Week 3 Flashcards
What is Sample Study?
Trying to estimate the value of a parameter for a population
ex) How long does each individual spend on the computer per day
What is Observational Study?
Trying two understand how 2 parameters relate to one another. You cant make a conclusion on a causality because of a lurking variable
ex) How does blood pressure relate to computer time
What is Experiment Study?
The basis of the scientific method and main purpose is to determine causality. It shows causality because you use randomization
What is Bias?
Error in methodology that results in a way systematic deviation of the result away from the truth and being pulled in one direction
What is Random Error?
The results are being pulled in different direction
What are Confounding variables?
is an outside influence that changes the effect of a dependent and independent variable
What is population?
group of people, objects or events
What is Target Population?
The larger population to which results of a study will be generalized too
What is Accessible Population?
The actual population of subjects available to be chosen for a study
What is Sample?
A subgroup of the population of interest (accessible population) which allows the results of research to be generalized to the population
What is Sampling bias?
Extent that systematically misrepresentation population can be conscious or unconscious
What is Sampling Error?
Extent that sample randomly misrepresents population
What are some Sampling techniques?
- Probability Sampling
- Non probability Sampling
What are some Probability Sampling?
- Simple Random Sampling
- Systematic Sampling (every 10th person)
- Stratified Random Sampling (specify # from each category)
- Cluster Sampling (multilayer/stage)
What are some Non Probability Sampling?
- Convenience Sampling (subjects are chosen on basis of availability
- Constructive Sampling ( Recruit all folks to meet inclusion/exclusion criteria)
- Volunteers
- Quota Sampling (like stratified, but not random
_Purposive Sampling (subjects hand picked by specific criteria)
What are some considerations with Non Probability Sampling?
- must be more careful with inferences and generalizations
2. Closely examine inclusion/exclusion criteria and baseline characteristics
What are some examples of Experimental Research?
- Randomized Trial
- Clinical Trial
- Randomized Clinical Trial
- Randomized Controlled Trial
What are extraneous variable?
Any factor not related to the purpose of the study, which may affect dependent variables. Also known as nuisance variables
What happens when extraneous variables become uncontrolled?
They become confounding variables
What is the purpose of Experimental Research?
they are designed to control confounding variable
What are the 3 essential components of Experimental Research?
- Include a control/comparison group
- Independent variables are being manipulated by the experimenter
- Subjects are randomly assigned to groups
What is a random assignment?
Each subject has an equal chance of being assigned to any group and help control for extraneous variables or prognostic indicators
What should Random Assignments do?
- Should results in homogenous groups at baseline except for the IV
- Concealing allocation: Separate people
- Sealed envelopes
What are ways to incorporate Missing Data in a study?
- Per Protocol Analysis: Analyze only those who complete the study
- Intention to Treat Analysis: Analyze in the group that they were assigned too and use Imputation (estimating missing data values)
- As treated (Observe data regardless of what happens and then move them)
What are some ways to control inter subject difference (confounders)
- Homogenous (only males, ages 30-40)
- build into study as IV (male/female)
- Matching (Age and Sex)
- Using subjects as their own control (within subject design)
What are some Experimental Design Validity?
- External Validity
- Construct
- Internal
- Statistical Conclusion
What are some threats to external Validity?
- Selection (Sampling)
- Setting (Military vs Civilian)
- History (Conclusions made 30 yrs ago might not be valid today)
What is a True Experimental Design?
- Gold Standard
- IV manipulated by researcher
- At least 2 comparison groups
- Subjects randomly assigned
What is Quasi experimental design?
- May lack randomization
2. May lack comparison groups
What is Between Subject Design?
Different people in each level of the IV
What is a Single Factor Design?
Used to compare 2 or more groups formed by random assignment and have a pretest and postest
What are the advantages of a single factor design?
- Less laborious and costly to preform
2. Appropriate when a pretest isnt practical or has the potential to sew the study results
What are the disadvantages of a single factor design?
You dont know if the groups were equivalent at the baseline
What is Multifactoral design?
Has 2 or more Independent variables
What is the randomized blocked design?
Same as multifactoral except 1 Independent variable is not randomized (3 or more way design)
What is within subject design?
Designed for repeated measures
What are some examples of Order Effects?
- Randomize Order
- Counter Balance (AB, BA, AB, BA)
- Crossover design( used when there are 2 levels of IV and the patients condition is stable over time and are randomized into 2 groups)
- Multifactor (all IVs are repeated or within subject)
What is Mixed Design?
- At least 1 IV between subject (independent factor)
- At least 1 IV within subject (repeated measure)
What is Efficacy?
The benefit of a treatment delivered in a under highly controlled and ideal environment
What is Effectiveness?
The benefit of a treatment delivered in a pragmatic manner under real world conditions
What are the Appraising therapy studies steps?
- Are the results valid
- What are the results?
- Will the results help care for my patients?
What is the PEDro Scale?
is a valid measure of the methodological quality of clinical trials: a demographic study.