Test 4 Flashcards
(137 cards)
The only design that allows for conclusion of cause and effect
experimental designs
The gold standard for EBP
experimental design
Similar to experimental design, assignment to groups is not random. Evidence of cause and effect is not as strong as a result
quasi-experimental design
Studies of intact groups:
case/control
ex post facto
causal-comparative
measurement of a group over time
time series designs
independent variable manipulated at some point in time, group serves as its own control
time series designs
enables researcher to determine effectiveness of intervention & quantify timing of effects
time series designs
Enables inference of results from a carefully selected sample to an overall population
inferential analysis
quantifies the potential effects of error on the results
inferential analysis
Statistical tests in quantitative analysis are selected:
a priori
Type of analysis is driven by:
goals of the analysis
assumptions of the data
number of variables in the analysis
Based on the assumption that the data fall into a specific distribution, usually the normal distribution
parametric
Specific to data that is not normally distributed
non-parametric
a single variable (descriptive and summary statistics, single dependent variable or one group)
univariate
two variables (relationship between two variables such as correlation, single variable predicts an outcome)
bivariate
Simultaneous analysis of multiple variables (greater than two) (multiple predictors on a single outcome or multiple factors on multiple outcomes, multiple groups and several effects)
multivariate
indicates that the probability the results are due to chance is very low
very low p-value
indicates that the test has statistical significance
low p-value
gives inferential analysis its strength
comparison of differences to standard error
Calculation of the probability of error
determines if the intervention has an effect that is real
statistical significance
quantify if the difference is important
clinical significance
Reflects the extent to which an intervention can make a real difference in patients’ lives
clinical significance
statistics that inform the importance of findings
confidence intervals
minimum important difference
effect size
this sample statistic equals the population parameter
estimation