7 mixed methods - quantitative and qualitative Flashcards
What are the main components of a qualitative analysis?
see the world of the participants
-> more insight as the focus is on textual rather than numerical data
why instead of how fast, how many
data must be collected in rigid and controlled way
-> validity as opposed to reliability
much more flexible and data-led
wording of the questions, style of asking the questions, order of the questions can vary
What types of qualitative studies can be conducted?
- reviews
pre-existing documentation (guidelines, policy documents, archived interviews, dairy entries, …) - observation
involves observing a particular setting or behaviour - or both
non-participatory - avoiding influence - monitoring from distance - interviews
structured, semi-structured, unstructred - focus groups
group interviews
Consider this scenario: Grace is a speech and language therapist and wants to develop a new approach to the treatment of language deficits following a left-hemisphere stroke. To do this, she first reviews the existing guidelines on post-stroke language treatment. She then spends a week making notes while working in her usual role on a stroke ward. This allows her to consider in more detail how the reviewed guidelines are actually implemented. Grace then finishes the first stage of her research by discussing the approach to post-stroke language treatment in a series of separate discussions with her colleagues.
Which of the main qualitative approaches would Grace be using in her research plan?
review then observation then interviews
What is one definition of qualitative research?
Qualitative research is defined as
“the study of the nature of phenomena”
, including
“their quality, different manifestations, the context in which they appear or the perspectives from which they can be perceived”,
but excluding
“their range, frequency and place in an objectively determined chain of cause and effect”.
How can qualitative methods be administered in health and health services research?
wide range of research questions generated by complexity of areas of interest
culture of the healthcare system, organisation, atmosphere are vital
→ diminish RCTs reliability
lay and professional health beliefs
qualitative description is a prerequisite of good quantitative research
naturalistic methods
development of concepts which help us to understand social phenomena in natural settings
‘What is X and how does X vary in different circumstances, and why?”
Answering the “what is X” question, though, is the foundation of quantification:
until something is classified it cannot be measured.
no a priori categories
How can qualitative methods complement quantitative ones?
- complementary rather than exclusive
- qualitative work can be conducted as an essential preliminary to quantitative research
- supplementation
triangulation/multimethod approach - exploring complex phenomena or areas not amenable to quantitative research
What is epistemology?
theory of knowledge; scientific study which deals with the nature and validity of knowledge
What is naturalistic research?
non-experimental research in naturally occurring settings
Social anthropology-social scientific study of peoples, cultures, and societies;
particularly associated with the study oftraditional cultures
What is induction?
process of moving from observations/data towards generalisations, hypotheses, or theory;
What is grounded theory?
hypothesising inductively from data, notably using subjects’ own categories, concepts, etc; opposite of deduction, process of data gathering to test predefined theory or hypotheses
What is purposive or systematic sampling?
deliberate choice of respondents, subjects, or settings, as opposed to statistical sampling, concerned with the representativeness of a sample in relation to a total population. Theoretical sampling links this to previously developed hypotheses or theories
What are fieldnotes?
collective term for records of observation, talk, interview transcripts, or documentary sources.
What is content analysis?
systematic examination of text (field notes) by identifying and
grouping themes and coding, classifying, and developing categories
What is analytic induction?
use of constant comparison specifically in developing hypotheses, which are then tested in further data collection and analysis
What is triangulation?
use of three or more different research methods in combination;
principally used as a check of validity
What are consensus techniques?
Delphi and nominal group techniques and consensus development conferences. They provide a way of synthesising information and dealing with conflicting evidence, with the aim of determining extent of agreement within a selected group
What is validity?
extent to which a measurement truly reflects the phenomenon under scrutiny
What is the hawthorne effect?
impact of the researcher on the research subjects or setting,
notably in changing their behaviour
What is reliability?
extent to which a measurement yields the same answer each time it is used
What are three components of reflixivity?
- demonstrating awareness of how biases may emerge
- thoughtfulness about, and attempts to minimise, the impact of the researcher on the data collected
- attempts to address bias through systematic and comprehensive analysis, and reflectiveness on the research methods
What are key features of semi-structured interviews?
scheduled in advance
location outside everyday events
organised around a set of predetermined questions
other questions emerge from the dialogue
usually last 30mins to several hours
What are key features of unstructured interviews?
guided conversations
originate from ethnographic and anthropological traditions
key informants are selected
interviewer elicits information
Who is a good informant?
