Week 4 Flashcards
(40 cards)
1
Q
What is qualitative research?
A
- Can be defined as the study of the nature of phenomena and is especially appropriate for answering questions of why something is (not) observed, assessing complex multi-component intervention and focussing on intervention improvement
- Observational
- Subjective
Lived experiences/patients perspectives
2
Q
Common qualitative data collection
A
- In depth interviewing
- Focus groups
- Participant observations
Ethnographic studies: Observational
3
Q
Qualitative data analysis and most common analysis methods
A
Transcription and coding
- Data cleaning
Categorization
- Recognizing relationships and developing categories
Most common analysis methods
- Thematic analysis
Content analysis
4
Q
Example of content analysis
A
- e.g. frequency of word finding difficulties: know what it is but can’t get it out
Difficulty of understanding
5
Q
Popularity of qualitative research
A
- Usually much cheaper than quantitative research
- No better way than qualitative research to understand in-depth the motivations and feelings of clients
- Qualitative research can improve the efficiency and effectiveness of quantitative research
Covid-19: No intervention research possible
6
Q
Limitations of qualitative research
A
- Qualitative research doesn’t distinguish these differences as well as qualitative research can
- Not representative of the population that is or interest to the researcher
The multitude of individuals who, without formal training, profess to be experts in the field
- Not representative of the population that is or interest to the researcher
7
Q
Mixed methods?
A
A combination of qualitative and quantitative research
8
Q
Qualitative vs Quantitative
A
Types of Questions: Probing vs limited probing
Sample size: Small vs large
Info Per respondent: Much vs varies
Type of analysis: Subjective, interpretive vs statistical
Type of research: Exploratory vs descriptive vs causal
9
Q
Common qualitative research methods:
A
Selection and recruitment of participants
Purposive sampling, Theoretical sampling
10
Q
Purposeful sampling
A
- Deliberate selection of specific individuals, events or settings because of the crucial information they can provide and cant be obtained sufficiently through other channels
- Selects cases with the purpose of providing a representative sample of the different processes involved rather than a representative sample from the population which its drawn
- Strategies include: criterion sampling, extreme case sampling, homogenous sampling, snowball sampling
- Snowball: Reasonably common, as it provides a convenient way of identifying individuals who are likely to have pertinent knowledge or experience
Sampling process begins with identification and inclusion in the study of one or more individuals who have the knowledge or experience and asking them to identify others with similar knowledge or experience
11
Q
Theoretical sampling
A
- Sampling strategy of grounding theory
- The sample evolves during the research, with sampling proceeding on theoretical grounds, rather than representative ground
- New units or cases are selected to be part of the sample on the basis of the need to fill out particular concepts or theoretical points
VIDEO: Theoretical sampling
- Also purposeful sampling, however purposeful isn’t always theoretical
- Purposeful: Sampling where you have a purpose and consciously make a decision of who you want
- Theoretical: Sampling that happens during the study after data analysis, as a theory forms, you recruit more people who you believe to be the most suitable people
12
Q
Sample size and saturation
A
- Small numbers of individuals, in comparison with most quantitative studies.
- However, they may generate a relatively large amount of data, as each of the individuals is studied, in-depth, to realise as fully as possible their potential to contribute to a rich description, and hence, an in-depth analysis, of the phenomenon of interest.
13
Q
Data Saturation
A
- Means of determining participant numbers required for the research design
- Considered to occur when little or no new data is being generated
- Saturation is considered to occur when little or no new data is being generated. This is the point at which recruitment of additional participants is unlikely to add to the understanding of the phenomenon under investigation.
- Papers may cease sampling when they reached saturation
- Data analysis often occurs as data is collected
Saturation is reached when the researchers fail to find new ideas, categories or themes emerging and they then decide that further data collection will not add anything more to the analysis and interpretation
14
Q
Methods of collecting data in qualitative studies
A
Common methods of data collection in qualitative research are:
- * Focus groups
- * Interviews (structured, semi-structured, in-depth)
- * Observation
- * Document review
- * Photographs
* Delphi Technique
15
Q
In-depth interviewing
A
- The process requires the interviewer to use semi-structured interview schedules rather than fixed questions and aims to engage the interviewee in conversation, to elicit their understandings and interpretations.
- Implicit in this method is the assumption that people have the particular and essential knowledge about the social world, which is obtainable through tapping into verbal messages.
