- Characteristics of qualitative research
- Natural settings - field research
- Behavior is studied as it occurs naturally
- Beliefs related to a natural setting
- Behavior is understood bests as it occurs without external
constraints or control
- The situational context is very important to understanding
behavior
- Data collection - data is collected directly from the source
- Observations
- Interviews
- Document analysis
- Rich narrative descriptions
- Process orientation - how and why behaviors occur
- Inductive data analysis
- Participant perspectives define what is "real"
- Emerging research design - the design plans change as data is collected,
analyzed, and understood
- Four type of qualitative designs
- Ethnography
- Case study
- Phenomenology
- Grounded theory
- Assumptions that differentiate qualitative and quantitative studies
- Epistemology
- Qualitative researchers believe there are multiple realities
represented by the participants' perspectives
- Quantitative researchers believe a single, objective reality exists
- Context
- Qualitative researchers believe context is critical to understanding
the phenomena being studied
- Quantitative researchers do not believe context is an important
factor
- Researcher bias
- Qualitative researchers believe the researcher's biases and
perspectives must be understood to interpret the results
- Quantitative researchers believe researcher bias is controlled
through the control of internal validity threats
- Ethnography (see Table 11.4)
- An ethnography is an in-depth description and interpretation of cultural
patterns and meanings within a culture or social group
- Culture - shared patterns of beliefs, normative expectations,
behaviors, and meanings
- Shared, not individualistic
- Problem statements
- Foreshadowed problem - a general framework for beginning a
qualitative study
- Specific question - a question(s) that emerge from the interactive
relationship between the problem and data
- Often found embedded in the data analysis
- Changing nature of questions often necessitates changes in
the design (i.e., an emergent design)
- Identifying and entering the research site
- Access to all parts of the site (e.g., the participants, documents,
physical locations, etc.)
- Rapport - need to be "integrated" within the site to gain the trust of
the participants
- Often site entry takes a long time
- Selecting participants
- Use of purposeful sampling strategies to select "information rich"
participants
- Purposeful sampling strategies
- Maximum variation - selecting individuals or cases to
represent extremes (e.g., very positive or very negative
attitudes, highest and lowest achieving students)
- Snowball (i.e., network) - initially selected participants
recommend others for involvement
- Sampling by case - selecting individuals or cases for their
unique characteristics
- Extreme
- Typical
- Unique
- Reputation
- Key informant - selecting an individual(s) particularly
knowledgeable about the setting and or topic
- Comprehensive - selecting all relevant individuals or cases
- Obtaining data
- Observation
- Unstructured in nature
- Comprehensive - continuous and total over an extended
period of time
- Participant-observer role of the researcher
- Continuum between complete participant and
complete observer
- Passive participant
- Moderate participant
- Active participant
- Complete participant
- Rare for an ethnographer to be a complete participant
- Use of field notes to record observations
- Two types of information
- Descriptions of what occurred
- Reflections of what the descriptions mean (i.e.,
speculations, emerging themes, patterns,
problems)
- Accuracy
- Extensive nature of notes
- Interviews
- Unstructured in nature
- Begins with a general idea of what needs to be asked and
moves to specific questions based on what the respondent
says
- Types of interviews
- Key informant
- Life history
- Focus group
- Tape recording and transcribing interviews afford the
opportunity to study the data carefully
- Document analysis
- Written records
- Print (e.g., minutes from meetings, reports,
yearbooks, articles, diaries)
- Non-print (e.g., recordings, videotapes, pictures)
- Types of sources
- Primary - original work
- Secondary - secondhand interpretations of original
work
- Commonly used to verify other observations or interview
data
- Data analysis
- Observations, interviews, and document analyses result in large
quantities of narrative data
- Analysis includes critically examining, summarizing, and
synthesizing the data
- Three stages of analysis
- Coding
- Organizing the data into reasonable, meaningful units
that are coded with words or very short phrases that
signify a category
- Emic categories - information provided by the
participants in their own language and
organizational units
- Etic categories - the researcher's interpretation
of emic data
- Use of major codes and sub-codes is common
- Summarizing the coded data
- Examining all similarly coded data and summarizing it
with a sentence or two that reflects its essence
- Computerized sorting of data is common and
effective
- Pattern seeking and synthesizing
- Synthesizing identifies the relationships among the
categories and patterns that suggest generalization
- The researcher interprets findings inductively,
synthesizes the information, and draws inferences
- Pattern seeking
- Begins with the researcher's informed hunches
and ideas
- Tentative patterns are identified and additional
data collected to determine if they are
consistent with those patterns
- Characterized by enlarging, combining,
subsuming, and creating new categories that
make sense
- Case studies (see Table 11.4)
- An in-depth analysis of one or more events, settings, programs, groups, or
other "bounded systems"
- Focus on one entity
- Defined by time and place
- Types of case studies
- Historical organizational - focus on the development of an
organization over time
- Observational - study of a single entity using participant
observation
- Life history (i.e., oral history) - a first-person narrative
completed with one person
- Situation analysis - a study of a specific event from multiple
perspective
- Multi-case - a study of several different independent entities
- Multi-site - a study of many sites and participants the main
purpose of which is to develop theory
- Concern with the limited generalizability of the findings
- Research problem statement
- Focus on in-depth description and understanding
- Use of a single major question and several sub-questions
- Emerging nature of the problems
- Identifying and entering the research site
- Selecting participants
- Participants are usually identified as a part of the site of the study
(e.g., a classroom, teachers in a specific department, etc.)
