This chapter discusses four alternatives to laboratory and survey research: field observations, focus groups, intensive interviews, and case studies. Field observation involves the study of a phenomenon in natural settings. The researcher may be a detached observer or a participant in the process under study. The main advantage of this technique is its flexibility; it can be used to develop hypotheses, to gather preliminary data, or to study groups that would otherwise be inaccessible. Its biggest disadvantage is the difficulty in achieving external validity.

The focus group, or group interviewing, is used to gather preliminary information for a research study or to gather qualitative data concerning a research question. The advantages of the focus group method are the ease of data collection and the depth of information that can be gathered. Among the disadvantages: the quality of information gathered during focus groups depends heavily on the group moderators' skill; focus groups can only complement other research because they provide qualitative not quantitative data.

Intensive interviewing is used to gather extremely detailed information from a small sample of respondents. The wealth of data that can be gathered with this method is its primary advantage. Because intensive interviewing is usually done with small, nonrandom samples, however, generalizability is sometimes a disadvantage. Interviewer bias can also be a disadvantage.

The case study method draws from as many data sources as possible to investigate an event. Case studies are particularly helpful when a researcher desires to explain or understand some phenomenon. Some problems with case studies are that they can lack scientific rigor, they can be time-consuming to conduct, and the data they provide can be difficult to generalize from and to summarize.

The quantitative approaches discussed in the preceding chapter are not suitable for all research problems. There may be certain situations in which a different technique is appropriate. This chapter outlines the major differences between the two methods and examines the most frequently used techniques of qualitative research.

5.1 Aims and Philosophy

Qualitative research differs from quantitative research along three main dimensions. First, the two methods have a different philosophy of reality. For a quantitative researcher, reality is objective; it exists apart from the researcher and is capable of being seen by all. In other words, it's out there. For the qualitative researcher, there is no one single reality. Each observer creates reality as part of the research process; it is subjective and exists only in reference to the observer. Further, the quantitative researcher believes that reality can be divided into component parts, and he or she gains knowledge of the whole by looking at these parts. On the other hand, the qualitative researcher examines the entire process believing that reality is holistic and cannot be subdivided.

Second, the two methods have different views of the individual. The quantitative researcher believes all human beings are basically similar and looks for general categories to summarize their behaviors or feelings. The qualitative investigator believes that human beings are all fundamentally different and cannot be pigeonholed.

Third, quantitative researchers aim to generate general laws of behavior and explain many things across many settings. In contrast, qualitative scholars attempt to produce a unique explanation about a given situation or individual. Whereas quantitative researchers strive for breadth, qualitative researchers strive for depth.

The practical differences between these approaches are perhaps most apparent in the research process. The following five major research areas describe significant differences between quantitative and qualitative research.

1. Role of the researcher. The quantitative researcher strives for objectivity and is separated from the data. The qualitative researcher is an integral part of the data; in fact, without the active participation of the researcher, no data exist.

2. Design. In quantitative methods, the design of the study is determined before it begins. In qualitative research, the design evolves during the research; it can be adjusted or changed as it progresses.

3. Setting. Quantitative researchers try to control contaminating and/or confounding variables by conducting their investigations in laboratory settings. Qualitative researchers conduct their studies in the field, in natural surroundings. They try to capture the normal flow of events, without trying to control the extraneous variables.

4. Measurement instruments. In quantitative research, these exist apart from the researcher. In fact, another party could use the instruments to collect data in the researcher's absence. In qualitative research, the investigator is the instrument; no other individual could fill in for the qualitative researcher.

5. Theory building. In the quantitative area, research is used to test theory and to ultimately support or reject it. In the qualitative area, theory is "data driven" and emerges as part of the research process, evolving from the data as they are collected.

These differences will become more apparent throughout this chapter. Four common qualitative techniques are discussed: field observations, focus groups, intensive interviews, and case studies.

5.2 Field Observations

Before 1980, field observation was rarely used in scientific research. It was reported that only 2%-3% of the articles published in journalism and broadcasting journals had employed the technique. Recently, however, field observations have become more common in the research literature.

Field observation is useful for collecting data as well as for generating hypotheses and theories. Like all qualitative techniques, it is more concerned with description and explanation than it is with measurement and quantification.

Field observations are classified along two major dimensions:

(1)       The degree to which the researcher participates in the behavior under observation; and

(2)       The degree to which the observation is concealed.

