1
Q
What are the 3 different kinds of decision making and give a brief definition on them?
A
- Anecdote or expert opinion: Assumption that what has worked in the past will continue to work.
- Press-cutting: Continually altering practice based on news clippings, internet, magazines etc.
- Cost-minimisation: Cost to patient vs. cost to developer.
2
Q
Define research.
A
The systematic and rigorous process of enquiry which aims to describe phenomena and to develop and test explanatory concepts and theories.
3
Q
Why do we do research?
A
To produce evidence to answer a question (not opinions - but data)
4
Q
What isn’t classified as research?
A
- Gathering facts or information
- Moving facts from one situation to another (regurgitation of another’s work with no interpretation)
- Work unrelated to real world application
- A marketing buzz word
5
Q
What is evidence?
A
Data used to support (or counter) a scientific theory or hypothesis.
6
Q
True or False: Research questions must be falsifiable?
A
True - All research questions must have the ability to be show to be false.
7
Q
What P value indicates a replicable result?
A
<0.05
8
Q
What is the difference between a primary and secondary source?
A
Primary Research: is information gathered through self-conducted research (eg. Experimentation or laboratory)
Secondary Research: is information gathered from previously conducted studies (eg. Lit reviews)
9
Q
What are different types of scholarly work?
A
- Research study (primary)
- Literature review (secondary)
- Theoretical discussion
- Opinion paper
- Program description
- Fiction writing
- Poetry
10
Q
What are the levels of evidence?
A
- Systematic reviews and meta analyses
- Peer-reviewed articles
- Government reports
- Books
- Blogs/websites
- Wikipedia
11
Q
What is the purpose of a literature review?
A
- summarise the results of previous research to form a foundation on which to build your own research.
- to collect ideas on how to gather data.
- to investigate methods of data analysis.
- to study instrumental & procedures that have been used.
- To assess the success/appropriateness of the various research designs of the studies already undertaken.
12
Q
What is the structure of an article?
A
- Front matter: Title, authors, journal, abstract.
- Introduction
- Methods and Methodology
- Results
- Discussion: Conclusion
- Back matter: References, acknowledgement, funding and conflicts of interest.
13
Q
Explain the PICO formula for developing a research question and give an example.
A
P: Patient, population or problem: What are the most important characteristics of the patient and their health status?
I: Intervention/Exposure: What main intervention are you considering (medical, surgical, preventative)?
C: Comparison: What are the alternative benchmark or gold standards being considered if any?
O: Outcome: What is the estimated likelihood of a clinical outcome attributable to a specific disease, condition or injury?
14
Q
What are some indication of a good research question?
A
- open ended and consider cause/effect.
- are researchable - can be answered with accessible research, facts and data.
- open the door for other areas of research and inquiries.
- often begin with how, why, what or which
- in the form of a question and therefore should end in a question mark (?).
15
Q
What are some indications of a poor research question?
A
- have the capacity to be answered in a simple format such as one word response, a number or a list.
- cannot be answered
- can only be answered with an opinion
- often begins with who, when, where, how much, how many.
16
Q
How does hypotheses development occur?
A
- establish a research question
- develop a theory
- identify variables
- identify hypotheses
- test hypothesis
17
Q
What is a hypothesis?
A
An assumption or proposed explanation for a phenomenon designed for arguments sake to prompt research.
18
Q
Define a null hypothesis?
A
The hypothesis that predicts no effect or relationship between variables.
19
Q
Define the alternate hypothesis?
A
The hypothesis that predicts statistical significance between two measured phenomenon.
20
Q
What factors must be included in a hypothesis?
A
- it must be falsifiable
- not moral or ethical questions
- a prediction of consequences (cause and effect)
- is considered valuable even if shown to be false
21
Q
Which of the following ends in a question mark? A research question or a hypothesis?
A
A research question as a hypothesis is a statement.
22
Q
Define qualitative research
A
Research that involves collecting and analysing non-numerical data to explore concepts, opinions and/or experiences. (Personal opinions)
23
Q
What are examples of qualitative data collection methods?
A
- Observation: recording what you have seen, heard, or encounter.
- Interviews: Personally asking questions in a 1-on-1 conversation.
- Focus Groups: Asking questions and generating discussion within a group of people.