- Internal sampling - selecting specific participants, times, and
documents within a site
- Obtaining data
- Data analysis
- Same procedures as in ethnographic data analysis
- Four types of data analysis
- Categorical aggregation - researcher codes data and
collects instances from which meanings will emerge
- Direct interpretation - use of a single example to illustrate
meaning
- Drawing patterns - examines the correspondence between
two or more categories or codes
- Naturalistic generalization - suggestions as to what others
can take from the research and apply to other situations
- Phenomenological studies (see Table 11.4)
- A phenomenological study describes and interprets the experiences of
participants to understand their perspectives
- Based on the belief that there are multiple ways of interpreting the
same experience and the meaning of that experience is what
constitutes reality
- Research problem - focused on what is essential for the meaning of the
event, episode, or interaction
- Selecting participants
- Participants are selected because they have lived or are living the
experience being investigated
- Participants will share their experiences
- Participants can articulate their feelings
- Obtaining data - in-depth, semi-structured, or unstructured interviews
- Data analysis
- Concerns that the analysis reflects the shared meanings and
consciousness of the participants
- Five step process
- A initial description of the researcher's experience with the
phenomena
- A statement how the participant's experience with the
phenomena are identified in the interview
- The creation of meaningful units from the statements using
participant's verbatim language to illustrate the units
- Separation of what was experienced from how it was
experienced
- The construction on an overall description of the experience
- Grounded theory studies (see Table 11.4)
- A grounded theory study discovers or generates a theory
- A theory is a set of propositions that pertain to a specific
experience, situation, or setting
- The contextual sensitivity of the theory is the basis for suggesting
the theory is "grounded" in the field data
- Research problems - broad general questions that focus on what
happened to people, why they believed it happened, and what it means to
them
- Selecting participants
- Obtaining data - in-depth unstructured interviews
- Data analysis
- Constant comparison - information from interviews is compared to
emerging themes as a part of a more encompassing theory
- Four step process
- Form initial categories with subcategories and descriptions
of extreme possibilities on a continuum
- Create a coding paradigm in which central tenets are
described with causal conditions, resultant actions,
conceptual conditions and consequences
- Write a story that integrates selective codes that have been
established and presents conditional propositions and
hypotheses
- Explicate the theory
- Credibility of qualitative research
- Credibility is the extent to which the data, data analysis, and conclusions
are believable and trustworthy
- Four technical issues related to credibility
- Triangulation - the comparison of results obtained from different
data collection methods (i.e., interview, observation, and document
analyses all lead to a similar conclusion)
- Reliability - the extent to which what is recorded as data is what
actually occurred in the setting (i.e., the accuracy of observations)
- Internal validity - the match between the researcher's categories
and interpretations and reality
- Threats related to observer effects are of paramount
concern
- Other threats include maturation, history, selection, attrition,
and subject effects
- External validity - generalizability
- Translatability and comparability are terms used to indicate
the extent to which the results can be used by other
researchers in other settings
- Generally weak in qualitative research
- Techniques to enhance credibility
- Triangulation
- Prolonged and persistent field work
- Copious field notes
- Low inference descriptors
- Mechanically recorded data
- Member checking
- Verbatim accounts
- Abundant use of detail
- Data collection in natural settings
- Researcher's role as participant observer
- Resources for qualitative research
- Criteria for evaluating qualitative research
- The researcher's background, interests, and potential bias should be clear
- Conceptual and/or theoretical frameworks for the study should be clear
- The method for selecting participants should be clear
- The level of the researcher's involvement in the setting should be
indicated
- The researcher should be trained in data collection procedures
- Credibility of the research should be addressed
- Descriptive data should be separated from the interpretations of the data
- The researcher should use multiple methods of data collection
- The duration of the study must be long enough
- Mixed method research designs
- Designs combining quantitative and qualitative approaches to collecting,
analyzing, interpreting, and reporting data
- Examples
- Use of a questionnaire to provide an overview of students'
attitudes toward drug testing programs followed by several
in-depth interviews of specific students with positive and
negative attitudes (i.e., maximum variation sampling) to
understand how those attitudes were shaped.
- A few interviews with students and a content analysis of
several surveys allowed a researcher to determine the
important factors around which an attitudinal scale on drug
testing programs was developed. Administration of this
survey gave an overall view of students' attitudes to a
specific program being used at a local school.
- The use of a scale addressing attitudes toward drug testing
programs could be administered to the students in a school.
Information from focus groups and interviews could be used
to confirm the conclusions drawn from the survey.
- Advantages
- Incorporates the strengths of both qualitative and quantitative
approaches
- Provides a more comprehensive view of the phenomena being
studied
- Does not limit the data being collected
- Disadvantages and limitations
- Requires expertise in both methods
- Requires extensive data collection and resources
- It is popular to claim the use of mixed method design even though
one method is used superficially
- Types of designs
- Explanatory
- Quantitative data are collected first with qualitative data
collection following
- See the example of an explanatory design discussed above
- Exploratory
- Qualitative data are collected first with quantitative data
collection following
- See the example of an exploratory design discussed above
- Triangulation
- Quantitative and qualitative data are collected at the same
time to provides a more comprehensive and complete set of
data
- See the example of a triangulation design discussed above
- Seven steps to conduct mixed method research
- Determine the feasibility of a mixed method design
- Determine the rationale for a mixed method design
- Determine the data collection strategies and designs
- Determine the specific research questions
- Collect data
- Analyze data
- Write the report
- Evaluating a mixed method design
- Each approach has been done well
- Examine the rationale and then apply appropriate quantitative or
qualitative criteria