Overt observation is represented by Quadrant 1. In this situation, the researcher is identified as such when the study begins. Those under observation are aware that they are being studied. Further, the researcher's role is only to observe, refraining from participation in the process under observation. Quadrant 2 represents overt participation. In this arrangement, the researcher is also known to those being observed, but unlike Quadrant 1, the researcher goes beyond the observer role and becomes a participant in the situation. Quadrant 3 represents the situation where the researcher's role is limited to that of observer, but those under observation are not aware they are being studied. A study in which the investigator participates in the process under investigation, but is not identified as a researcher, is represented by Quadrant 4, see Figure 5.1.

To illustrate the distinction between the various approaches, assume a researcher wants to observe and analyze the dynamics of writing comedy for television. The researcher could choose the covert observer technique and perhaps pretend to be doing something else (such as fixing a typewriter) while actually observing the TV writing team at work. Alternatively, the researcher could be introduced as someone doing a study of comedy writing and allowed to watch the team in action. If the research question is best answered by active participation, the investigator might be introduced as a researcher but would still participate in the writing process. If the covert participant strategy is used, the researcher might be introduced as a new writer just joining the group (such an arrangement might be made with the head writer who would be the only person to know the true identity of the researcher).

Figure 5.1: Dimensions of field observation

Overt

Covert

The choice of technique depends upon the research problem and the degree of cooperation available from the group or individual being observed, as well as ethical considerations. Covert participation may affect subjects' behavior and also raises the ethical question of deception. On the other hand, the information gathered may be more valid if subjects are unaware of being scrutinized.

5.2.1 Advantages of Field Observations

Field observation is not an appropriate technique for every research question, owing to the lack of control and quantification, but it does possess several unique advantages. For one thing, many mass media problems and questions cannot be studied using any other methodology. Field observation often helps the researcher to define basic background information necessary to frame a hypothesis and to isolate independent and dependent variables. For example, a researcher interested in how creative decisions in advertising are made could observe several decision-making sessions to see what actually transpires. Field observations often make excellent pilot studies in that they identify important variables and provide useful preliminary information. In addition, since the data are gathered firsthand, observation is not dependent on the subjects' ability or willingness to report their behavior. For example, young children may lack the reading or verbal skills necessary to respond to a questionnaire concerning their play behavior, but such data are easily gathered by the observational technique.

A field observation is not always used as a preliminary step to other approaches, however. In many cases it alone is the only appropriate approach, especially when quantification is difficult. Field observation is particularly suitable for a study of the gate keeping process in a network television news department, because quantification of gate keeping is rather tenuous. Field observation may also provide access to groups that would otherwise be difficult to observe or examine. For example, a questionnaire sent to a group of producers of X-rated movies is not likely to have a high return rate. An observer, however, may be able to establish enough mutual trust with such a group to persuade them to respond to rigorous questioning.

Field observation is usually inexpensive. In most cases, writing materials or a small tape recorder will suffice. Expenses increase if the problem under study requires a large number of observers, extensive travel, or special equipment (such as video recording machines).

Perhaps the most noteworthy advantage of field observation is that the study takes place in the natural setting of the activity being observed and can, thus, provide data rich in detail and subtlety. Many mass media situations, such as a family watching television, are complex and are constantly subjected to intervening influences. Field observation, because of the opportunity for careful examination, allows observers to identify these otherwise unknown variables.

5.2.2 Disadvantages of Field Observations

On the negative side, field observation is a bad choice if the researcher is concerned with external validity. This difficulty is partly due to the potentially questionable representativeness of the observations made and partly to problems in sampling. Observing the television viewing behavior of a group of children at a day-care center can provide valuable insights into the social setting of television viewing, but it probably has little correlation to what preschoolers do in other places and under different circumstances.

Moreover, since field observation relies heavily on a researcher's perceptions and judgments as well as on preconceived notions about the material under study, experimenter bias may unavoidably favor specific preconceptions of results, while observations to the contrary are ignored or distorted. This, primarily, is why one observer is rarely used in a field observation study. Observations need to be cross-validated by second or third observers.

Finally, like field experiments, field observations suffer from the problem of reactivity. The very process of being observed may influence the behavior under study. Of course, reactivity can be a problem with other research methods, but it is most often mentioned as a criticism of field observation. Scholars provide some perspective on observer effects using data taken from an observational study of families' TV viewing behavior They found that the presence of an observer in the house did have some impact on family members. About 20% of parents and 25% of children reported that their overall behavior was affected by the presence of an observer. The majority of those who were affected thought that they became nicer or more polite and formal because of the observer's presence. When it came to differences in the key behavior under study, 87% said that the observer's presence had no effect on their TV viewing activity. Additionally, among those who reported an observer effect, there were no systematic differences in the distribution of changes. About the same number said that they watched more because the observer, as they said, watched less. Obviously, additional studies of different groups in different settings are needed before this problem is fully understood, but Lull's data suggest that although reactivity is a problem with observational techniques, its impact may not be as drastic as some suggest.

