Qualitative Research Methods

Qualitative research

Research whose findings are not arrived at through statistical means, but rather by attempting to understand behavior in a natural setting, through meaningful interaction with subjects

Credibility

Whether or not a study's finding represent a trustworthy interpretation of the data; should be possible to check how the results of the study were obtained

Rapport

a trusting and open relationship between the interviewer and the participant -- important in helping establish credibility of results

Iterative Questioning

Researcher looks for ambiguous participant responses, then returns to same topic later, seeking to ask further/new/differently worded questions -- aim is to see where participants are being less-than-honest, trying to "look good" in their responses

Credibility checks

this refers to checking accuracy of data by asking participants themselves to read transcripts of interviews of field notes of observations and confirm that the notes/transcripts are an accurate representation of what they said (meant) or did

"Thick" descriptions

rich and detailed descriptions of participant behavior/responses, which allows another person to rightly understand the behavior/responses -- gives context = helps improve credibility

Generalizability

Findings of a study are relevant outside the immediate context of that one particular study -- less important/less possible in qualitative research, where specific/unique groups are the population being studied

Transferability

findings of the study can be generalized from the study's participants to other people outside the study -- helped if "thick" descriptions available to readers of data

theoretical generalizability

The application of conclusions from findings based on a sample or case to larger sociological processes and theories about the world (e.g., schema, intelligence) -- can make this generalizability if we've reached data saturation, and have obtained "thick" description, and have confidence our research was (relatively) bias-free

Purposive sampling

Targets a particular group of people (as opposed to random sampling) -- participants are intentionally chosen on the basis of particular characteristics involved in the research -- quick, useful if target population is hard to find -- but, can reflect researcher bias

Non-probability sampling

the probability of any particular member of the population being chosen is unknown -- allows researcher to study a particular group who share some unique trait or behavior -- used typically in qualitative research

Quota sampling

A very tailored sample that's in proportion to some characteristic or trait of a population you're wanting to study (e.g., in total population, men = 45%, women = 55%; a sample of 100 includes 45 men, 55 women)

Snowball sampling

Sample achieved by asking a participant to suggest others who might be willing or appropriate for the study -- useful in hard-to-track populations, such as drug users, etc. -- but, introduces confidentiality concerns (participants know each other!), could be a biased sample

Convenience (a.k.a. opportunity) sampling

A sample made up of people who are easy to reach, near at hand - easy and cheap, but may not reflect target population's diversity and so may not be representative

Data saturation

When researchers conclude that further sampling and study will no longer lead to helpful new information

Triangulation

Use of different approaches to gathering/interpreting data in order to improve trustworthiness of findings

Rich data

Highly detailed, specific information gathered by researchers that is open to multiple interpretations -- boosts credibility

Data triangulation

Use of different data — Data from different sources (schools, individuals, teams, etc.), or data from different time periods

Methodological triangulation

Using different methods on a single study (case studies, observations, interviews, quantitative tests/data), to learn as much as can be learned about a behavior being studied

Researcher triangulation

Using different people as researchers, different observers to eliminate researcher bias — consulting colleagues on methods and data

Theory triangulation

Examining the data through different theoretical perspectives -- different levels of analysis or different theories within the same level of analysis

Reflexivity

Assumption that it's important for the researcher to be aware of his/her contribution (by acknowledging own background + beliefs) to the research process -- occurring throughout the research, which allows him/her to reflect on how bias may occur and influence the findings

Personal reflexivity

reflecting upon personal experiences and values, and how they influence research

Epistemological reflexivity

determining if the research was limited by the research question or methodological design, and if a different approach could have brought a different/better understanding.

Participant bias

Ideas about the research, researcher that can affect data -- participants want to "help out" the researcher by giving info they think the researcher is "looking for" -- or, participant feels they have to behave in certain ways in order to please the researcher -- can make data less credible

Researcher bias

Allowing researcher's beliefs to affect research process -- by handpicking participants, or focusing on some parts of data and not others -- reduces credibility of findings

Semi-structured interviews

Research type which, while having a set of goals beforehand, is generally flexible in how the questions are worded and in what order they occur, which allows the interviewer to go more in depth on certain questions -- focused on a particular topic, can ask participants for elaboration = richer data; less biased by researchers' preconceptions -- but does not allow to pursue themes not prepared in advance, one-on-one can create artificial results (ecological validity), data analysis is time consuming

Inductive Content Analysis

Assumes that theory can be derived from the data -- not that the data be made to fit a theory -- researcher reads and rereads the transcript of the interview and produces notes to assist analysis -- next, emerging themes identified/recorded, after which they will be together into higher-order themes, which are then organised into tables which will assist the researcher with analysis of the data