- In-depth interviewing can yield informative data about a wide range of health issues, including sensitive personal issues, which people may find challenging to talk about. In this instance, the interviewers would need training and supervision to support data collection.
- Some issues may be investigated using a survey that uses flexible, open-ended questions, however, data obtained from this method is less detailed and is unlikely to provide the “thick” descriptive data sought by qualitative researchers.
VIDEO: In-depth interviewing
16
Q
Focus groups
A
- Focus groups are a qualitative data collection method that is commonly used in health research. It requires group interaction and interactive discussion of a particular topic of interest.
- Typically, there is a moderator (or facilitator) who acts as the leader of the group.
- The participants (usually between eight and ten) interact with the other members of the group and express their views, through participation in the group’s discussion of the issues, during the focus group session.
- Thus, there is a range of thoughts, feelings, and behaviours, about which many people would not be prepared to make disclosures in a group setting, but which they may be prepared to discuss in a one-to-one interview
- . Another drawback of focus groups is that they can come to be dominated by a few members of the group and the data then becomes biased towards the comments of these group members.
17
Q
Observation
A
The process involves the researcher compiling a primary record, or “thick” record, which requires extended participant observation and detailed recordings of what was observed through the taking of field notes. In ethnographic studies of organisations, the researcher seeks to observe “patterns of interest” which are usually those activities where people frequently come together to “act”.
18
Q
Content analysis
A
Content analysis is a form of data analysis used in both qualitative and quantitative research. It involves the identification of codes, for a number of concepts, before searching for their occurrence in the data. It is a deductive methodology, in contrast to the inductive methodology of thematic analysis
19
Q
Thematic analysis
A
- Identification of themes, through careful reading and rereading of the transcripts of interviews or focus groups.
- Thematic analysis involves the careful reading through each individual transcript and the creation of an interpretative description of what is being said.
- The transcripts are then examined as a set, to determine what is being said by the participants as a group.
- Thus, thematic analysis involves searching across a data set, such as a number of interview transcripts or a number of focus group transcripts, to find repeating patterns of meaning.
- These patterns are then described and interpreted and presented as themes in a report on the study, often accompanied by illustrative quotations taken from the transcripts.
- In contrast to the deductive procedure of content analysis, thematic analysis is inductive, building up concepts and theories from the data.
- Tange of unobtrusive methods that do not involve direct contact with the informants that may be used in qualitative research
Some make use of data that has been publish e.g. libraries, press, media etc.
20
Q
Methodological rigour
A
- Trustworthiness of qualitative research
Similar but different to validity and reliability (quant)
21
Q
Processes of assuring methodological rigor
A
- Audit trail: keeping detailed records of processes involved in each step in conducting the study. Can be followed by others as part of the process of making an assessment of trustworthiness
- Triangulation: use of multiple methods, researchers, data sources or theories in a research project. Recognizes the value of different methods of data collection and/or different methods of data analysis, in teasing out authentic answers to complex questions about health related issues
Ethical considerations: Pertaining to risk, benefit and consent
- Triangulation: use of multiple methods, researchers, data sources or theories in a research project. Recognizes the value of different methods of data collection and/or different methods of data analysis, in teasing out authentic answers to complex questions about health related issues
22
Q
Qualitative research and EBP
A
- Allows for in-depth understandings of peoples illnesses
EBP requires AHP to consider various factors when making decisions about care and treatment: FAME framework
23
Q
What is FAME
A
- Feasibility (the extent to which an activity or intervention is practical or viable in a context or situation – including cost-effectiveness).
- Appropriateness (the extent to which an intervention or activity fits with a context or situation).
- Meaningfulness (refers to how an intervention or activity is experienced by an individual or group and the meanings they ascribe to that experience).
- Effectiveness (the extent to which an intervention achieves the intended result or outcome).
- Can lead to information that informs decision-making
Draws on evidence from qualitative studies that lead to the development of health promotion programs
24
Q
What is NHMRC
A
- Expanded evidence hierarchy
- Provides guides to levels of evidence for a greater range
- Focused on quantitative research
1. Qualitative or mixed-methods systematic review
2. Qualitative or mixed-methods synthesis
3. Single qualitative study
4. Systematic review of expert opinion
5. Expert opinion
Qualitative evidence syntheses are used to inform the development of clinical guidelines for national organizations such as the UK National Institute for Health and Clinical Excellence
25
Systematic reviews of qualitative research
- Methods for systematic reviews of qualitative are not as well developed as quantitative
- There is a Cochrane Qualitative and Implementation Methods Group and considerable interest in how qualitative research can best be evaluated and synthesised. Standards for Reporting Qualitative Research (SRQR) and the Enhancing Transparency in Reporting the Synthesis of Qualitative Research (ENTREQ) Statement are available to authors and journals, and this should improve the quality of reporting individual studies over time.
