1
Q
What are ways to maximize reliability
A
- Standardize measurement methods
- Train and test observers
- Refine and Automate Instruments
- Blind to reduce differential bias across groups
2
Q
What are ways to maximize validity
A
- Random Assignment
- Use of Control Groups
- Blinding (can prevent compensatory reactions and imitation of treatment)
- Selection of Homogeneous Subjects
- Blocking (build extraneous attribute variables into design by using them as independent variables)
- Matching (Age, Sex, Etc)
- Using subjects as their own control
- Analysis of Covariance
3
Q
What are sources of knowledge
A
- Tradition
- Authority
- Trial and Error
- Logical Reasoning
- Scientific Method
4
Q
Define Inductive and Deductive reasoning
A
- Inductive Reasoning - Specific to general
- Deductive Reasoning - General to specific
5
Q
What are the four steps for identifying the research question
A
- From a problem area, identify the research problem
- Justification for the research problem
- Develop the research question
- State hypotheses
6
Q
Why should theories be Economical or “Parsimonious” as well as important
A
- Economical (Parsimonious) – Explain the most with the fewest variables. If two variables explain 75% of the problem, but adding 5 more variables only explains an additional 3%, it would be parsimonious to stick with just the initial two.
- Important - “so what”. Should be significant for those who will use it
7
Q
What are some ways we can ensure protection of human rights
A
- Beneficence: the obligation to attend to the well-being of individuals
- Justice: refers to fairness in the research process; fair selection of subjects to equally distribute the benefits and burdens
- Use of control groups when effective therapeutic methods exist may present problem
8
Q
What are continuous variables
A
- Can take on any value along a continuum within a defined range
- Infinite fractional values can occur
- Accuracy of measurement relies upon precision of measuring tool
9
Q
What are Discrete variables
A
- Described only in whole units
- Example: Heart beat…not measure in half beats
10
Q
What are Dichotomous variables
A
- When a variable can take on only two values….it is described as being dichotomous
- Example: Male/Female
11
Q
What is a Nominal Variable
A
- Objects or people are assigned to categories according to some criterion
- Categories coded by name, number, etc although none of the categories has any quantitative value
- Categories are used purely as labels
- Example: Blood Type, Handedness, Side of Hemiplegic involvement
12
Q
What is an Ordinal Variable
A
- Categories are rank ordered on the basis of an operationally defined characteristic
- Data organized into adjacent categories with a greater than less than relationship, one is greater or less than the other
- May have unequal intervals
- Example: None
13
Q
What is an Interval Variable
A
- Rank order characteristic similar to Ordinal
- Unlike ordinal, demonstrates equal distances or intervals
- Example: Degrees in C or F
14
Q
What is a Ratio Variable
A
- Interval Scale with a true zero
- Therefore, no negative values
- Measurement of zero represents complete absence of property being tested
- Example: ROM, Height, Weight
15
Q
What are some issues with measuring constructs
A
- Constructs are not easily defined and as such are difficult to measure
- Example: Intelligence is not determined based on one measurement of reading ability or memory but is more complex
16
Q
Discuss the importance of reliability in clinical measurement
A
- Extent to which a measurement is consistent and free from error
- Measurements have to be reliable to be valid. No validity without first reliability
- Reliability is fundamental to all aspects of measurement, without it we cannot have confidence in the data we collect or draw conclusions from those data
17
Q
Explain the concept of measurement error
A
- The difference between the true value and the observed value
- Related to the “noise” that gets in the way of our finding the true score
18
Q
Distinguish between random and systematic error
A
- Systematic Errors: Predictable errors of measurement, consistently overestimate or underestimate
- Random Errors: Due to chance and can affect a subjects score in an unpredictable way from trial to trial
19
Q
Identify typical sources of measurement error
A
- Rater - Person taking measurement
- Instrument
- Variability of characteristic being measured
20
Q
Describe the effect of regression toward the mean in repeated measurement
A
- Measurement errors are random and therefore normally distributed and will equal out over time
- Extreme scores on a pretest will move closer or regress toward the group average
21
Q
Discuss how concepts of agreement and correlation relate to reliability
A
- Correlation reflects the degree of association between two sets of data but does not tell you anything about agreement
22
Q
Define and give an example of intra-rater reliability
