Which of the following data collection techniques is not applicable for qualitative research

6.1 Interviews

The first and most popular data collection technique for eliciting idiosyncratic emotion content is structured and semistructured in-depth interviews with open-ended and probing questions activating an athlete’s recall of thoughts and feelings prior to and during best and worst performances. It is usually recommended in pilot and exploratory studies to generate idiosyncratic labels and then to aggregate most selected items into a stimulus list or standardized emotion scale. Identifying idiosyncratic content and personally meaningful labels to describe athletes’ subjective experiences is a clear advantage of these interviews over questionnaires with researcher-generated items and “fixed” emotion content. Experienced and verbally skillful athletes can provide detailed and meaningful accounts of their experiences (how they felt) and meta-experiences (how they interpreted and coped with these feelings) prior to and/or during the competition. Athletes who are less aware of their performance-related experiences might require some assistance initially in structuring and focusing their recall.

Data collection and analysis

Case study research typically includes multiple data collection techniques and data are collected from multiple sources. Data collection techniques include interviews, observations (direct and participant), questionnaires, and relevant documents (Yin, 2014). For detailed discussions of questionnaires, interviews and observation, see Chapter 16: Questionnaires, individual interviews, and focus group interviews and Chapter 17: Observation. The use of multiple data collection techniques and sources strengthens the credibility of outcomes and enables different interpretations and meanings to be included in data analysis. This is known as triangulation (Flick, 2014).

In case study research, the data collected are usually qualitative (words, meanings, views) but can also be quantitative (descriptive numbers, tables). Qualitative data analysis may be used in theory building and theory testing. Theory building may use the grounded theory approach. Theory testing typically involves pattern matching (Yin, 2014). This is based on the comparison of predicted outcomes with observed data. Qualitative data analysis is usually highly iterative. Visual displays of qualitative data using matrices (classifications of data using two or more dimensions) may be used to discover connections between the coded segments (Crabtree & Miller, 1999; Miles et al., 2014). Data analysis may be undertaken within a case and also between cases in multiple case study research (Eisenhardt, 1989). Quantitative data is typically presented in descriptive, tabular form and used to highlight characteristics of case study organisations and interviewees. See also Chapter 18: Quantitative data analysis.

Solicited and Spontaneous Data

A distinction that involves all primary data collection techniques is that between data that are solicited and data that are spontaneous. In experiments, surveys, and much qualitative research, the researcher uses a stimulus (experimental variable, survey question, or open question) to elicit information from the research subjects. Explicitly soliciting information has the advantage that the researcher can design the data collection to optimally provide data given the research question. However, the disadvantage is that the research subjects are aware that they are taking part in a scientific study. As a consequence, they may react to the announcement of the study topic, the institution that sponsors the study or carries it out, the individual experimenter or interviewer, and so on. It is not clear whether the recorded behavior or response is the “true” behavior, that is, whether it is the same behavior that would have occurred naturally, if it had not been elicited.

The possible reactivity of research subjects can be circumvented by observing natural activities or the traces they leave behind, without disturbing the research subjects in any way. Nonreactive or nonintrusive primary data collection methods include (covert) observation and monitoring. Observation, which can be done in the actual location or remotely using video technology, can lead to both quantitative and qualitative data. Increasingly, technological advances make it possible to monitor activities without disturbing the subjects. For instance, media research in general no longer relies on a panel of respondents who report on their television viewing; instead, in selected households a monitoring device is installed in the television that monitors the television use and transmits the information to the researchers without disturbing the respondents. Scanning devices are used to monitor consumer behavior. Internet behavior can also be monitored. For instance, when people visit a specific Web site, it is simple to monitor which banners and buttons they click on, how long they stay on the page, where they come from, and where they go to when they leave the site. All this provides information without directly involving the subjects in any way.

Interviews

Interviewing constitutes probably the most common and popular qualitative data collection technique. It normally involves a ‘dialogue’ with the researcher setting the agenda and asking questions and the interviewee being cast in the role of respondent. Nevertheless, interviews as a specific type of dialogue can be more or less structured. In structured interviews – rarely used in qualitative research – both the wording and the order of the questions are the same from one interview to another. In unstructured interviews, on the other hand, a free-flowing conversational style is adopted and respondents are encouraged to raise issues not originally included in the interview schedule. Biographical interviews which aim at the elicitation of research participants' personal stories with minimum researcher prompting constitute a paradigmatic example of unstructured interviews. Finally, in semistructured interviews, which are most commonly used in qualitative research, the researcher sets the agenda on the basis of their own interests and topics, but allows room for the participants' more spontaneous descriptions and narratives. Other distinctions are between one-to-one versus group interviewing, face-to-face versus telephone interviewing or interviewing through the Internet (Madill and Gough, 2008).

This article begins with a broad discussion of interviewing as a data-collection technique. It then offers a more theoretically nuanced understanding of how interviews and interviewing are conceptualized and used by quantitative and qualitative researchers. The remainder of the article focuses on three qualitative interview techniques: in-depth, ethnographic, and focus group interviews. It is suggested that these three styles are similar in that they all involve asking questions and receiving responses from research participants; however, they vary in terms of their emphasis on the inner self, social context, and group dynamics.

