week 2 Flashcards
(53 cards)
1
Q
define popualtion
A
when all the members of a group are surveyed (e.g., all OTs in Victoria; all parents of 5 year old children in Frankston; all patients admitted to a rural hospital ER in 2017).
2
Q
define sample
A
subgroup / subsample of a population who are surveyed; is the part of the population that one uses to administer a survey & collect information.
3
Q
define concept
A
an abstraction formed by generalisations from similar phenomena or similar attributes.
4
Q
define construct
A
is a concept that is systematically defined to be used in scientific theory
5
Q
types of survey questions
A
1) Those that ask about behaviors or facts
- Non-threatening behavior questions
- Threatening behavior questions
- Demographics
2) Those that ask about psychological states or attitudes
3) Those that ask about knowledge
6
Q
What is a good question?: Summary Points
A
- One that yields a truthful, accurate answer
- One that asks for one answer on one dimension
- One that accommodates all possible contingencies of response
- One that uses specific, simple language
- One that has mutually exclusive response options
- One that produces variability in response
- One that minimises social desirability
- One that is pre-tested
7
Q
steps that a researcher might consider when dealing with highly sensiittvie survey material
A
- Minimise a sense of a judgment.
- Maximise the importance of accuracy.
- Use self-administered data collection procedures.
- Ensure confidentiality & anonymity
8
Q
Five types of closed-ended questions
A
1) Likert-scale
2) Multiple-choice
3) Ordinal
4) Categorical
5) Numerical
9
Q
Advantages of open ended questions:
A
- Allows researcher to obtain answers that were not anticipated.
- May describe more closely the real views of respondents.
- Respondents like the opportunity to answer questions in their own words.
- Are appropriate when the list of possible answers is longer than is feasible to present to respondents.
10
Q
Advantages of closed questions:
A
- Respondents can more reliably answer questions when the response alternatives are provided.
- Researcher can more reliably interpret the meaning of answers when they are provided.
- Avoids the context where respondents will provide rare answers that are not analytically useful.
- Ensures a conformity and consistency of respondents’ answers to questions.
11
Q
3 types of measures
A
- Nominal: people / events are sorted into unordered categories
- Ordinal: people / events are ordered or placed in ordered categories along a single dimension.
- Interval data: numbers are attached that provide meaningful information about the distance between ordered stimuli or classes….interval data are rare in survey research.
12
Q
survey errors
A
sampling error
coverage error
measurement error
non response error
13
Q
whats sampling error
A
- The results of surveying only some, and not all, elements or parts of the population
- The extent to which the precision of the sample survey estimates are limited by the number of persons surveyed
14
Q
whats coverage error
A
- The result of not allowing all members of the target survey population to have an equal or known nonzero chance of being sampled for participation in the survey
- When the ‘list’ from which the sample is drawn does not include all elements of the population under consideration
15
Q
whats measurement error
A
- Respondents’ answers to the survey questions are inaccurate, imprecise, or cannot be compared in any useful way to other respondents’ answers
- Results from poor wording or phrasing of the survey questions or the overall construction of the questionnaire
16
Q
whats non response error
A
- The result of people who respond to a survey being different from sampled individuals who did not respond, in a way that is relevant to the study; non-responders to a survey being fundamentally different that the participants who did complete a survey
- When a significant number of people in the same do not respond AND have different characteristics from those who do respond
17
Q
multi method research vs mixed methods
A
Multimethod research designs:
- Single research approach e.g. qualitative or quantitative
- Two or more data collection methods in single study – Qualitative study – known as data triangulation e.g. interview, observations – Quantitative study – e.g. survey with study participant and a survey completed by carer (proxy report)
VS
Mixed methods research design
- Quantitative & qualitative approach used in single study
- Concurrent or Sequential design
18
Q
mixed methods benifits
A
- broadens the scope of study
- answer complex research questions
- brings together qual and quant
can take multiple approaches
19
Q
mixed methods challenges
A
- study takes longer in all phase design
- researcher requires expertise qual and quant research paradigms
20
Q
3 different approaches to mixed methods research
A
convergent
explanatory
exploratory
21
Q
whats convergent parallel mixed methods design
A
- Qual & quant research traditions equal
- Provides comprehensive analysis of research problem
- Both qual & quant data collected at same time
- Interpretation of results integrated
- May highlight contradictions in qualitative & quantitative findings
22
Q
whats explanatory sequential mixed methods design
A
- Quantitative 1st to explain relationships follow up with qualitative and iterpretation
23
Q
whats exploratory sequential mixed methods design
A
- Qualitative 1st to explore peoples experiences builds to quantitative to interpretation
24
Q
reliability
A
Refers to how well the data collected by using a questionnaire can be reproduced. Types: test-retest, internal consistency, intra-rater reliability & inter-rater reliability.