- Surveys: Distributing questionnaires with open-ended questions
24
Q
True or false: Typically qualitative research is generated from a smaller sample size than quantitative?
A
True
25
Define quantitative research.
Research that involves collecting and analysing numerical data and statistics to systemically measure variables and test hypotheses.
26
Give 3 examples of qualitative research questions.
- How does social media impact the body images of teenagers?
- How do children and adults interpret healthy eating?
- How do Australian adults interpret the mental health crisis within teenagers?
27
What are examples of quantitive data collection methods?
- Questionnaires: Must be able to be quantified into numerical values.
- Laboratory experiments
- Observational checklists
- Field trials
- Longitudinal studies
28
With regards to quantitative and qualitative research which type of research is INDUCTIVE and which is DEDUCTIVE?
Qualitative is INDUCTIVE: Explores, generates ideas from data.
Quantitative is DEDUCTIVE: Begins with a hypothesis, draws conclusions based on data.
29
Define objective measurement and give an example.
Measurement based on factual data with no personal interpretation. Independently verifiable.
E.g. Blood pressure reading, HR, RR, amount of time it takes to complete a given task, distance between town points.
30
Define subjective measurements and give an example.
Measurements based on a person’s interpretation, belief, feeling or experience.
E.g. How much pain are you in? What symptoms are you experiencing? What are your concerns?
31
Define social desirability bias and give an example.
A type of response bias where the participants answer in a manner that will be viewed favourably by others.
E.g. Self-reported weight, pain scale, admitting to mental health issues.
32
Define social desirability bias and give 3 examples.
When a participant responses to a question in a manner that will be viewed favourably by others.
E.g. Reporting a lower body weight, expressing experiencing less pain then they are to seem tougher, down playing mental health issues/symptoms.
33
What does the term “saturation” mean in qualitative research?
When multiple interviews/focus groups have been completed and consistently answer the same resulting in no new information or data gathered.
34
Needing to explore individuals stories to describe the lives of people is an an example of what qualitative research design?
Narrative research design
35
Needing to explore what happens, to describe and interpret a case is an an example of what qualitative research design?
Case study research design
36
Needing to explore a group of people to describe their shared culture is an an example of what qualitative research design?
Ethnographic research deign
37
Needing to explore the different experiences of individuals to develop a theory is an an example of what qualitative research design?
Grounded theory research design
38
What is congruence in regard to qualitative data analysis?
Looking for consistencies or themes repeated by different participants.
39
When describing qualitative research would you use the terms “validity and reliability” or “trustworthiness and authenticity” ?
Trustworthiness and authenticity
40
Define “triangulation”
Evidence of the same conclusion from multiple sources of data collection, interviewers etc.
41
What is member checking?
Showing interviewees your summary to ensure it is an accurate reflection.
42
Define “bracketing”.
Acknowledging your own predispositions and biases and seek to control them.
43
What is the benefit of prolonged engagement within qualitative research?
The more time spent with a person the high the likelihood that you will get to know them and interpret the data collected from an individual accurately.
44
What is meant by “interviewer corroboration”?
Independently coding transcripts to ensure consistent interpretations. (Having a second review).
45
Define experimental research.
Where we manipulate something to see the cause of the manipulation. (Intervention studies).
46
Define non-experimental research.
Non intervening matters of data collection. Observation, description, surveys. (Observation studies).
47
What are the types of validity?
- Face validity: face values.
- Content validity: ensuring the measurements gathered capture the whole construct.
- Construct validity: How well are we measuring the individual construct and not a different construct.
- Criterion or predictive validity: the extent to which the measurement at this point and time predicts what we want it to in the future.
48
What is the difference between validity and reliability?
Validity refers to the accuracy of a measure.
Reliability refers to the consistency of a measure.
49
What are the two types of reliability?
Internal consistency: The extent to which multiple items in a scale all relate to each other (questionnaires that have multiple questions essentially asking the same thing).
Test re-test: Multiple tests to ensure the same measurements (stepping on and off the same set of scales and receiving the same numerical weight each time)
50
What is the difference between cross sectional and longitudinal research?
Cross sectional is observational/non experimental studies that involve the analysis of data collection from a population at one pacific point in time.
Longitudinal studies can be experimental or non-experimental and track participants across several time points.
