Here are the characteristics of a good qualitative data: 1) naturalistic (derived from actual participation or analysis of a subjective data, 2) “rich” and “deep” data (that is, specifics of dynamics of an event or context can be discerned or analyzed), 3) subjective (data should be perceptions of the people in the environment), 4) credible (that is, the data are derived from actual experiences of the people involved – the source of data), and 5) confirmable (that is, the data derived may be collaborated by other subjective sources).
There is though an additional characteristic (but not required) of a good qualitative data. In some cases, social scientists attribute a good qualitative data based on its transformability into quantitative data (this is though not necessary). Types of Qualitative Data Collection (Qualitative Methods, 2006: URL cited) There are generally four qualitative data collection methods that are frequently used in the social sciences. Here are as follows: 1) participant observation, 2) direct observation, 3) unstructured interviewing, and 4) case studies.
There are though variations in qualitative data collection methods. In anthropology, ethnography is used as the primary mode of qualitative data collection. In a sense, it is case study on a wide range (all aspects of a culture are examined and analyzed). In psychology, psychoanalysis methods of qualitative data collection are used to code and validate a person’s psychological standing or perhaps his/her state of mind. For simplicity’s sake, we shall not tackle on these methods. Participant Observation.
This method requires that the researcher become a participant in an event or the place being observed. This approach allows the researcher to know the specifics as well as the intent of an activity or the people involved. Without bias or prejudgment, this method becomes more pronounced when the researcher is accepted as a natural part of the culture, assuming that the observations are natural phenomena. Here, the researcher collects first-hand qualitative data, and hence allows him/her to relate it simultaneously with the event or activity (or culture).
Direct Observation. This is a different from the previous method in a number of ways. First, the researcher is not a participant in the context or event. The researcher in this case does not in any way mingle or influence the actions of the participants in a context or event. Doing so would undermine data authenticity as well as validation (see Hawthorne Effect). Second, direct observation is a detached perspective. Technologies replace actual participation as a measuring tool for validation and procurement of good qualitative data (as presented earlier).
Third, the researcher is observing sampled situations or groups of people; in no way the researcher is immersed in the activity or event. Lastly, direct observations are usually shorter in scope than participant observation in terms of data viability as well as practicality. Unstructured Interviewing or In-Depth Interviewing. Here the researcher and the respondent have direct interaction. The researcher usually uses a short guide to his interview questions (unstructured) or core concepts to ask about. The interviewer may ask additional or supporting questions that are relevant or connected to the main problem of the research.
This allows the researcher flexibility in structuring his/her qualitative data as well as representative tools like bodily gestures and facial expressions. The protocol however in this type of method is that the interviewer respects the principle of confidentiality. Only information approved by the interviewee can be released to the public or to academic associations. Case Studies. This is the frequently used research method in the social sciences (especially in anthropology and sociology). This involves an
In a sense, this is a combination of structured interview, participant observation, and direct observation. Using all the methods in one setting allows the researcher to get the whole picture of the problem. It also allows him/her to determine the variables or factors at play without undermining validity. Most of the time, if only one method is utilized, there is a tendency for variables to be neglected or misrepresented. There was a case when voting behavior was concluded to be attributed to the party affiliation of the group being studied (participant observation was the only method used) – note that only one variable was used.
When the study was replicated using combinations of methods, party affiliation accounted only 19% of the relations (when the qualitative data was converted to quantitative data). Although this is not to say that a combination of data is more desirable or more academically reliable (this would depend on the context of the research problem), it is often noted that this type of method has all the requirements for procuring good qualitative data. Process for Analyzing Qualitative Data
Qualitative data analysis is composed of three general processes: 1) noticing things, 2) collecting things, and 3) thinking about things. These three general processes are connected are related with each other. We shall discuss each of the processes below. Noticing Things. This refers to the general observation of an event or context and the manner by which it is coded. It generally means “making observations, writing field notes, tape recording, interviews, gathering documents, etc. When you do this you are producing a record of the things hat you have noticed”(Seidel, 1998:3). Collecting Things. This process is similar to solving jigsaw puzzles (Seidel, 1998:5). The data coded are assembled or disassembled into groups. In this way, relations can easily be extracted. Thinking About Things. This is generally the theoretical part of the research process. Each part of the “puzzle” are examined and related to the main problem. After relationships between variables are stated, they are then referred to the main problem (as well as the specific propositions).