In any case, at least two strategies are available to diminish the impact of selective perception and reactance. One is to use several observers to cross-validate the results. A second strategy has to do with the notion of triangulation - the supplementing of observational data with data gathered by other means (questionnaires, existing records, and so on). Accuracy is sought by using multiple data collection methods.

5.2.3 Field Observation Techniques

There are at least six stages in a typical field observation study: choosing the research site, gaining access, sampling, collecting data, analyzing data, and exiting.

5.3 Choosing the Research Site

The choice of a research site depends upon the general nature of the research question. The area of inquiry usually suggests a behavior or a phenomenon of interest. Once that is identified, the next step is to select a setting in which the behavior or phenomenon occurs with sufficient frequency to make observation worthwhile. The setting should also accommodate the recording forms and instruments the observer plans to use. For example, if videotaping certain scenes is planned, there must be enough light available for the camera to operate.

It is recommended that the researcher select two or three possible research sites and then "hang around"  each of them to discover their main advantages and disadvantages. He goes on to caution researchers that the site must be permanent and stable enough to permit observations over a period of time.

5.3.1 Gaining Access

Once the site is selected, the next step is to establish contact.  It is noted that the degree of difficulty faced by researchers in gaining access to settings is a function of two factors: (1) how public the setting is, and (2) the willingness of the subjects in the setting to be observed. The easiest setting to gain access to is one that is open to the public and where people have little reason to keep their behavior secret (for example, TV watching in public places such as airports, bars, dormitory viewing rooms). The most difficult setting to gain access to is one where entry is restricted and where participants have good reason to keep their activities secret (for example, the behavior of hostage takers).

Observation of a formal group (such as a film production crew) often requires permission from management and perhaps union officials. School systems and other bureaucracies usually have a special unit to handle requests from researchers and to assist them in obtaining necessary permissions.

Gaining permission to conduct field observation research requires persistence and public relations skills. Researchers must determine how much to disclose about the nature of the research. In most cases, it is not necessary to provide a complete explanation of the hypothesis and procedures, unless there may be objections to sensitive areas. Researchers interested in observing which family member actually controls the television set might explain that they are studying patterns of family communication. Once the contact has been made, it is necessary to establish a rapport with the subjects). Bogdan and Taylor (1984) suggested the following techniques for building rapport: establish common interests with the participants; start relationships slowly; if appropriate, participate in common events and activities; and do not disrupt participants' normal routines.

5.3.2 Sampling

Sampling in field observation is more ambiguous than in most other research approaches. In the first place, there is the problem of how many individuals or groups to observe. If the focus of the study is communication in the newsroom, how many newsrooms should be observed? If the topic is family viewing of television, how many families should be included? Unfortunately, there are no guidelines to help answer these questions. The research problem and the goals of the study are often used as indicators for sample size: if the results are intended for generalization to a population, one subject or group is probably inadequate.

Another problem is deciding what behavior episodes or segments to sample. The observer cannot be everywhere and see everything, so what is observed becomes a de facto sample of what is not observed. If an observer views one staff meeting in the newsroom, this meeting represents other, unobserved meetings; one conversation at the coffee machine is a sample of all such conversations. In many cases researchers cannot adhere closely to the principles of probability sampling, but they should keep in mind the general notion of representativeness.

Most field observations use purposive sampling: observers draw on their knowledge of the subject(s) under study and sample only from the behaviors or events that are relevant. In many cases, previous experience and study of the activity in question will suggest what needs to be examined. In a study of newsroom decision making, for example, researchers would want to observe staff meetings, since they are obviously an important part of the process. However, restricting the sampling to observations of staff meetings would be a mistake; many decisions are made at the water fountain, over lunch, and in the hallways. Experienced observers tend not to isolate a specific situation but rather to consider even the most insignificant situation for potential analysis. For most field observation, researchers need to spend some time simply getting the feel of the situation and absorbing the pertinent aspects of the environment before beginning a detailed analysis.

5.3.3 Collecting Data

The traditional tools of data collection—the notebook and pen—have given way to radically new equipment in many cases, due to recent advances in electronics. For example, television cameras may be installed in a small sample of households to document the families' television-viewing behavior. Two cameras, automatically activate when the television set is turned on, videotaped the scene in front of the set. However, while a camera is able to record more information than an observer with a notebook, the problems in finding consenting families, maintaining the equipment, and interpreting tapes shot at low light levels made the project difficult.

