Ch 9 Flashcards
diff types of comparison
compare same group at different times or after an intervention
diff groups at same time
methods to limit influence of confounding factors (5)
randomization, crossover, homogeneity, matching, statistical control
who can be “blind”
participants, researchers, other staff, data collectors
**more blinding = less bias
2 diff time frames
cross-sectional, longitudinal
relative timing (2 types)
retrospective or perspective
3 criteria for establishing causal relationships
temporal (requires cause to precede effect)
relationship (relationship requires an association to be shown between the cause and effect)
confounders (requires it to be proven that the effect cannot be caused by a third variable)
3 diff types of experimental designs
randomized controlled trials
pre-test/post-test
crossover
what is crossover
give more than one treatment but separate the treatments and use randomization to determine the ordering
what is usual care
when there is a therapy in place for something and you want to compare it with the new intervention
what is alternative treatment
testing out two treatments for one outcome to see which one gets better results
is alternative treatment design ethical?
YES
Ethical since both treatments should benefit the patient, but if outcomes are similar the study loses power to determine if an intervention is better
attention control condition
when the control group gets attention but not the intervention.
An example of this is both independent and control groups receiving education on weight loss, but the intervention group is given a supplement to take in addition to the education
what is delayed treatment?
both groups receive intervention, but the control group waits until later to get it.
This option is the most ethical of the C options
limitations of RCTs
inability to test all interventions in this format, the inability to randomize certain variables, and difficulties getting administrative approval to carry out a study
what is a quasi-experimental study
examines an intervention but does not include randomization
the intervention is offered to everyone in a population. To perform a comparison the researchers would find another area with a similar population
what might a quasi-experimental study call a control group?
comparison group
method to increase the reliability of the results in a quasi-experimental design
use a time-series design (collects data over an extended period, introducing an intervention, then collecting data for a length of time)
disadvantages of quasi-experimental designs
less generalizability of the findings due to increased bias, and the increased chance for confounding variables to influence the results
nonexperimental study AKA…
observational study
cohort design or prospective design begins with…
a presumed cause, looks for a presumed effect
A case-control design involves starting with…
the effect or outcome and looking retrospectively at patient data to observe for potential causative variables
what is a descriptive study?
observes for the prevalence of something without determining a relationship
what is homogeneity?
a method of restricting a study to individuals who have confounding variables and subsequently canceling out the effect of the confounding variable. For instance, if age was a confounding variable then participants could be restricted to a certain age range
what is matching?
developing a comparison group with the same confounding variables as the control group
statistical power: the larger the sample size…
the more powerful
what is internal validity
the extent it can be inferred that the independent variable causes the outcome
threats to internal validity (5)
temporal ambiguity (when the cause does not occur before the effect)
selection (selecting participants rather than using randomization)
history (outside events that can influence participants cooperation with the study)
maturation (time’s effect on the outcome)
attrition
types of studies more likely to face threats to internal validity
Quasi-experimental studies and correlational studies
construct validity
involves inferences from the particulars of the study to the higher order constructs they are intended to represent
Which of the following is not a factor in establishing a causal relationship? A) Confounders B) Relationship C) Temporal D) Sample
Answer D: Sample
Confounders, relationship, and temporal are the three traits of establishing a causal relationship. The sample impacts statistical power.
Which of the following is true of a randomized controlled trial? A) Also known as observational study B) Uses a control group C) Researchers select the participants D) Retrospective
Answer: B, uses a control group
Randomized controlled trials use an intervention, have a control group, randomized participants, and are primary studies.
What is a method or intervention that quasi-experimental studies can use to increase the quality of the results if it does not use a comparison group? A) Time-series Design B) Increase sample size C) Randomization D) Identify confounding variables
Answer A, time-series design
Quasi-experimental studies do not use randomization, otherwise it would be structured as a RCT.
Without a comparison group the biggest disadvantage of a quasi-experimental study is influences from a specific time period on the outcome. A time-series design collects time before and after the intervention to allow data to be compared to observe a change. Increasing the sample size increases internal validity but does not impact the time confounding variable. Identifying confounding variables is important but it does not change the results.
Which of the following are methods to control a study? (Select all that apply). A) Heterogeneity B) Matching C) Randomization D) Script
Answer: B, C, D, matching, randomization, and script.
Homogeneity is used not heterogeneity. Matching, Randomization, Script, Blinding, and Statistical control are all methods of maintaining control in a study.
What type of study collects data at multiple points or over an extended period of time? A) Observational B) Quasi-experimental C) Randomized Controlled Trial D) Longitudinal
Answer D, longitudinal
This is the definition of a longitudinal design. The other three are research design methods and are not specific to a time.