51
What is the difference between “correlation” and “causation”?
Correlation is the statistical measure that describes the size and direct relation between two or more variables.
Causation indicates that one event is the result of the occurrence of the other event.
52
Does longitudinal or cross-sectional research provide answers regarding correlation and causation?
Longitudinal
53
Longitudinal studies can be both prospective and retrospective, define these two types of studies.
Prospective: Study design and measures carefully chosen to test specific hypothesis (predictive).
Retrospective: Analysis of previously collected data (not planned).
54
Define sampling.
The activity of determining how many will participate in the study and identifying who those individuals will be.
55
What does it mean to have a representative sample?
When researchers select individuals that are typical of the wider population in key aspects, gender, age, SES.
56
Define generalisability.
The extension of research findings and conclusions from a sample group to the wider population.
57
Define exhaustive sampling.
Inclusion of every single individual in the target population.
E.g. having every single student from a unit provide feedback.
58
What is the difference between response rate and completion rate?
Response rate is the number of people who part took in the survey in comparison to the number of the total sample population.
Completion rate is the number of those who actually completed the entirety of the survey.
59
What are the 5 types of probability sampling?
- Random sampling
- Systematic random sampling
- Stratified random sampling
- Cluster sampling
- Multistage sampling
60
Is randomly selected individuals from a list of all members of the interested population a method of probability or non-probability sampling, and if so what type?
Probability: Random sampling
61
Is randomly picking a starting point within a list of all members then selecting every nth individual on the list a method of probability or non-probability sampling, and if so what type?
Probability: Systematic random sampling
62
Is dividing the population into subgroups and randomly selecting participants from each group a method of probability or non-probability sampling, and if so what type?
Probability: Stratified random sampling
63
Is deciding the population into clusters, randomly selecting a cluster and including all people from that cluster a method of probability or non-probability sampling, and if so what type?
Probability: Cluster sampling
64
Is research carried out in various stages, incorporating multiple versions of other sampling methods over time a method of probability or non-probability sampling, and if so what type?
Probability: Multistage sampling
65
What is one easy way to determine if probability or non-probability sampling methods have been used?
Probability sampling methods require a pre-existing list of the entire population of the study.
66
Is selecting participants who are available and accessible a method of probability or non-probability sampling, and if so what type?
Non-probability: Convenience sampling
67
Is selecting participants “on purpose” because they are considered to be most appropriate for the study a method of probability or non-probability sampling, and if so what type?
Non-probability: Purposive sampling
68
Is asking pre-existing participants (or others that meet inclusion criteria) to recommend others they know who may also meet the study inclusion criteria a method of probability or non-probability sampling, and if so what type?
Non-probability: Snowball sampling
69
What is the purpose of quantitative research?
To generalise to the greater population and therefore an increased sample size will increase the generalisability.
70
Who is the key governing body for health and medical research in Australia and what role do they play?
National Health and Medical Research Council (NHMRC).
Create ethics guidelines, fund majority of research.
71
What are the two key questions that need considering when determining the level of NHMRC hierarchy evidence falls under?
1. Is there a comparable control group or is it a single group being studied?
2. Is there any evidence of randomisation of participants into comparison vs control group?
72
What are the levels of NHMRC hierarchy of evidence?
I: Systematic reviews of level II studies.
II: Randomised controlled trial (RCT).
III-1: Comparative pseudo-randomised studies.
III-2: Comparative non-randomised studies.
III-3: Comparative historical groupings studies.
IV: Single group/case studies.
73
What is attrition in regard to longitudinal studies?
The drop out rate during intervention, potential for only the good students to complete the study - causes loss of generalisability.
74
What is the Hawthorne effect?
A type of reactivity in which individuals modify or improve an aspect of their behaviour in response to their awareness of being observed.
75
Differences in how data is measured between groups is known as?
Detection bias
76
What is allocation bias and how can this be accounted for?
When baseline groups present differences. Randomisation into groups should account for this.
77
When other confounding factors contribute to an individuals performance at the end of an experiment this is know as?
Performance bias
78
When the group characteristics are no longer equal at the end of a study this is known as?
Attrition bias
79
Measurement bias is caused by?
Outcomes being poorly or inaccurately measured.
80
What are the three elements considered when assessing the GRADE of research?
