Lecture 5 Flashcards
descriptive analysis
describe the data-
central tendency
spread
inferential analysis
parametric or non
is it valid to generalise
degree finding is due to chance
test for differences between variables or groups
compare means/ median
compare proportions
if normally distributed we use
a parametric
if not normally distributed
we use a non parametric test
parametric test
to determine if there is a significant difference between 2 groups
independent t test
test differnce between 2 independent variables
p value
probability of something occuring to chance
if p value is less then 0,05
means it is less then 5 percent chance due to chance
paired t test
difference between 2 related variables
changes within each subjects
difference of 2 measures
parametric test ANOVA
difference between group means after other variance in outcome variables.
bonferroni adjustment
reduce the chance of making a type 1 error
correlation analysis of data
1- positive
neg 1- negative correlation
0- no correlation
statistical significance
difference is unlikely to occured by chance
clinical importance
difference is large enough to be important clinically
clinical importance
minimum amount of change in a score that an individual patient would identify as meaningful and which would indicate a change in care of a patient
MCID
minimal clinically important difference
MID
minimal important difference
interpreting results
P value
MCID
effect side
confidence interval