Cipriani!1 Patten Part H Outline 1. Sampling in Qualitative Research: I - Purposive sampling: Selecting a group of individuals who are likely to have relevant information to the research topic. Participants fit into a broad category. - Purposive criterion sampling: Specific criteria applied in the selection of a sample group. - Demographic variables Age Race/Ethnicity Marital status Number of children Highest educational level achieved Type of employment - Demographic sampling can be used for convenience to obtain predicted results. This is a flaw in sampling. 2. Sampling in Qualitative Research: II - Qualitative researchers can be interested in studying problems in specific locations or institutions. - Should seek participants from a variety of sources. This is a strength of qualitative research.
Cipriani!2 - Qualitative researchers typically use fewer participants than quantitative researchers Qualitative research is more expensive Also more time consuming Qualitative researchers almost never test for statistical significance and therefore do not need samples that are so large - Saturation: Researcher identifies emerging themes and continues to research to the point at which he or she does not find any new themes or trends. 3. Measurement in Qualitative Research: I - Semistructured interviews: Face to face tape recorded interviews (most commonly used method of data collection). Does not have to ask only predetermined questions - Interview protocol: written directions for conducting an interview with standard predetermined questions that have been pilot tested. - It is important to first establish rapport with the subject of the interview so that they are more inclined to be honest. - Interviewers must be experienced and unbiased! - Predetermined questions should touch on demographic information so that it can be referenced in a research report. 4. Measurement in Qualitative Research: II - Interviews are flawed because they expose the subjects perception of reality not what actually happened.
Cipriani!3 - Phenomenological approach: Examining participants perceptions. Almost al qualitative research contains some element of this. - Focus group: a small group of participants (6-12) that are used to discuss a research topic. Led by a facilitator. reveals the evolution of perceptions in a social context - Nonparticipant observation: researcher observes individuals as an outsider. - Participant observation: researcher becomes a member of the group without the group s knowledge. This method has serious ethical problems 5. Quality Control in Qualitative Research - Data triangulation: to establish dependability and credibility of data, researchers can use multiple sources for obtaining data on research topic. - Methods triangulation: One type of participant is used to provide data. Two or more methods are used to collect data. - Researcher triangulation: having a specific and focused research theme. Reduces possibility of results of research representing only the idiosyncratic views of one researcher. - Interobserver agreement: conducting initial analysis alone and then agreeing on or reaching the same result. This ensures that the research is dependable. - Member checking: Participant researchers review the results of the analysis. 6. Grounded Theory for Data Analysis
Cipriani!4 - Grounded theory: an inductive method of analysis that can lead to theories of behavior. Theories are developed based on data in qualitative research. - Open coding: sorting interview data and coding it according to name of participant or topic. - Axial coding: transcripts and data sources are reexamined and relationships between categories and themes that were identified in open coding. Types of relationships examined (a) temporal (b) causal (c) associational (d) valence (e) spatial - Core category: describe the process that leads to the relationships identified. - Constant comparison: constant comparison of each new element of data with previous elements. 7. Consensual Qualitative Research: Analysis - Consensual qualitative research: the goal is to get all researchers to come to a consensus about the meaning of the data. Code into domains: segmenting data into groups according to the topics they cover.
Cipriani!5 Develop core ideas within domains: writing short summaries to reduce original ideas of participants to shorter more cohesive statements. Cross analysis: group data into categories based on similarities. Higher level of generalization. Stability check: reexamining data and interview transcripts. - Internal stability: determining the extent to which each category was general, typical, or variant. - CQR requires the use of an auditor as an outside check.