- Quality of evidence
- Balance between harms and benefits
- Are incremental health benefits worth the cost?
81
With regard to the assessment of evidence quality within the GRADE evaluation what attributes are considered?
Study design: As per NHMRC hierarchy
Study quality: imprecise or sparse data, forms of bias, strength of association of findings.
Direct applicability: dependent on the extent to which the studies participants/interventions/outcomes are similar to those of interest.
82
Define the difference between a strong and weak recommendation.
Strong: We can be reasonably sure that a treatment’s benefit will exceed its costs and associated risks. Given an informed choice, most patients would choose this treatment.
Weak, conditional or discretionary: There is a fine balance between cost and benefit of the treatment and the treatment effects cannot be certain. Given an informed choice, patients may prefer other treatment options.
83
What is the difference between categorical and continuous variables? Give 3 examples of both.
Categorical: work to group data into categories. E.g. sex (male, female, other), age groups (18-30, 31-45, 46-65), university (ECU, Curtin, UWA)
Continuous: work on a continuum. Age (0-100), time taken to complete a task, distance between two points.
84
What is range in statistics?
The difference between the highest and lowest scores obtained for a variable.
85
Define standard deviation.
The measure of how dispersed the data is around the mean value for a continuous variable.
If the individual scores are grouped heavily around the mean, the set standard of deviation is going to be low and therefore increase the confidence in the reflection of the mean in reality.
86
Explain the difference between independent and dependant variables.
Independent variable stands alone and isn’t changed by the other variables you are trying to measure.
Dependant variable value depends on that of another.
87
What p value is typically considered statistically significant?
<0.05
88
What are the three types of T tests? And what are they for?
1. One-sample T test: Comparison of a mean value to a pre-existing value.
2. Independent T test: Comparison between a categorical and continuous variable.
3. Paired-samples T test: Measurements taken across multiple time points within the same group.
89
How does the One Way ANOVA differ from the Independent t-tests?
Has the capacity to compare across 3 or more groups.
90
What are the ECU Ethical Principals?
- Research Merit and Integrity: Worthwhile, robust, competent, honesty.
- Justice: Participant recruitment, burden, exploitation.
- Beneficence: Benefits need to outweigh the risks to participants.
- Respect: Privacy, confidentiality, cultural sensitivity.
91
Define “autonomy”.
Participants free to determine their own actions.
92
Define “beneficence”.
Research must act to benefit human kind.
92
Define “justice”
Consideration of pressure to participate or not withdraw.
92
Define “non-maleficence”.
Avoid/minimise harm to participants (physical, social, psychological, economic, legal)
93
Define “human dignity”
Informed consent and impaired consent
94
Define “confidentiality”
Privacy and anonymity of participants
95
Define “veracity”
Not lying to the participants or within the reporting of the results.
96
What is mixed method research?
The combination of quantitative and qualitative research within a single study.
97
How is convergent mixed-method research performed?
Quantitive and qualitative run concurrently but separately.
98
How is sequential mixed-method research performed?
Qualitative conducted first, followed by quantitative (or vice-versa).
99
How is multiphase mixed-method research performed?
Through different designs for different participant cohorts.
100
How is embedded mixed-method research performed?
Quantitative and qualitative data collection is undertaken within the same data collection instrument.
101
What does PICO stand for?
P: Population/participants
- Age
- Sex
- Race
- Primary problem
- Health status
I: Intervention
- Diagnostic tests
- Medication
- Procedure
C: Comparison
- Another test, medication or procedure
- Watchful waiting
O: Outcome
- Accurate diagnosis
- Relieve or improve symptoms
- Maintain function
102
What is a sampling error?
An error in a statistical analysis arising from the unrepresentativeness of the sample taken
103
What is the Chi-Square test?
A test that measures how a model or expectations compare to the actual observed data
104
Explain Pearson's Correlation coefficient (r).
A measure of the strength of a linear association between two variables and is denoted by r.
The scale ranges from -1 to +1
<0 = indicates a negative association (when one variable increases and the other decreases)
0 = indicates no association between variables.
>0 = indicates a positive association (when one variable increases so does the other)
105
Is Alan going insane?
3 apples
Research Methodology
flashcards
Decks in class (1)
# Cards