2 Data Collection Technique

The clinician him- or herself is, however, also a data collection “technique”. She or he is a critical element in the clinician-patient system (Henderson, 1935) and thus is an integral part of the data to be derived from the transaction between adult and child in the assessment setting. She or he is also critical to the collection of ecologically-important data from the nonclinical environment via the clinical interview.

Adult-Child System/Transaction. The behaviors of adult and child in the clinical setting are reciprocal. The adult naturally supports the transaction by supplying what is needed to facilitate optimal communication in the dyad. This requires the clinician to be aware of his or her own behavioral baseline, to monitor any change from baseline that this particular child under this particular demand elicits, and to actively test the hypotheses that such behavioral change sets up. Thus, observing that one is slowing, simplifying, repeating, and/or rephrasing one’s utterances in the course of ongoing conversation sets up a hypothesis of potential language impairment and requires that the examiner examine in detail the child’s language processing skills, both in linguistic interactions and on specific tests of language capacities—as well as other, not overtly related, skills that may also depend on the integrity of left hemisphere brain mechanisms. (These must be derived from both language and nonlanguage behavioral domains. Deficits in language alone would not be a sufficient test of the neuropsychological hypothesis, that is, one specified in terms of a neural substrate: such would only provide information at the psychological level of analysis.) Such a hypothesis also, however, requires that the examiner actively look for, and evaluate the impact of, other reasons for slowed output or need for repetition, such as a general rate of processing deficit, attentional instability, or hearing impairment. These would then be seen in the context of a different diagnostic behavioral cluster. Note that the change in the examiner’s behavior elicited during the interaction with the child will be a member of the diagnostic behavioral cluster, equivalent in this respect to test scores, quality of performance, historic variables, and so on.

The analytic interview. Interviewing technique, the ability to elicit information from caretakers, teachers, and so on that is as free from bias as possible, is crucial to any psychological assessment approach. Good interviewing technique is thus a sine qua non of the clinician’s armamentarium and should be undertaken in systematic fashion (Maloney & Ward, 1976). The interview is an intrinsic part of the neuropsychological assessment (as opposed to testing), and not separate from it. It is thus governed by the research design and theoretical principles of the assessment. Given this, interviewing strategies need to be extended and tailored to the neuropsychological context specifically. Interviewing is an active process in which no observation is taken “cold,” all observations are analyzed in light of their potential neuropsychological source or implications. Interviewees are thus queried to elucidate the actual behavior (rather than an interpreted version thereof) that they are describing. Strategies include: query providing a targeted contrast of a descriptive label (e.g., a child’s response of “This is boring” elicits “Is it boring-easy or boring-hard?”); clinical analysis of a descriptive label (a parent or teacher description of anxiety cues the skilled examiner to consider the actual behaviors that would lead the layperson to use the label “anxiety”—such as press of speech or motor activity—and to actively query the quality of speech and/or motor patterns with a view to evaluating the possibility of neuropsychological, rather than emotional, factors contributing to the observed behavior); and elicitation of relevant anecdotes (a complaint of memory problems in a child leads the clinician to ask for a specific example of the kind of situation in which the problem occurs—so that he or she can consider it from a broader neuropsychological perspective that may well include language processing or attentional issues, for example). The data from this analytic interview technique is crosschecked (where possible) against reports from other individuals/sources, and/or the neuropsychological hypotheses to which they give rise are tested against other types of assessment information (i.e., multimethod, multi-trait analysis).

Methods of Ecological Oral History

Ecological oral histories thus can be used for a wide variety of research projects. Data collection techniques often follow those used for other kinds of oral histories, although some novel methods are used as well. Many ecological oral histories follow techniques used in oral history and therefore use in-depth, semistructured interviews. In this type of interview, the researcher develops a set of questions to be asked, but does not specify the order in which they will be asked. Rather, the researcher strives to create a conversation with the participant(s) and asks questions where relevant or as a prompt to begin or redirect the interview. Unlike oral histories, however, ecological oral histories often utilize more interactive methods. Interactive methods include ambulatory interviews, mapping, and discussion of photographs. In ambulatory interviews, the researcher and informant or a small group of informants walk through a piece of land and identify plants or discuss changes and other features of the landscape. These kinds of interviews are often very effective and yield information not only on the landscape, but also on the relationship between the participants and the environment (i.e., nature–society interactions). They are used extensively in studies seeking to document TEK and also in research seeking to understand the social–political dynamics of land use. Other forms of interactive interviews use mapping or photographs of places. Participatory mapping is often considered a separate research technique in its own right and is used in nature–society studies or land-use planning research projects in addition to ecological oral histories. In these interviews, researchers ask people to draw a map of a place and then discuss what is included. Mapping provides a unique picture of landscape as seen from the perspective of the research participants. It can also illuminate important information on landscape features that may not be obvious to the researchers or the relative importance of particular resources. Photos of places are similarly used to invite discussion. Researchers either use photographs they have obtained or give participants cameras to take photographs of places or objects of importance to them. The researcher then discusses the photographs with the participants in order to lend a focus to the discussion. All these types of interviews can be done with groups as well as individuals. Group interviews are useful for obtaining accurate information on plant names and other aspects of shared knowledge of ecosystems, but they are equally useful for gathering information on the social politics of environments. A group context can help to illuminate which inconsistencies in oral histories are based on a lack of knowledge and which are based on competing claims. Members of the group correct each other or arrive at a consensus through discussion, helping to ensure accuracy as well as illuminating other dynamics that may be useful to the research.