Similarly, it was noted that although the advantages offered by audio and video recording are tempting, there are five major drawbacks to their use:

¨      Recording devices take time away from the research process because they need regular calibration and adjustment to work properly.

¨      The frame of the recording is different from the frame of the observer; a human observer's field of view is about 180°, whereas a camera's is about 60°.

¨      Recordings have to be catalogued, indexed, and transcribed, adding extra work to the project.

¨      Recordings take behavior out of context.

¨      Recordings tend to atomize (fragment) behavior and distract attention from the whole process.

Consequently, researchers must weigh the pros and cons carefully before deciding to incorporate recording equipment into the observational design.

Note taking in the covert participant situation requires special attention. Continually scribbling away on a notepad is certain to draw attention and suspicion to the note taker and might expose the researcher's real purpose in a particular setting. In a situation of this type, it is advisable to make mental notes and transcribe them at the first opportunity. If the researcher is initially identified as such, the problem of note taking is somewhat alleviated. Nonetheless, it is not recommended that the observer spend all of his or her time furiously taking notes. Subjects are already aware of being observed, and conspicuous note taking could make them more uneasy. Brief notes jotted down during natural breaks in a situation attract a minimum of attention and can be expanded at a later time.

The field notes constitute the basic corpus of data in any field study. In them, the observers record not only what happened and what was said, but also personal impressions, feelings, and interpretations of what was observed. A general procedure is to separate personal opinions from the descriptive narrative by enclosing the former in brackets.

How much should be recorded? It is always better to record too much information than too little. An apparently irrelevant observation made during the first viewing session might become significant during the course of the project. If the material is sensitive, or if the researcher does not wish to make it known that research is taking place, the notes may be written in abbreviated form or in code.

5.3.4 Analyzing data

In field observation, data analysis consists primarily of filing and content analysis. Constructing a filing system is an important step in observation. The purpose of the filing system is to arrange raw field data in an orderly format to enable systematic retrieval later (the precise filing categories are determined by the data). Using the hypothetical study of decision making in the newsroom, filing categories might include the headings "Relationships," "Interaction—Horizontal," "Interaction—Vertical," and "Disputes." An observation may be placed in more than one category. It is a good idea to make multiple copies of all notes, and periodic filing of notes throughout the observation period will save time and confusion later.

A rough content analysis is performed to search for consistent patterns once all the notes have been ascribed to their proper files. Perhaps most decisions in the newsroom are made in informal settings such as hallways rather than in formal settings such as conference rooms. Perhaps most decisions are made with little superior-subordinate consultation. At the same time, deviations from the norm should be investigated. Perhaps all reporters except one are typically asked their opinions on the newsworthiness of events. Why the exception?

The overall goal of data analysis in field observation is to arrive at a general understanding of the phenomenon under study. In this regard, the observer has the advantage of flexibility. In laboratory and other research approaches, investigators must at some point commit themselves to a particular design or questionnaire. If it subsequently turns out that a crucial variable was left out, there is little that can be done. In field observation, the researcher can analyze data during the course of the study and change the research design accordingly.

5.3.5 Exiting

A participant must also have a plan for leaving the setting or the group under study. Of course, if the participant is known to everyone, exiting will not be a problem. Exiting from a setting that participants regularly enter and leave is also not a problem. Exiting can be difficult, however, when participation is covert. In some instances, the group may have become dependent on the researcher in some way and the departure may have a negative effect on the group as a whole. In other cases, the sudden revelation that a group has been infiltrated or taken in by an outsider might be unpleasant or distressing to some. The researcher has an ethical obligation to do everything possible to prevent psychological, emotional, or physical injury to those being studied. Consequently, leaving the scene must be handled with diplomacy and tact.

5.4 Focus Groups

The focus group, or group interviewing, is a research strategy for understanding audience/ consumer attitudes and behavior. From 6 to 12 people are interviewed simultaneously, with a moderator leading the respondents in a relatively free discussion about the focal topic. The identifying characteristic of the focus group is controlled group discussion, which is employed to gather preliminary information for a research project, to help develop questionnaire items for survey research, or to understand the reasons behind a particular phenomenon.

5.4.1 Advantages of Focus Groups

 One advantage of focus groups is that they allow for the collection of preliminary information about a topic or phenomenon. Focus groups may be used in pilot studies to detect ideas that will be investigated further using another research method, such as a telephone survey, or another qualitative method.

A second important advantage is that focus groups can be conducted very quickly. The major portion of time is spent recruiting the respondents. A good research company that specializes in recruiting for focus groups can usually recruit respondents in about 7—10 days, depending on the type of person required.

