Search
Create
Log in
Sign up
Log in
Sign up
BCPS statistics
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (68)
TEST TO CONFIRM NORMAL DATA
KOLMOGOROV-SMIRNOV
TEST FOR CONTINUOUS DATA (2 GROUPS RELATED)
PAIRED T-TEST
TEST FOR CONTINUOUS DATA (2 GROUPS INDEPENDENT)
T-TEST
TEST FOR CONTINUOUS DATA (>2 GROUPS AND INDEPENDENT, NO FACTORS)
ANOVA (ONE-WAY)
TEST FOR FOR CONTINUOUS DATA (>2 GROUPS AND RELATED, NO FACTORS)
REPEATED-MEASURES ANOVA
TESTS FOR ORDINAL INDEPENDENT DATA (2 GROUPS)
WILCOXON RANK SUM
MANN-WHITNEY U TEST
WILCOXON-MANN-WHITNEY
TEST FOR ORDINAL INDEPENDENT DATA (>2 GROUPS)
KRUSKAL-WALLIS
TESTS FOR NOMINAL INDEPENDENT DATA (2 GROUPS)
CHI SQUARE (X2) OR FISHER EXACT
TESTS FOR NOMINAL RELATED DATA (2 GROUPS)
MCNEMAR TEST
TESTS FOR NOMINAL INDEPENDENT DATA (>2 GROUPS)
CHI SQUARE (X2)
TESTS FOR ORDINAL RELATED DATA (2 GROUPS)
WILCOXON SIGNED-RANK
SIGN TEST
TESTS FOR ORDINAL RELATED DATA (>2 GROUPS)
FRIEDMAN ANOVA
TEST FOR NOMINAL RELATED DATA (> 2 GROUPS)
COCHRAN Q
TYPE 1 ERROR
FOUND A DIFFERENCE WHERE NONE WAS
TYPE 2 ERROR
FAILED TO FIND DIFFERENCE WHERE THERE WAS ONE
WHAT IS A SIGNIFICANT CONFIDENCE INTERVAL FOR OR/RR?
DOES NOT INCLUDE "1"
DIFFERENCE BETWEEN CI AND PVALUE
CI SHOWS MAGNITUDE OF DIFFERENCE AS WELL AS SIGNIFICANCE
DIFFERENCE BETWEEN REGRESSION AND CORRELATION
CORRELATION SHOWS ASSOCIATION, REGRESSION SHOWS PREDICTABILITY
TESTS FOR SURVIVAL ANALYSIS
KAPLAN-MEIER (calculates)
LOG RANK TEST (compares 2 groups)
COX PROPORTIONAL HAZARDS (evaluates several variables)
SPECIFICITY
TRUE NEGATIVES
= B / B+D
SENSITIVITY
TRUE POSITIVES
= A / A+C
POSITIVE PREDICTIVE VALUE
= A / A+B
NEGATIVE PREDICTIVE VALUE
= C / C+D
MEASURES OF RISK ASSESSMENT
ODDS RATIO
RELATIVE RISK
ABSOLUTE RISK REDUCTION
NNT
ODDS RATIO
(A/C) / (B/D)
RELATIVE RISK
(A / A+B) / (C / C+D)
ABSOLUTE RISK REDUCTION
(A / A+B) - (C / C+D)
NNT
1/ARR
P-value represent
probability of a Type I error (Found a difference when there was none)
Alpha
p-value
If p = 0.01
There is a 1% chance that a Type 1 error is committed ( Found a difference when there was none)
Beta represents
probability of a type II error (FAILED TO FIND DIFFERENCE WHERE THERE WAS ONE)
Beta
Power
Power =
1-beta
Higher Beta
less power (closed to worse power)
Confidence Interval
range of values you believe the true value lies between
Confidence interval not statically significant for a study
contain 0 (not OR study)
Examples nominal data
Yes or No variables or sex, mortality, dz presence, race, marital status
Nominal data
2 or more groups with no definite order
Ordinal data
have a specific order but no clear difference between each value
Examples of ordinal data
pain scale
NYHA scale
Likert scele
Continuous variables
distinct measurable difference between values ca be ratio or interval
non parametric variables
nominal and ordinal data
Parametric variables
continuous (ratio and interval)
Parametric data
must be normal distributed
What is the easiest way to determine if you have 'normal' or gaussian distribution?
look at median and mean, if they are close it is normal distribution
standard deviation
used for continuous data
Which types of descriptive statistics are sensitive to outliers?
mean is sensitive, median is not
mean can be used for which variables
nominal and continuous
median can be used for which variables
Ordinal and continous
mode can be used for which variable
Ordinal, nominal, and continous
SEM
Quantifies the uncertainty and the estimate of the mean
Coefficient of variation equation
SD/mean * 100%
__ % OF THE SAMPLE DATA FALLS IN 1 SD
68%
__% OF THE SAMPLE DATA FALLS IN 2 SD
95%
__% OF THE SAMPLE DATA FALLS IN 3 SD
99%
Do not use correlation
When there is a non linear relationship
Mean used for what type of data?
-continuous and
-normally distributed
Standard deviation data has to be...
-continuous
-normally distributed
What does iqr encompass?
-the 25th to 75th percentile
Best measurement of central tendency for continuous data
Means and SD
Does the CI and the p-value need to be reported?
No
means and SD should not be reported with which type of data
Ordinal
Standard deviation is
Most common measure used to describe a spread of data
Variance equation
Standard deviation squared
Coefficient of variation meaning
Relates the mean and the SD
Another indicator data isn't normally distributed
If the SD is larger relative to the mean it's likely it isn't normally distributed
Why do you need to calculate SEM?
-calculate confidence intervals
-hypothesis testing
-deception
;