- knowledgeable about the topic
- able to reflect and provide detailed experiential information
- willing to talk
What needs to be set before the interview?
- purpose
- clarification of topic
- format
- appropriate length
- assurance of confidentiality
- digital recorder? permission
- assure participant they may ask for clarification
- assure participant they can decline to answer
- assure participant there will be opportunities to ask questions
What are the phases of an interview?
- building rapport
- apprehension phase (general convo about interview, prompt questions)
- exploration phase (more in depth, open-ended questions, probing)
- co-operative phase (free discussion, not a chat)
- participation phase (greatest rapport, not always reached, interviewee guides and teaches the researcher)
- concluding the interview (positive note, humor)
What are possible probing techniques to be used during the exploration phase of an interview?
silent - thinking time allowed
echo - repeats the point
verbal agreement - expresses interest
tell me more - expand
long question - short one, then longer one
leading - explain reasoning
baiting - interviewer gives impression that they might be aware of certain information, prompting the participant to explain further
What is important about the transcription process of an interview?
verbatim
practising the technique!
What are ethical duties during and after an interview?
confidentiality
protection of study participants such that individual identities are not linked to information provided and are never publicly divulged
consequences
interview might be the only place where the topic is discussed
→ strong feelings
→ reflective discussions
What is mixed method research?
combination of qual and quan
-> gain a more in depth understanding
-> strengthen both validity and reliability
methodological eclecticism
(combination might cancel out their respective weaknesses)
How did mixed methods originate?
late 1970s from applied fields in social and behavioural sciences
formative component to summative component
causal effects as opposed to causal mechanisms
narratives might be intrinsically more interesting than numbers
What are possible designs of a mixed method approach?
convergent parallel design
(go in parallel, do not influence each other)
exploratory sequential design
(one follows after the other)
+ nested
(both administered at the same time)
What types of research problems can be adressed through mixed method?
assessing complex multi-component interventions or systems
what works for whom, when, how and why?
focussing on intervention improvement
What are primary characteristics of mixed method approaches?
paradigm pluralism
(variety of paradigms may serve as underlying philosophy)
emphasis on diversity at all levels of the research entreprise
emphasis on continua rather than a set of dichotomies
Iteratice, cyclical approach to research
deductive and inductive logic
context of discovery
context of justification
focus on the research question in determining the methods employed within any given study
set of basic signature research designs and analytical processes
What is the balance and comprise tendency?
- decline in conftrontationalism
- paths for fruitful dialog between paradigms
- cooperate and work with one another
What is the issue related to the paradigms?
three accepted paradigmatic positions:
- pragmatism and its interpretations
Howe (1988) - everyday pragmatism - frameworks associated with the axiological assumption
researchers need to be explicit about how to adress issues of social justice power differences and ethical implications - dialectical stance, which involves using multiple assumptive frameworks within the same study
paradigms are being mixed → away from attributes of paradigms but
distinctive attributes
-> maybe we need to deconstruct paradigms and open up
- world views
- epistemological stances
- model examples
- shared beliefs in community of scholars
What are issues related to the language of MMR?
bilingual - quan+qual
developing new language that crosses methodological designs
What are issues in conducting MMR?
- A diversity of options
many data analysis procedures are design-bound
there is no correct design
frameworks should be mixed
design is separable from research purpose
- data analysis
identification of analogous analytical processes
generation of unique lexicon of MMR analysis procedures indigenous to
the area
What are considerations when evaluating MMR?
- Does the research question justify the use of mixed methods?
- Is the method sequence clearly described, logical in flow and well aligned with study aims?
- Is data collection and analysis clearly described and well aligned with study aims?
- Does one method dominate the other or are they equally important?
- Did the use of one method limit or confound the other method?
- When, how and by whom is data integration (mixing) achieved?
What are critiques of MMR?
- incompatibility thesis
- MMR subordinates qual methods to a secondary position to quan methods
- logistical concerns, costs
- methodological bilingualism is unworkable
- expertise
- fitting into MMR literature - writing good articles
- prematurely headed toward fixed consensus for social inquiry
What categories would a decision tree for statistical approaches have?
- group differences or correlation/prediction
- how many dependent variables
- how many independent
- How many levels in the independent variables?
- is it categorical?
- are the same participants used in all conditions?