8.2.1 Motivation

Different sources of data can be complementary. In this research we need to capture the complementarity of conventional data collection techniques with emerging data collection streams. What is the relation between these two paradigms? Do Big Data collection techniques obviate or underline the need for traditional data collection? One proposition is to consider Big data as a continuous update using inexpensive Big Data of large cross sectional survey data but methods to do this are not available widely. The opposite is also possible. Trends can be observed with Big Data and then an in-depth cross sectional survey is needed for behavioral analysis that explains the trends.

Willingness to Participate in Surveys and Response Rate

European national time-use surveys in the late 20th Century received relatively high response rates, irrespective of data collection technique. In the 21st Century response rates have decreased dramatically and even 40% is now regarded as rather high. There are many possible explanations to this development; people get more and more aware of integrity issues, they are too stressed and occupied to fill in the diary forms, or they do not find any meaning in doing so. Whatever reason, national statistical bureaus make efforts to overcome the problem. Creating new ways to collect diaries by using smartphone applications (apps) is one opportunity, another is to create simplified diary sheets, with a total of no more than 30 activity categories. In such a diary form, people are asked to choose between activity categories and draw a line in a time slot for the activities relevant for her/him. Still there is no consensus on what is the “new way” to collect time-space diaries from a sample, which is representative for a population.

There are research studies where apps with GPS functionality are utilized to perform studies on daily mobility. GPS provides precise data about places and movements, but information about activities will be less known if the app is not combined with a diary asking for activities. In some countries, data from telephone operators is used by researchers to map peoples' movements between and visits to places. This kind of data will not give information about the activities performed, apart from what can be derived from what is possible to do at the placed visited. However, there are many different kinds of activities that can be performed at many places, for example, both at home and in a shopping center, so the activity information derived from place location will be relatively unsatisfactory. In addition, in studies based on data from telephone operators it is not always the case that the individuals have given their consent to be part of the study.

As mentioned above, there are also smartphone apps developed for collecting more or less the same data as in the traditional pen and paper diaries. Vrotsou et al. suggest the use of an app developed where the mobile sets the time and the diarist fills in activity, place, togetherness, and other information depending on the purpose of the study in the course of the day. This app calls for the diarist's constant awareness and there is a risk that the diarist sometimes will forget to fill in the diary. Therefore, the app is complemented with a function where the diarist can edit the diary afterward.

In small-scale studies, where a personal relation develops between the diarist and the responsible researcher or professional therapist, the diarist is often more motivated to participate in the study. The diary process will provide the diarist with new knowledge on a personal level that makes it rewarding to write diaries. This is shown, for example, by the use of time-geographic time-space diaries in occupational therapy.

Step 5: Review New Evidence

Evidence-based practitioners collect data that allow the team to make decisions about whether or not the intervention is working. The data collection technique should match the practical question that is being asked. Procedures for determining the best type of data to collect are comprehensively outlined in books dedicated to this topic (e.g., Kazdin, 2011). However the appropriateness of a given data collection is also based on who is collecting the data and their experience with data collection. Issues relevant to data collection are briefly discussed further in Chapter 11, Progress Monitoring.

Evidence-based practitioners collect data on treatment fidelity, quality of adherence, and the extent to which the treatment has been implemented as planned. Treatment fidelity data are essential both for determining if the treatment is feasible and if the target client is actually accessing the treatment. A treatment that is not being accurately implemented should not be rejected unless the team determines it is not feasible. First, the evidence-based practitioner must facilitate a problem-solving discussion to overcome obstacles. Even highly qualified professionals can follow the steps of a protocol but miss an important component, so the quality of implementation should be evaluated. The steps of a treatment can be technically implemented, but an essential aspect of one or more of the steps may not be sufficient. This is addressed in more detail in Chapter 11, Progress Monitoring. Evidence-based practitioners compare when the intervention was implemented in relation to the proposed schedule in the implementation plan. It is also important to identify barriers that prevent scheduled implementation and determine if these barriers have been adequately addressed.

Several forms of evidence will need to be collected on an ongoing basis once the treatment is implemented. For example, client preference should be assessed again after the treatment has been initiated. In addition, evidence-based practitioners assess the tolerability of the treatment. That is, can the target client tolerate the treatment as indicated by positive or negative enthusiasm or affect. Consumer satisfaction data are collected from the target, stakeholder, and leader clients. Through intervention implementation, these consumers may determine that they find a treatment unacceptable or that unanticipated barriers to treatment fidelity make the intervention unfeasible. Evidence-based practitioners continue supporting the team based on ongoing data collection as further determinations about the treatment are made.