Chapter 11 • Qualitative Data Collection and Analysis
Sampling techniques in qualitative research are intentional, as opposed to random.
This type of sampling is known as purposeful sampling.
There are several specific types of purposeful sampling:
In maximum variation sampling, the researcher selects cases that differ on an important characteristic.
Extreme case sampling focuses on the sampling of an outlying case.
Typical sampling involves the selection of a person or site that is typical to outsiders.
Theory or concept sampling helps the researcher generate or discover a new theory or concept.
In homogeneous sampling, sites or individuals are selected because they possess a similar trait.
Critical sampling focuses on individuals or sites that represent in dramatic terms the phenomenon being studied.
Opportunistic sampling allows the researcher to sample for new and different information as questions emerge in the study.
Snowball sampling relies on participants to recommend other potential participants for the study.
Confirming or disconfirming sampling allows the researcher to seek additional data to confirm or disconfirm preliminary findings.
Qualitative research sample sizes are typically very small, although they may range from a single individual or site to as many as 30 or 40.
It is the researcher’s responsibility to strike a balance between the amount of data to be collected and the depth of data sought.
There are numerous ways to collect qualitative research data.
Observations involve carefully watching and systematically recording what you see and hear in a setting.
Observations may be structured, unstructured, or semistructured.
Researchers may assume several roles along the participant–observer continuum.
When the researcher is in an observer role, participants may not even know they are being observed.
An observer as participant is primarily an observer but has some interaction in the setting.
A participant as observer acts as an observer but also interacts more formally with participants.
A full participant is a researcher who is also a fully functioning member of the community.
Observations are recorded in the form of field notes.
When observing and taking field notes, it is good practice to include observer’s comments, which are preliminary interpretations of observational data.
Interviews are formal conversations between the researcher and participants in the study.
Interviews may be conducted individually or in groups, known as focus groups.
Before interviewing participants, it is best to prepare an interview guide to be closely followed during data collection.
Interviews may be structured, semistructured, or open-ended.
Journals—including student journals, teacher journals, and class journals—can also be used to collect qualitative data at the research site.
Validity of research data deals with the extent to which the data collected accurately measure what the researcher intended to measure.
When establishing the validity of qualitative data, researchers are concerned with the data’s trustworthiness.
Trustworthiness is established by examining four criteria: credibility, transferability, dependability, and confirmability.
Additional criteria that can be used to establish the validity of qualitative research include descriptive validity, interpretive validity, theoretical validity, evaluative validity, and generalizability.
Triangulation is a process of using multiple methods, data collection strategies, data sources, and sometimes multiple researchers to enhance validity.
Persistent and prolonged participation in the study site will also enhance the validity of the research.
Enlisting other professionals to help review and critique your data collection and analysis to enhance the study’s validity is known as peer debriefing.
Having an outsider review the final report is called an external audit, and can also enhance validity.
Member checking is a process of asking participants to review the accuracy of the research report.
Reflexivity—the process of documenting and evaluating your interpretations, assumptions, and biases—also aids in establishing validity.
Although there is not a single method for analyzing qualitative data, the general approach is a process of inductive analysis.
Inductive analysis focuses on three main steps: organization of the data, description of coded themes, and interpretation of those themes.
Numerous software programs are available to assist the researcher in organizing and coding data.
Qualitative data analysis is complex and time-consuming. Multiple interpretations are a distinct possibility.