25
validity
Refers to how well the items of a questionnaire measure what they intend to measure.
Types: face, content, criterion-related, concurrent, predictive, construct, convergent/divergent, factorial, discriminant, diagnostic.
26
weaknesses of closed questions
- Cannot capture in-depth or spontaneous responses.
- Can sometimes bias answers by forcing the respondent to choose between alternative responses.
- Limits chance to consider options that they had not previously considered.
Example: Why do students not buy prescribed texts for the courses they are enrolled in?
1. Too expensive
2. Not relevant
3. Too heavy to carry
4. Colour of cover is not attractive
5. Required readings are not relevant
27
weaknesses of open ended questions
- Requires more effort from the respondent.
- Possibility of interviewer variability.
- Are more time-consuming to complete hence may limit the number of questions that can be asked.
- Answers are more difficult to score and analyse which makes comparisons challenging to complete.
28
Factors to consider when designing survey questions
Memory: avoid over-taxing the respondent’s memory.
Motivation: ask questions that are relevant to the respondent.
Communication: ensure that what we are asking is clear & understandable.
Knowledge: only ask for information the respondent is likely to have or be familiar with
29
types of response formats
- One response pick list
- Multiple response pick list
- Narrative comments
- Short answer
- Dichotomous answers: yes/no; true/ false
- Rank ordering
- Matrix
30
suggestions for survey item success
- Ensure response categories for closed questions are mutually exclusive / do not overlap.
- All potential options / categories are exhaustive and if necessary include the option “other, please indicate/specify”.
- Be aware that response options can send a message about the type and range of ideas, in addition to concepts the respondent should be thinking about.
- An open space should be provided in self-completion questionnaires for free comment which can improve response rates.
- Use simple language.
- Avoid use of technical terms, professional jargon, & abbreviations.
- Avoid words that may have more than one meaning.
- Avoid questions that are not sufficiently specific. Example: “Does your astrological sign impact on the amount of rain that falls on your garden?”
31
things to avoid in surveys
- Ambiguity. (can affect comparability)
- Double-barrel questions.
- Double negatives.
- Leading questions.
Use of loaded questions and concepts.
- Use of presuming questions. (assumes people already have knowledge of a topic)
32
print details for survey
- Font sizes less than 10 should be avoided.
| - Overuse of Upper Case Letters, italics & multiple types of Font should be avoided.