The cost of focus groups also makes the approach an attractive research method; most focus groups can be conducted for about $1,000-$3,000 per group, depending on the type of respondent required for the group, the part of the country in which the group is conducted, and the moderator or company used to conduct the group. When respondents are difficult to recruit, or the topic requires a specially trained moderator, a focus group may cost several thousand dollars. The price, however, is not excessive if the groups provide valuable data for future research studies.

Researchers also like focus groups because of the flexibility in question design and follow-up. In conventional surveys, interviewers work from a rigid series of questions and are instructed to follow explicit directions in asking the questions. A moderator in a focus group, on the other hand, works from a list of broad questions as well as more refined probe questions; hence, follow-up on important points raised by participants in the group is easy. The ability to clear up confusing responses from respondents makes focus groups valuable in the research process.

Most professional focus group moderators or research companies use a procedure known as an extended focus group, in which respondents are required to complete a written questionnaire before the start of the group. The pregroup questionnaire, which basically covers the material that will be discussed during the group session, serves to "force" the respondents to commit to a particular answer or position before entering the group session. This commitment eliminates one potential problem created by group dynamics, namely, the person who does not wish to offer an opinion because he or she is in minority.

Finally, focus group responses are often more complete and less inhibited than those from individual interviews. One respondent's remarks tend to stimulate others to pursue lines of thinking that might not have been brought out in an individual situation. With a competent moderator, the discussion can have a beneficial snowball effect, as one respondent comments on the views of another. A skilled moderator can also detect the opinions and attitudes of those who are less articulate by noting facial expressions and other nonverbal behavior while others are speaking.

 5.4.2 Disadvantages of Focus Groups

Focus group research is not totally free from complications; the approach is far from perfect. Some of the problems are discussed here.

Some groups become dominated by a self-appointed group leader who monopolizes the conversation and attempts to impose her or his opinion on the other participants. Such a person usually draws the resentment of the other participants and may have an extremely adverse effect on the performance of the group. The moderator needs to control such situations tactfully before they get out of hand.

Gathering quantitative data is inappropriate for a focus group. If quantification is important, it is wise to supplement the focus group with other research tools that permit more specific questions to be addressed to a more representative sample. Many people unfamiliar with focus group research incorrectly assume that the method will answer questions of "how many" or "how much." Focus group research is intended to gather qualitative data to answer questions such as "why" or "how." Many times people who hire a person or company to conduct a focus group are disgruntled with the results because they expected exact numbers and percentages. Focus groups do not provide such information.

As suggested earlier, focus groups depend heavily on the skills of the moderator, who must know when to probe for further information, when to stop respondents from discussing irrelevant topics, and how to get all respondents involved in the discussion. All these things must be accomplished with professionalism and care, since one sarcastic or inappropriate comment to a respondent may have a chilling effect on the group's performance.

There are other drawbacks, as well. The small focus group samples are composed of volunteers and do not necessarily represent the population from which they were drawn, the recording equipment or other physical characteristics of the location may inhibit respondents, and if the respondents are allowed to stray too far from the topic under consideration, the data produced may not be useful.

5.4.3 Methodology of Focus Groups

There are seven basic steps in focus group research.

1. Define the problem. This step is similar in all types of scientific research: a well-defined problem is established, either on the basis of some previous investigation or out of curiosity. For example, many television production companies that produce pilot programs for potential series will conduct 10-50 focus groups with target viewers to determine their reactions to each concept.

2. Select a sample. Because focus groups are small, researchers must define a narrow audience for the study. The type of sample depends on the purpose of the focus group: the sample might consist of consumers who use a particular type of laundry detergent, men aged 18—34 who listen to a certain type of music, or teenagers who purchase more than 10 record albums a year.

3. Determine the number of groups necessary. To help eliminate part of the problem of selecting a representative group, most researchers conduct two or more focus groups on the same topic. Results can then be compared to determine whether any similarities or differences exist; or, one group may be used as a basis for comparison to the other group. A focus group study using only one group is rare, since there is no way to know if the results are group-specific or characteristic of a wider audience.

4. Prepare the study mechanics. A more detailed description of the mechanical aspects of focus groups is in; suffice it to say here that this step includes arranging for the recruitment of respondents (by telephone or possibly by shopping center intercept), reserving the facilities at which the groups will be conducted, and deciding what type of recording (audio and/or video) will be used. The moderator must be selected and briefed about the purpose of the group. In addition, the researcher needs to determine the amount of co-op money each respondent will receive for participating. Respondents usually receive between $10 and $50 for attending, although professionals such as doctors and lawyers may require up to $100 or more for co-op.