A
- Scores should match when the same examiner tests the same subjects on two or more occasions
23
Q
Define and give an example of inter-rater reliability
A
- 2 or more examiners test the same subjects for the same characteristic using the same measure, scores should match
24
Q
Define and give an example of test retest reliability
A
- Used to establish that an instrument is capable of measuring a variable with consistency
- A test is administered to the same group of subjects on more than one occasion; scores should match
25
Discuss how generalizability theory influences the interpretation of reliability
- An individual score can be thought of as a sample from a universe of possible scores that might have been obtained under the same testing conditions
- Single measurement becomes the best estimate of a true score under those conditions
- Reliability then interpreted in relation to a set of specific testing conditions
- Error is divided into components (facets)
26
Relate reliability to the concept of minimal detectable difference
- When we measure a difference between two scores, some portion of that change may be due to error and some portion may be real
- Minimal detectable change is used to define the amount of the change in the variable that must be achieved to reflect a true difference
27
What are some ways to improve reliability of a measurement
- Standardize measurement methods
- Train and test observers
- Refine and Automate Instruments
- Blind to reduce differential bias across groups
28
What is the research question
- Asks "What does the study plan to answer?"
- Example: Do runners using a non rearfoot strike pattern (NRFS) have fewer injuries than runners using a rearfoot strike pattern (RFS)?
- NOT Is a NRFS pattern a better way to run? Define “better”.....cant answer that question
29
What is a theoretical framework or rationale and how does it form the framework for a research question
- presents a logical argument that shows how and why the question was developed
- Will support the research question, guide decisions in designing the study, and provide basis for interpreting results
30
What is an operational definition
- Defines a variable according to its unique meaning within a study
- Should be sufficiently detailed so that another researcher could replicate the procedure or condition
- Should differentiate the various levels of the variable
31
What is an independent variable
- What you manipulate or specify, also called factors
| - Will predict or cause a given outcome
32
What is a dependent variable
- What you measure
| - a response or effect that is presumed to vary depending on the dependent variable
33
What are the characteristics of good research hypotheses
- Must be testable and based on a sound rationale
34
What are directional and non directional research hypotheses
- Non directional describe the relationship between variables
- "There is a difference between A and B"
- Directional not only describes relationship but also assigns a direction to that difference
- "A is greater than B"
35
Distinguish between primary and secondary sources of information
- Primary source is a report or document provided directly by the person who authored it
- Secondary source is a description or review of one or more studies presented by someone other than the original author
36
Distinguish between populations and samples
- The larger group to which results are generalized is the population
- Through a process of sampling, a researcher chooses a subgroup of the population used as a reference group for estimating characteristics or drawing conclusions about the population - This group is the sample
37
Define the concepts of sampling bias and sampling error
Sampling Bias - Extent that sample systematically misrepresents population----Can be conscious or unconscious
Sampling Error - Extent that sample randomly misrepresents population
38
Summarize the difference between target and accessible populations
- Target population: Universe of interest; Overall group of people to which the researcher intends to generalize the findings of the study
- Accessible population: Portion of the target population that has a chance of being selected
39
Describe the purpose of inclusion and exclusion criteria in sampling for research studies
- Inclusion Criteria describe the primary traits of the target and accessible populations that will qualify someone as a subject
- Exclusion Criteria indicate the factors that would preclude someone from being a subject
40
Contrast different types of probability and non-probability sampling procedures in terms of generalizability
- Probability sampling is through simple random selection giving every member of a population of an equal opportunity or probability of being selected
- Non probability samples are made by nonrandom methods. This limits the ability to generalize the outcomes beyond the specific sample studied.
