, Hammond and Fife-Schaw 1995:238). They may often lie due to feelings of embarrassment, inadequacy, lack of knowledge on the topic, nervousness, memory loss or confusion. On the contrary, they may also provide very elaborate answers in an attempt to figure out the purpose of the study (Wimmer and Dominick 1997:162). Validity and reliability of the interview data may be influenced by these (Breakwell, Hammond and Fife-Schaw 1995:238-239). "Interviewing is a complex and demanding technique" (Frey and Oishi 1995:02).

These interview approaches can be better understood by the use of examples to illustrate the style. David Morley’s study of the former television programme Nationwide used the focus group interview method to investigate "the extent to which individual interpretation of programmes could be shown to vary systematically in relation to ...sociocultural background (Morley 1981b:56). He chose to work with groups rather than individuals because he believed that individuals were not seen in their social context (Morley 1980:33). The groups were very traditional focus groups as they were already meeting as part of an educational course. They therefore knew each other and were easy to access (Morley 1986:40). The groups each watched two programmes which were then directly followed by a discussion lasting about 30 minutes. Morley chose to use an unstructured approach to the interview with open-ended questions allowing for diversity in response. His study therefore lacked standardisation. He stated however that the reasoning behind this choice was that "it is not simply the substance of the answer which is important, it is also the form of its expression which constitutes its meaning" (Morley 1980:31). One problem with Morley’s research is that the viewing and interviewing did not take place in the usual domestic setting (Morley 1992:133 and Morley 1986:40). Morley did however accept himself that his research was subject to the usual limitations of the interview technique (Morley 1981b:67).

David Buckingham’s study of young viewers watching Eastenders in 1987 also used the focus group method . His focus groups were smaller in size than many other researchers have used (only 5 people). The majority of the groups were friendship groups, already established outside of the study and therefore they all knew each other well (Buckingham 1987:158-160). Very few, in fact only 3 out of the 12 groups, had members entirely of the same sex. This is in contrast to Nichols ‘ suggestion (1991:14). Open-ended discussion was once again used just as in David Morley’s study. David Buckingham notes that his intention was "to make these discussions as open-ended as possible and to avoid directing them towards particular issues" (Buckingham 1987:158) He began with background questions about viewing habits and then asked each member which was their favourite and least favourite character and why. This generated into a more open, general discussion where the group could define their own agenda (Buckingham 1987:158).

Elihu Katz and Tamar Liebes used very similar methodologies to those of Buckingham and Morley. Focus groups of 6 people (3 married couples) were used with 1 of the couples being initially selected and the other 2 couples being friends of them. Like Morley and Buckingham’s study, they all knew each other. [WWW document] (URL http://www.aber.ac.uk/media/Modules/TF33120/katzlieb.html). The situation in which the viewing and discussion took place was the hosts living room; a much more realistic and natural setting than Morley’s was. Discussion afterwards about the programme Dallas that they were viewing, generally lasted an hour. It was guided by the researcher using both open and closed questions. The interview style generally became more structured as the interview progressed with more closed questions being introduced, requiring a more focused answer. "The participants were from similar age groups and educational backgrounds - all were lower middle class with high school education or less" [WWW document] (URL http://www.aber.ac.uk/media/Modules/TF33120/katzlieb.html). Katz and Liebes therefore followed in Nichols’ guidelines (1991:14) ensuring that they were all of similar age and background therefore reducing the effects of confounding variables.

A second study by David Morley looked at the use of television within the family setting. Interviews took place within the respondent’s own home with initially just the two parents being interviewed together (the children were allowed to join in later). The interviews lasted one to two hours (as suggested by Wimmer and Dominick 1997:455). They were tape recorded and transcribed in full. The method used was an unstructured discussion using open-ended questions with the chance of probing further for greater depth in the respondent’s answer (Morley 1986:51-52). Problems with David Morley’s research into the use of television includes the use of small samples from one area of the country. This means that the results cannot be generalised very widely. Children were also very rarely included in the interview itself due to the many difficulties involved in doing this. This means that the results may in fact not be a very accurate representation of the entire family’s use of television (Morley 1986:52 and 174).

A final observation of a study by James Lull (1990) indicates a different methodological approach. Each individual within a family was interviewed separately because Lull believed that "the research was conducted from a perspective that regards the individual as participating in strategic patterns of interaction which involve the employment of media as personal resources for accomplishing social tasks" (Lull 1990:51). He used a structured interview style with a formal interview schedule. This standardised the interview ensuring that each individual interviewed received the same questions in the same order (Lull 1990:52). Due to the trained observers living with the families for the two days prior to the interview, it is likely that the families’ responses to the interview itself would be better, both in terms of content and style. The families had had a chance to get to know the researcher prior to the interview enabling them to develop a sense of trust. "An unusual degree of trust is likely to lead to willingness on the part of the subjects to answer the questions carefully and with validity. This is especially advantageous when the questions are of a sensitive nature" (Lull 1990:53).

Having looked at the available research evidence concerning the use of interviews, it is clear that there are varying types of interview as well as various styles an interview can take. It has also been made clear, that many research studies using the interview method also use another methodology as well to allow for more accurate results and greater understanding. Interviewing is a difficult method to employ properly, relying on the interviewer themselves to enable an objective interview to be undertaken. It is a technique employed extensively in television viewing and which appears very effective for this field. It can provide valuable data either for personal reference only or as a means of gathering information to pursue further research using a different method. "Interviewing provides an opportunity for combining practical, analytical and interpretative approaches to media" (Jensen and Jankowski 1991:223).

References

Monday 19th April 1999