5. Prepare the focus group materials. Each aspect of a focus group must be planned in detail; nothing should be left to chance — in particular, the moderator must not be allowed to wing it. The screener questionnaire is developed to produce the correct respondents; recordings and other materials the subjects will hear or see are prepared; any questionnaires the subjects will complete are produced (including the presession questionnaire); and a list of questions is developed for the presession questionnaire and the moderator's guide.

Generally, a focus group session begins with some type of shared experience, so that the individuals have a common base from which to start the discussion. The members may listen to or view a tape or examine a new product, or they may simply be asked how they answered question 1 on the presession questionnaire.

The existence of a moderator's guide does not mean that the moderator cannot ask questions not contained in the guide. Quite the opposite is true. The significant quality of a focus group is that it allows the moderator to probe comments that respondents make during the session. A professional moderator is often able to develop a line of questioning that no one thought about before the group began, and many times the questioning provides extremely important information. Professional moderators who have this skill receive very substantial fees for conducting focus groups.

6. Conduct the session. Focus groups may be conducted in a variety of settings, from professional conference rooms equipped with two-way mirrors to hotel rooms rented for the occasion. In most situations, a professional conference room is used. Hotel and motel rooms are used when a focus facility is not located close by.

7. Analyze the data and prepare a summary report. The written summary of focus group interviews depends on the needs of the study and the amount of time and money available. At one extreme, the moderator/researcher may simply write a brief synopsis of what was said and offer an interpretation of the subjects' responses. For a more elaborate content analysis, or a more complete description of what happened, the sessions can be transcribed so that the moderator/ researcher can scan the comments and develop a category system, coding each comment into the appropriate category. For example, a researcher who notices that most respondents focus on the price of a new product can establish a content category labeled "Price," code all statements in the transcript referring to price, and arrange these statements under the general heading. The same technique is followed for other content categories. When the coding is completed, the researcher makes summastatements about the number, tone, and consistency of the comments that fall into each category. Needless to say, this approach requires some expenditure of time and money on the researcher's (or client's) part.

5.5 Intensive Interviews

Intensive interviews, or in-depth interviews, are essentially a hybrid of the one-on-one personal interview approach discussed in Chapter 6. Intensive interviews are unique in that they:

¨      Generally use smaller samples.

¨      Provide very detailed information about the reasons why respondents give specific answers. Elaborate data concerning respondents' opinions, values, motivations, recollections, experiences, and feelings are obtained.

¨      Allow for lengthy observation of respondents' nonverbal responses.

¨      Are usually very long. Unlike personal interviews used in survey research that may last only a few minutes, an intensive interview may last several hours, and may take more than one session.

¨      Are customized to individual respondents. In a personal interview, all respondents are asked the same questions. Intensive interviews allow interviewers to form questions based on each respondent's answers.

¨      Can be influenced by the interview climate. To a greater extent than with personal interviews, the success of intensive interviews depends on the rapport established between the interviewer and respondent.

5.5.1 Advantages and Disadvantages of Intensive Interviews

As is probably obvious, the biggest advantage of the in-depth interview is the wealth of detail that it provides. Further, when compared to more traditional survey methods, intensive interviewing provides more accurate responses on sensitive issues. The rapport between respondent and interviewer makes it easier to approach certain topics that might be taboo in other approaches. In addition, there may be certain groups for which intensive interviewing is the only practical technique. For example, a study of the media habits of U.S. senators would be hard to do as an observational study. Also, it would be difficult to get a sample of senators to take the time to respond to a survey questionnaire. But in some cases, such persons might be willing to talk to an interviewer.

On the negative side, generalizability is sometimes a problem. Intensive interviewing is typically done with a small, nonrandom sample. Further, since interviews are usually non-standardized, each respondent may answer a slightly different version of a question. In fact, it is very likely that a particular respondent may answer questions not asked of any other respondent. Another disadvantage of in-depth interviews is that they are especially sensitive to interviewer bias. In a long interview, it's possible for a respondent to learn a good deal of information about the interviewer. Despite practice and training, some interviewers may inadvertently communicate their attitudes through loaded questions, nonverbal cues, or tone of voice. The effect this may have on the validity of the respondent's answers is hard to gauge. Finally, intensive interviewing presents problems in data analysis. A researcher given the same body of data taken from an interview may wind up with interpretations significantly different from the original investigator.

5.5.2 Procedures

The procedures for conducting intensive interviews are similar to those used in personal interviews in reference to problem definition, respondent recruiting, and data collection and analysis. The primary differences with intensive interviews are:

¨      Co-op payments are usually higher, generally from $50-$1,000.