41
Discuss issues in recruiting an adequate sample size
- Must consider the effect of sample size on their analytic process, sometimes the restrictions set for inclusion and exclusion criteria must be modified to obtain a large enough sample, and the implications of this process must be considered
- In Qualitative studies, samples may be small, as compared to quantitative studies that will require statistical comparisons
42
What is PICO and what is its importance
- Population
- Intervention
- Comparisons
- Outcomes
- These can be important to emphasize when searching the literature for articles relevant to your clinical question
43
What Three criteria must be satisfied to constitute authorship
- Substantial contribution to conception and design OR acquisition of data OR analysis and interpretation of the data
- Drafting the manuscript OR reviewing critically for important intellectual content
- Final approval of the version to be published
44
Describe Honorary Authorship
- Also known as “gift or guest” authorship
- A listed author who has not contributed at all, or not contributed significantly to authorship criteria
- Inclusion based on belief that presence will add credibility and will assist in journal publication OR
“Gift” to seek favor, provide payback, etc…
45
Describe Coercion Authorship
- A form of honorary authorship not initiated by the primary author
- Senior member of lab or team pressures openly or through subtle coercion*
Either way: IT’S NOT RIGHT!
46
Describe Ghost Authorship
- An individual makes substantial contributions to the work and they are not listed as an author
Happens When:
- Industry does not want to be associated with the work and acknowledge the conflict of interest
- Someone leaves the project early
- Junior members are left out in the cold*
47
What are the 5 Main paper sections of the Consort Statement
- Title and Abstract
- Introduction
- Methods
- Results
- Discussion
48
What is the sub-cateogory in the Introduction section of the Consort Statement
- Background
49
What are the sub-catergories in the Methods section of the Consort Statement
- Trial Design
- Participants
- Interventions
- Outcomes
- Sample Size
- Randomization - Sequence generation
- Randomization - Allocation concealment
- Randomization - Implementation
- Blinding
- Statistical methods
50
Discuss the testing effect and how it can change the reliability of a measure
- When the test itself is responsible for observed change in a measure variable, the change is considered the testing effect
- Such effects can be manifested as systematic error creating consistent changes across all subjects
51
What is measurement validity
- Concerns the extent to which an instrument measures what it is intended to measure
52
What is Face Validity
- Indicates that an instrument appears to measure what it is supposed to and that it is a plausible method for doing so
- No method of judging "how much" of it an instrument has, face validity is an all or none type of validity
53
What is Content Validity
- Indicates that the items that make up an instrument adequately sample the universe of content that defines the variable being measured.
- Most useful with questionnaires and inventories
54
What is criterion related validity
- Indicates that the outcomes of one instrument, the target test, can be used as a substitute measure for an established reference standard criterion test
- Can be tested as concurrent or predictive validity
55
What is Concurrent validity
- Establishes validity when two measures are taken at relatively the same time
- Most often used when the target test is considered more efficient than the gold standard and, therefore, can be used instead of the gold standard
56
What is Predictive validity
- Establishes that the outcome of the target test can be used to predict a future criterion score or outcome
57
What is construct validity
- Establishes the ability of an instrument to measure an abstract construct and the degree to which the instrument reflects the theoretical components of the construct
58
What is convergent validity
- A method of construct validation
- indicates that two measures believed to reflect the same underlying phenomenon will yield similar results or will correlate highly
59
What is Discriminant validity
- A method of construct validation
- indicates that different results, or low correlations, are expected from measures that are believed to assess different characteristics
- Opposite of convergent......you would expect this to NOT correlate highly with a test that measures a different construct from the one you are trying to measure
60
What are the issues interpreting change scores with Nominal, Ordinal, Interval, and Ratio variables
Nominal- Cannot be subtracted and thus cannot demonstrate change scores
Ordinal- The greatest risk of misinference is with ordinal data because distance between intervals is not known and may not be equal
Interval- Presents a problem for evaluating change because although we can determine distance of change, may not know the true amount of change
Ratio- Only way to measure true change because all measures are known quantities
61
What is responsiveness of the instrument
- The instruments ability to detect minimal change over time
62
Explain the concept of Minimally Clinically Important Difference
- The smallest distance in a measured variable that signifies an important rather than trivial difference in the patients condition
- Also defined as the smallest difference a patient would perceive as beneficial and that would result in a change in the management of the patient
Statistics
flashcards
Decks in class (7)
# Cards
