Specific tests Flashcards
Chi square test
definition and assumptions
for both 2 and 2+ groups, chi square test compares group proportions and if they are different from that expected by chance
assumption: usually chi-square distribution fro nominal type data
no cell with expected count of <5
Fisher’s exact test
like chi, but good for groups of 2 or more with EXPECTED cell count of 5+
McNemar Test
2 groups, and if they are from same group
Cochran
same principle as chi square and assumptions, yet mathematical factors in concept of paired or related data
Log Rang
Nominal, compares proportions of, or time to, event occurences between groups
Contingency coefficient
nominal, correlation test
provides a quantitative measure of the strength and direction of a relationship between variables
r = values range from ( -1.- + 1.0 )
Logistic Regression
provides a measure if the relationship between variables by allowing the prediction about the dependent variable, or outcome, knowing the value/rank of others independent variables (IVs)
alos able to calculate OR for a measure of association
bonferroni test of inequality
adjusts the p value for # of comparisons being made
VERY conservative
Mann-Whitney test
2 groups, independent
same as X^2 but for ordinal data
Wilcoxon Signed Rank test
for related data of 2 groups
compares median values between groups
Friedman test
compares the median values between groups
if 3+ group comparison significant, must perform post-hox test to determine where differences are
Student Newman-Keul test
post-hoc test for 3 or more group comparisons
compares all pairwise comparisons possible
all groups must be equal in size
Dunnett test
post-hoc test for 3 or more group comparisons
compares all pairwise comparisons against a single control
all groups must be equal in size
Dunn test
post-hoc test for 3 or more group comparisons
compares all possible multiple comparisons
useful when all groups are not equal size
student t test
2 groups, interval data
compares the mean values between groups
ANOVA
1 dependent variable
compares the means of all groups along with intra and inter group variations against a single DV
“Analysis of Variance”
Manova
2+ dependent variables
compares means of all groups (along with intra and inter group) variations against multiple DVs
ANCOVA
3+ groups
Analysis of Covariance
compares the means of all groups against a single DV while also controlling for the co-variance confounders
MANCOVA
multiple analysis of Co-Variance
compares the means of all groups against multiple DVs while also controlling for the co-variance of confounders
Tukey or Scheffe tests
compares all possible multiple comparisons
all groups must be equal in size
tukey test
slightly more conservative than the stu N.K
Scheffe test
less affected by violations in normality and homogeneity of variances
more conservative than tukey
kappa statistic
agreement between evaluators
consistency of “decisions” “determinations”
kappa interpretation
+1 = the observers perfectly “classify” everyone exactly the same way
O = there is no relationship at all between the observers
“classifications” above the agreement that would be expected by chance
-1 = the observers “classify” everyone exactly the opposite of each other
kappa can be + or -
\+ = good agreement - = bad agreement
Prediction =
regression test
regression measures the relationship between variables by allowing the prediction about dependent variables knowing the value rank of others independent variables
Correlation tests =
what kind of relationship is it? is it linear? for example?
Survival =
time to event test
compares how long a “period” of time to event between groups
Association =
regression test
Mantel Haenszel Test
test of ORs for adjusted/crude ORs testing for confounding