¨      The amount of data is tremendous. Analysis may take several weeks to several months.

¨      Interviewers get extremely tired and bored. Interviews must be scheduled several hours apart, which makes data collection take much longer.

¨      It is very difficult to arrange intensive interviews because of the time required. This is especially true with respondents who are professionals.

¨      Small samples do not allow for generalization to the target population.

5.6 Case Studies

The case study method is another common qualitative research technique. Simply put, a case study uses as many data sources as possible to investigate systematically an individual, group, organization, or event. Case studies are performed when a researcher desires to understand or explain a phenomenon. Case studies are frequently used in medicine, anthropology, clinical psychology, management science, and history. Sigmund Freud wrote case studies of his patients; economists wrote case studies of the cable TV industry for the FCC; the list is endless.

On a more formal level, a case study was defined as an empirical inquiry that uses multiple sources of evidence to investigate a contemporary phenomenon within its real-life context in which the boundaries between the phenomenon and its context are not clearly evident. This definition highlights how a case study differs from other research strategies. For example, an experiment separates phenomenon from real-life context. The context is controlled by the laboratory environment. The survey technique tries to define the phenomenon under study narrowly enough to limit the number of variables to be examined. Case study research includes both single and multiple cases. Comparative case study research, frequently used in political science, is an example of the multiple case study technique.

Four essential characteristics of case study research:

1. Particularistic. This means that the case study focuses on a particular situation, event, program, or phenomenon, making it a good method for studying practical real-life problems.

2. Descriptive. The final result of a case study is a detailed description of the topic under study

3. Heuristic. A case study helps people to understand what's being studied. New interpretations, new perspectives, new meaning, and fresh insights are all goals of a case study.

4. Inductive. Most case studies depend on inductive reasoning. Principles and generalizations emerge from an examination of the data. Many case studies attempt to discover new relationships rather than verify existing hypotheses.

5.6.1 Advantages of Case Studies

The case study method is most valuable when the researcher wants to obtain a wealth of information about the research topic. Case studies provide tremendous detail. Many times researchers want such detail when they don't know exactly what they are looking for. The case study is particularly advantageous to the researcher who is trying to find clues and ideas for further research. This is not to suggest, however, that case studies are to be used only at the exploratory stage of research. The method can also be used to gather descriptive and explanatory data.

The case study technique can suggest why something has occurred. For example, in many cities in the mid-1980s, cable companies asked to be released from certain promises made when negotiating for a franchise. To learn why this occurred, a multiple case study approach, examining several cities, could have been used. Other research techniques, such as the survey, might not be able to get at all the possible reasons behind this phenomenon. Ideally, case studies should be used in combination with theory to achieve maximum understanding.

The case study method also affords the researcher the ability to deal with a wide spectrum of evidence. Documents, historical artifacts, systematic interviews, direct observations, and even traditional surveys can all be incorporated into a case study. In fact, the more data sources that can be brought to bear in a case, the more likely it is that the study will be valid.

5.6.2 Disadvantages of Case Studies

There are three main criticisms. The first has to do with a general lack of scientific rigor in many case studies. It was observed that in too many times, the case study investigator has been sloppy, and has allowed equivocal evidence or biased views to influence the findings and conclusions. It is easy to do a sloppy case study; rigorous case studies require a good deal of time and effort.

The second criticism is that the case study is not easily open to generalization. If the main goal of the researcher is to make statistically based normative statements about the frequency of occurrence of a phenomenon in a defined population, some other method may be more appropriate. This is not to say that the results of all case studies are idiosyncratic and unique. In fact, if generalizing theoretic propositions is the main goal, the case study method is perfectly suited to the task.

Finally, like participant observation, case studies are likely to be time-consuming and may occasionally produce massive quantities of data that are hard to summarize. Consequently, fellow researchers are forced to "wait years for the results of the research, which too often are poorly presented. Some authors, however, are experimenting with nontraditional methods of reporting to overcome this last criticism.

5.6.3 Conducting a Case Study

The precise method of conducting a case study has not been as well documented as the more traditional techniques of the survey and the experiment. Nonetheless, there appear to be five distinct stages in carrying out a case study: design, pilot study, data collection, data analysis, and report writing.

Design

The first concern in a case study is what to ask. The case study is most appropriate for questions that begin with "how" or "why." A research question that is clear and precise will focus the remainder of the efforts in a case study. A second design concern is what to analyze. What exactly constitutes a "case"? In many instances, a case may be an individual, several individuals, or an event or events. If information is gathered about each relevant individual, the results are reported in the single or multiple case study format; in other instances, however, the precise boundaries of the case are harder to pinpoint. A case might be a specific decision, a particular organization at a certain point in time, a program, or some other discrete event. One rough guide for determining what to use as the unit of analysis is the available research literature. Since researchers want to compare their findings with the results of previous research, it is sometimes a good idea not to stray too far from what was done in past research.