33
whats pretesting/piloting
- Ask 2-3 colleagues/peers to review your questionnaire critically
- Select people who are similar to your target participant group
- Obtain feedback about the form and content of the questionnaire
- Assess whether the questions generate the information needed
- Revise the questionnaire based on the feedback you receive
34
what does the national statement on ethical conduct in human research focus on and whats it used to
focus:
- Promoting ethical human research
- Respect & protection of participants
- Research that benefits the community
used to:
- Inform design of human research studies
- Guide ethical review of research by human research ethics committees, &
- Conduct of human research A
35
when was the Australia’s National statement on ethical conduct in human research made
National statement (2007, updated 2015) developed by:
- National Health and Medical Research Council (NHMRC)
- Australian Research Council (ARC)
- Australian Vice-Chancellor’s Committee
36
whats the four principles of biomedical ethics
respecting autonomy
beneficience
non maleficence
justice
37
whats respecting autonomy
- Participants right to make informed choice to partake or not, or partake & withdraw from a study
- Participant right to confidentiality
38
whats beneficence
- Benefit of research for participant or others must Beneficence outweigh any potential risks
39
whats non malefienece
- Research must identify & manage any risks (harm, Non discomfort or inconvenience) to participants or others
40
Values underpinning ethical conduct of research:
```
research merit
research integrity
justice
benifiencence
repect
```
41
whats research merit
- Research has benefit e.g. contribution to knowledge or to improve social welfare, individual wellbeing or researcher skills & expertise
- Research design & methods suitable to answer research aim
- Based on thorough study of current literature
- Respect for participants upheld in research design, implementation & dissemination
- Conducted or supervised by people with necessary skills, experience, &/or qualifications
- Conducted using research appropriate resources
42
whats research integrity
- researchers must be committed to:
- “Searching for knowledge and understanding
- Following recognized principles of research conduct e.g. research designs
- Conducting research honestly”
- Disseminate and communicate results (whether favourable or not) to contribute to community knowledge & allow scrutiny
43
potential risks in research
Evaluating risk:
- likelihood, severity & consequences of risk to participants & non-participants
- Strategies to manage risk
Levels of risk:
- harm, discomfort (low risk) or inconvenience (negligible risk)
Risk can result from:
- Research process
- Data collection & storage
- Dissemination of findings e.g. publication, conferences
44
potential befits of research
To knowledge
- “Gains in knowledge, insight or understanding” (p 14, NHMRC, 2007/2015) of research topic
- Increase skills
To participants
- Opportunity to reflect on & share knowledge of research topic based on own experience
- Sense of contributing or giving back to society
- Benefit from new technologies or interventions
To broader community
- Increase public knowledge of issues
- Improve community knowledge & resources e.g. health professionals, government
- Increase ability to address issues e.g. management of health conditions, inequality or disadvantage
45
three forms of research misconduct in relation to reporting results
- Fabrication – “making up data or results and recording or reporting them”
- Falsification – misrepresentation of data due to: – manipulation of “research materials, equipment or processes” or – “Data or results being omitted”
- Plagiarism – misuse “of another person’s ideas, processes, results, or words without giving appropriate credit”
46
vulnerable populations
Women who are pregnant and the human foetus
- Children and young people (under 18 years of age)
Vulnerable populations also include people who are:
- In a dependent or unequal relationships
- Dependent on medical care who may be unable to give consent
- Involved in illegal activities
- Have a cognitive impairment, intellectual disability or mental illness
- Aboriginal and Torres Strait Islander peoples and
- From other countries or culturally and linguistically diverse background
47
citation mining
or sometimes referred to as 'snow balling' or 'pearl growing') is a literature searching technique where references from key articles are used to add to the evidence base.
48
whats PIO stand for
P: Patient/Population/Problem
I: Interest (event, activity, experience or process)
Co: Context (setting or characteristics)
49
whats grey literature
* Information that is not traditionally published, for example conference papers, government reports and statistics. It is not peer reviewed but provided it comes from reliable sources, it is a valuable and often very current source of information. See the grey literature library guide for more information and grey literature sources.
* Use known reputable sites such as Australian Bureau of Statistics or the Australian Institute of Health and Welfare.
50
concurrent design
: Qualitative and quantitative data sets are collected at the same time and their results compared.
51
sequential design
One data set follows another and extends or explores the ndings from the rst set
52
q2 or q squared
: The combination of quantitative and qualitative approaches in a research study. It is the concept used in mixed methods research design.
53
Respondents should have the following information before being asked to answer questions:
- The name of the organisation carrying out the research
- He sponsorship
- Brief description of purpose of research
- Statement on how answers will be protected with confidentiality
- Assurance that cooperation is voluntary
- Assurance that respondents can skip any questions