26
What is qualitative coding
- Making sure data is valid
- Transparency
- Deductive coding: pre-established codes and apply to a data set, pre-existing data or research. Then code data set only using those codes
- Inductive: Create codes based on the data collected. No predetermined sets, developed as you review data e.g. conversation you wouldn't know which way It will flow
- You can combine deductive and inductive: hybrid approach
Hybrid approach: Start with a set of priori codes (deductive), add new codes based on the data (inductive)
27
Stages of coding
Initial coding: general overview of data by reading and understanding it
-
Line-by-line coding: delve into the data and organize it into a formal set of codes
28
What is qualitative coding?
- The process of labelling and organising qualitative data (e.g. interview transcripts) to identify patterns, themes and categories
- Coding helps researchers analyse data systematically and transparently
29
5 Common coding methods
In Vivo coding:
- Uses participants exact words as codes
- Useful when working with unique language, jargon or culturally specific expressions
e.g. if a patient says 'I feel like ill never be independent again' the code might be never be independent
30
Process coding
- Focuses on actions or processes in the data
- Codes often use verbs ending in -ing (struggling, adjusting, relearning)
e.g. adjusting to physical limitations, struggling with motivation
31
Descriptive coding
- Summarises a passage of text using a single word or short phrase
- Helps with organizing large datasets
- e.g. a paragraph discussing how a stroke patients family supports them could be coded as family encouragement
32
Structural coding
- Labels sections of text based on their question type or purpose
- Often used for open-ended survey questions
e.g. if an interview asks 'what challenges do patients face in recovery' responses might be labelled under 'challenges in rehabilitation
33
Values coding
- Captures participants beliefs, values and attitudes
- Looks for words like, I believe, I feel, its important that
e.g. need for independence, fear of dependency, importance of small victories
34
Step 1 for initial coding
Objective: Identify key ideas and broad categories in the text.
· Read through the data (e.g., interview transcript) once to familiarise yourself.
· Identify major topics or recurring concepts.
· Assign broad, descriptive codes to sections of text (e.g., “patient emotions,” “family support,” “recovery progress”).
Keep codes flexible; they can be refined later.
35
Step 2 for line by line coding (detailed analysis)
Objective: Break down text into smaller segments and refine codes.
· Read the transcript line by line and assign specific codes to each segment.
· If a broad code from initial coding needs more detail, create sub-codes (e.g., “patient emotions” → “frustration,” “motivation,” “fear of failure”).
· Capture nuanced meanings rather than just broad topics.
Codes should be consistent across different sections of data.
36
Step 3 of line by line coding: Categorization of codes
Objective: Organise similar codes into broader categories for easier analysis.
· Review your codes and group them into themes or categories.
· Identify patterns in the data (e.g., multiple participants mentioning the same concern).
Adjust or refine codes as needed.
37
Step 4 line- by line coding
Objective: Develop overarching themes from coded data.
· Themes are higher-level concepts that describe broader findings.
· Example: Codes like "frustration," "motivation," and "patient progress" may fall under the theme "Emotional Responses to Rehabilitation".
Compare themes across different participants to identify commonalities and differences.
38
Tips for effective coding
✅ Keep a codebook – Maintain a document with code definitions to ensure consistency. ✅ Be flexible – Initial codes may change as new insights emerge. ✅ Code everything first – Don’t rush to themes too quickly; let patterns emerge. ✅ Revisit codes – Adjust and refine codes as more data is analysed. ✅ Use multiple coding methods – Combining approaches (e.g., in vivo + process coding) provides deeper insights.
39
Reflection questions for analysis
· What patterns or themes emerged from the coding process?
· How did different coding methods change the interpretation of data?
· Were there any surprising insights in the data?
How do these findings relate to real-world allied health practice?
40
The delphi technique
The Delphi technique is a structured communication method that utilizes expert opinions to generate consensus or forecast future trends. It involves a panel of experts who anonymously provide feedback in multiple rounds, allowing for the collective knowledge to guide the process. This iterative process helps to refine opinions and arrive at a more informed group consensus