Pilot Study

Before the pilot study is conducted, the case study researcher must construct a study protocol. This document contains the procedures to be used in the study and also includes the data-gathering instrument or instruments. A good case study protocol contains the procedures necessary for gaining access to a particular person or organization and the methods for accessing records. It also contains the schedule of data collection and addresses the problems of logistics. For example, the protocol should note whether a copy machine will be available in the field to duplicate records, whether office space is available to the researchers, and what will be needed in the way of supplies. The protocol should also list the questions central to the inquiry and the possible sources of information to be tapped in answering these questions. If interviews are to be used in the case study, the protocol should contain the questions to be asked.

Once the protocol has been developed, the researcher is ready to go into the field for the pilot study. A pilot study is used to refine both the research design and the field procedures. Variables that were not foreseen during the design phase can crop up during the pilot study, and problems with the protocol or with study logistics can also be uncovered. The pilot study also allows the researchers to try different data-gathering approaches and to observe different activities from several trial perspectives. The results of the pilot study are used to revise and polish study protocol.

Data Collection

At least four sources of data can be used in case studies. Documents, which represent a rich data source, may take the form of letters, memos, minutes, agendas, historical records, brochures, pamphlets, posters, and so on. A second source is the interview. Some case studies make use of survey research methods and ask respondents to fill out questionnaires, others may use intensive interviewing.

Observation/participation is the third data collection technique. The same general comments made about this technique earlier in this chapter apply to the case study method as well. The last source of evidence used in case studies is the physical artifact—a tool, a piece of furniture, or even a computer printout. Although artifacts are commonly used as a data source in anthropology and history, they are seldom used in mass media case study research. (They are, however, frequently used in legal research concerning the media.)

Most case study researchers recommend using multiple sources of data, thus affording triangulation of the phenomenon under study. In addition, multiple sources help the case study researcher improve the reliability and validity of the study. Not surprisingly, a study of the case study method found that the ones that used multiple sources of evidence were rated higher than those relying on a single source.

Data Analysis

Unlike more quantitative research techniques, there are no specific formulas or "cookbook" techniques to guide the researcher in analyzing the data. Consequently, this stage is probably the most difficult in the case study method. Although it is hard to generalize to all case study situations, three broad analytic strategies were suggested: pattern matching, explanation building, and time series.

In the pattern-matching strategy, an empirically based pattern is compared with a predicted pattern or several alternative predicted patterns. For instance, suppose a newspaper is about to institute a new management tool: a regular series of meetings between top management and reporters, excluding editors. Based on organizational theory, a researcher might predict certain outcomes, namely, more stress between editors and reporters, increased productivity, weakened supervisory links, and so on. If analysis of the case study data indicates that these results did in fact occur, some conclusions about the management change can be made. If the predicted pattern did not match the actual one, the initial study propositions would have to be questioned.

In the analytic strategy of explanation building, the researcher tries to construct an explanation about the case by making statements about the cause or causes of the phenomenon under study. This method can take several forms. Typically, however, an investigator drafts an initial theoretical statement about some process or outcome, compares the findings of an initial case study against the statement, revises the statement, analyzes a second comparable case, and repeats this process as many times as necessary. For example, to explain why some new communication technologies are failing, a researcher might suggest lack of managerial expertise as an initial proposition. But an investigator who examined the subscription television industry might find that lack of management expertise is only part of the problem—inadequate market research is also contributory.

Armed with the revised version of the explanatory statement, the researcher would next examine the direct broadcast satellite industry to see whether this explanation needs to be further refined, and so on, until a full and satisfactory answer is achieved.

In the analytic strategy of time series analysis, the investigator tries to compare a series of data points to some theoretic trend that was predicted before the research, or to some rival trend. If, for instance, several cities have experienced newspaper strikes, a case study investigator might generate predictions about the changes in information-seeking behaviors of residents in these communities and conduct a case study to see whether these predictions were supported.

Report Writing

The case study report can take several forms. The report can follow the traditional research study format: problem, methods, findings, and discussion. Or it can use a nontraditional technique. Some case studies are best suited for a chronological arrangement, whereas case studies that are comparative in nature can be reported from that perspective. No matter what form is chosen, the researcher must consider the intended audience of the report. A case study report written for policy makers would be done in a style different from one that was to be published in a scholarly journal.