Experimental design Flashcards
purpose of measurement
- quantify the presence of something (an attribute)
- to examine difference in the degree to which an attribute is present
goal of measurement
to quantify a single isolated attribute in a standardized manner
instrument reliability
instruments are considered to be reliable when they provide consistent measures over time under the same conditions
data is collected…
via numerical data, questionnaires, scales measurement data
terms to look for instrument reliability
test-retest reliability
interrater reliability
Cronbach’s alpha >0.80
instrument validity
instruments are considered to be valid when they actually measure what they are suppose to
terms to look for for instrument validity
content validity
- asks experts to judge items on a tool
criterion-validity
- compare results with an established tool
sample size
must be sufficient to support the research purpose and design
* power analysis
- terms EFFECT SIZE
internal validity
applies to the inferences about relationships among study variables
threats to internal validity
- homogeneous samples at baseline
- randomizations
- rigorous data collection (instrument reliability and validity)
- attrition: drop out rates
- response rate
- large enough sample size? (power analysis)
external validity
the degree to which study findings are generalizable to other populations
data analysis
p-value
probability statistics
P-value < 0.05 means the likelihood of the results being due to chance is less than 1/20
Therefore, the effect is statistically significant and the intervention is believe to have caused the change in outcome.
confounding variable
when there is an alternative explanation for the outcome of a study that was NOT taken into account in the study design