Participant observation

where the observer takes part in the situation that is being studied at the same time that they are performing research -- increases data that could lead to more valid interpretations, but can also lead to researcher favoring group, being "too involved" to be objective

Structured interview

A selection interview that consists of a predetermined set of questions for the interviewer to ask, asked in a fixed order -- useful for interviewing large numbers of participants, limits variables by having same questions for all

Non-participant observation

The researcher observes participants with or without their knowledge, and is not part of the group being studied -- greater objectivity, but greater chance of researcher imposing his/her own views/understandings on observed behaviors

Naturalistic observation

Observation takes place in the participant's' normal environment -- greater ecological validity, and can observe groups that otherwise might not be "observable" -- but outside variables may impact observed behavior, leading to misinterpretation by researcher

Unstructured interview

An interview in which the question-answer sequence is spontaneous, open-ended, and flexible -- each question is driven by participant's response to the last question -- gives greatest insight into participant's thoughts/emotions, but can be prone to wander off-topic.

Overt observation

Participants know they're being observed, are informed about research and given informed consent -- strong ethical considerations -- but quality of data depends on researcher/participant relationship -- participants may be more likely to alter their behavior due to being aware of researcher's presence

Focus group interview

A research technique in which a small group of persons (usually 8 to 12) comes together for an intensive discussion about a particular topic, with the conversation guided by a trained moderator using an unstructured method of inquiry -- quick way to gather data from multiple participants, more natural setting = greater ecological validity, participant interaction = richer data

Covert observation

Participants either don't know they are being observed (possible ethical issues = no informed consent) -- OR they aren't informed about the specific aims of the research (researchers make up a justification as to why they are there) -- important for when researcher's presence may be especially likely to affect behavior of participants

structured observation

researcher identifies beforehand which behaviors are to be observed and recorded -- allows for easier data triangulation and better credibility, but researchers may miss behaviors that aren't "on their list" of things to look for

Grounded theory

An inductive method that allows the data to "speak for itself," to naturally lead researchers to a theory that's clearly rooted IN the data

unstructured observation

method of observation in which the researcher has a great deal of flexibility in terms of what to note and record -- may allow for more behaviors to be noticed/recorded, but may leave lesser basis of comparison for multiple researchers

Verbatim transcription

Word for word transcription. Includes all dialogue spoken, fillers, false starts, and repetitions.

Case study

an in-depth investigation of a human experience -- can cover anywhere from one person to entire organizations

Postmodern transcription

The copied data not only the words of the interview, but other non-verbal cues as well such as pauses, body languages, sighs, etc.

acquiescence bias

a tendency for respondents to agree with all or most questions asked of them in a survey -- researcher must make open-ended questions, to combat this

Memos

Notes included in data analysis that explain to readers how and why interpretive decisions re: the data were made -- increases "thickness" of data, increases credibility of study

social desirability bias

the tendency to respond to questions in a socially desirable manner -- researcher must ask questions in non-judgmental ways, ask open-ended questions, to combat this

Thematic analysis

The most common form of analysis in qualitative research. It emphasizes pinpointing, recording, and examining patterns (or "themes") within data

dominant respondent bias

Occurs in a group interview setting when one of the participants influences the behavior and responses of the others -- researchers must be trained in how to ensure ALL participants have a chance to be heard

Sensitivity bias

The tendency of the participant to answer regular questions honestly, but distort their responses to questions on sensitive subjects (e.g., race, religion) -- researcher must build good rapport to combat this

confirmation bias

a tendency to search for information that supports our preconceptions and to ignore or distort contradictory evidence -- researcher must use reflexivity, to combat this

leading questions bias

occurs when respondents in an interview are inclined to answer in a certain way because the wording of the question encourages them to do so -- questions ought to be open-ended, allowing for multiple responses

question order bias

the tendency for earlier questions on a questionnaire to influence respondents' answers to later questions -- combat by asking general questions before specific ones, positive questions before negative ones, and behavior questions before attitude ones

sampling bias

A problem that occurs when a sample is not representative of the population from which it is drawn -- use random sampling, if possible -- if not, at least use different participants in different studies!

Cost-benefit analysis

When considering research that's potentially ethically problematic, the decision-making process in which adherence to ethical standards is weighed against the potential gain in knowledge for humanity -- usually determined in advance of research being conducted by an ethics committee at a university/institution

Ethics committee

A group of people within a research institution that must approve a study before it begins -- responsible for conducting the necessary cost-benefit analysis before a research study begins