Evidence-Based Practice Flashcards
Independent Variable
activity or factor believed to bring about change in dependent variable
Dependent Variable
change or difference resulting from intervention; outcome
Null hypothesis
no relationship exists between variables
Data types: Nominal, Ordinal, Interval, Ratio
Nominal: categories based on characteristics (male/female)
Ordinal: ranked categories (GPA, MMT)
Interval: classified based on scale w/o true zero
Ratio: classifies based on equal interval true zero
Effect size
the size of differences between sample means
generalizability
degree at which the study’s findings based on a sample apply to the entire population
internal validity
the degree to which the observed differences on the dependent variable are a result of manipulation of the independent variable
external validity
the degree to which the results are generalizable to individuals outside the study
face validity
the assumption of validity based on the appearance of an instrument as a reasonable measure.
content validity
the degree to which an instrument measures an intended content area
concurrent validity
teh degree to which the scores on one test are related to the scores on another test
predictive validity
the degree to which a test is able to predict future performance
construct validity
the degree a test measures intended hypothesis
hawthorne effect
the subject’s knowledge of participation in an experiment influences the results
interrater reliability
the degree 2 or more raters can get same rating
intrarater reliability
one rater, multiple ratings
test-retest reliability
test is stable over time
split-half reliability
one half of test compared to the other for internal consistency.
sensitivity
test’s ability to correctly identify people with condition
specificity
tests ability to correctly identify people without condition
predictive value
tests ability to estimate likelihood person will test (+) for condition
Cohort Study
prospective study of participants with condition compared with matches group without condition
case-control study
retrospective study of individuals with similar condition
Range
difference between highest score and lowest score
Standard Deviation
Determination of variability of scores from mean
Normal distribution
bell curve, mean/median/mode close to same; 68% between 1 and -1 SD
Inferential Statistics- purpose
to determine how likely the results of a study of a sample can be generalized to a population
Alpha level
preselected level of significance- probability level; measured in P
Standard error
expected chance of variation among the means from sampling error
type 1 error
null hypothesis is rejected by the researcher when it is true
type 2 error
null hypothesis is not rejected when it is false
Parametric statistics- defined
testing is based on population parameters; includes tests of significance based on interval/ratio data; Must be:
1) Normal distribution
2) random sampling
3) equal variance in groups
T-test for independent samples
Parametric test to compare 2 independent groups by random assignment (i.e. test whether hand splint improves pain in RA pts
T-test for paired samples
compares the difference between matches samples; can be one-tailed (directional hypothesis (i.e. does the intervention improve outcomes?) or two-tailed based on non-directional hypothesis (i.e. either group could have better outcomes) ** T-test can only compare 2 groups
ANOVA
Analysis of Variance: parametric to compare 3 or more independent tx; Simple: 3 tests of posttest scoares are compared from 3 categories; OR Factorial ANOVA: compares multiple groups on 2 or more independent variables
ANCOVA
Analysis of covariance: parametric comparing 2 or more treatmetn groups while controlling for effects of other variables
Nonparametric Statistics- defined
testing not based on population parameters; includes tests on nominal or ordinal data; Used when:
1) Can’t meet parametric assumptions
Chi square test
Nonparametric test to compare data in the form of frequency counts in 2 or more mutually exclusive categories (rate tx performance)
Correlational Statistics
determine the relative strength of a relationship between variables
Pearson product-moment coefficient(r)
used to correllate continuous data with underlying normal distribution on interval/ratio scales
Spearman’s rank correlation coefficient (r)
nonparametric test used to correlate ordinal data (verbal vs. reading comprehension scores)
Point biserial correlation
one variable is dichotomous (nominal) and the other is ratio or interval (relationship between elbow flexor spasticity and side of stroke)
Rank biserial correlation
one variable is dichotomous (nominal) and the other is ordinal (relationship between gender and functional ability)
Intraclass Correlation coefficient (ICC):
reliability coefficient based on analysis of variance
Common variance
representation of the degree that variation in one variable is attributable to another variable
Linear regression
used to establish the relationship between two variables as a basis for prediction