Statistics Flashcards
Sensitivity definition
true positive rate proportion of all people with the disease that test (+) for disease when disease is present values approaching 100% desirable for ruling out disease
Sensitivity equation
= TP / (TP + FN) = 1 - (false negative rate)
specificity definition
true negative rate proportion of people without the disease who test (-) values approaching 100% desirable for ruling in disease
specificity equation
= TN / (FP + TN) = 1 - (false positie rate)
positive predictive value definition
proportion of positive test results that are true positives
positive predictive value equation
= TP / (TP + FP)
negative predictive value definition
proportion of negative test results that are true negative
negative predictive value equation
= TN / (TN+ FN)
Incidence looks at the number of _____ cases in a SPECIFIC period of time
new cases
prevalence looks at _____ cases in a SPECIFIC period of time
all existing cases
incidence equation
incidence rate = (# of new cases in a specific time period)/ (population at risk during same time period)
prevalence equation
prevalence = (# of existing cases) / (population at risk) approximately equal to (incidence rate) X (average disease duration)
Risk / disease ratio table
Disease
+ -
risk factor + a b
- c d
odds ratio
- study used in
- definition
case control studies
odds that the group with the disease (cases) was exposed to a risk factor (a/c) **divided by **the odds that the group without the disease (controls) was exposed (b/d)
odds ratio equation
= (a/c) / (b/d)
= ad/ bc
relative risk
- study used in
- definition
cohort study
risk of developing disease in exposed group **divided by ** risk in the unexposed group
relative risk equation
= (a/(a+b)) / (c/(c+d))
attributable risk definition
difference in risk between exposed and unexposed groups
or
the proportion of disease occurence that are attributable to the exposure
attributable risk equation
= (a / (a+b)) - (c/(c+d))
absolute risk reduction equation
ARR = (control event risk) - (experiment event rate)
number needed to treat
- definition
- equation
number of patients who need to be treated for 1 patient to benefit
= 1/ ARR
number needed to harm
- definition
- equation
number of patients who need to be exposed to a risk factor for 1 patient to be unharmed
= 1/ attributable risk
Standard error of Mean (SEM) =
(standard deviation) / (square root of sample size)
as sample size increases, standard error of mean decreases
normal distrubtion
bell shaped curve (gaussian)
mean = median = mode
positive skew vs. negative skew
positive: mean > median > mode
asymmetry with longer tail on right
negative: mode > median > mean
asymmetry with longer tail on left
Null hypothesis (Ho)….. hypothesis of no difference
Alternative (H1)… hypotheis of some difference
chart
reality
H1 Ho
study result H1 power (1 - beta) alpha
Ho beta correct
Type I error (alpha)
accept alternative and reject null hypothesis
states that there is an effect or difference when none exists
if P < .05
-there is a 5% chance that the data will show something that is not really there
type II error (beta)
accept null… reject alternative hypothesis
stating that there is not an effect or difference when one exists
false negative
power (1- beta)
probability of rejecting null hypothesis when it is in fact false
or
the likelihood of finding a difference if one in fact exists
Confidence interval equation
CI = [mean - Z(SEM)] to [mean + Z(SEM)]
SEM = standard error of mean
Z = Z-score
for the 95% CI: z = 1.96
for the 99% CI: z = 2.58
if 95% CI for a mean difference between two variables includes 0, then there is _____ difference and Ho is _____
no significant difference
Ho is not rejected
if the 95% CI for odds ratio or relative risk includes 1, Ho is ____
not rejected
If the CIs between 2 groups do not overlap, then ____
if the CIs between two groups do overlap, then ____
do not: significant difference exists
do: usually not significant difference exists
test that checks for the difference between the means of **two **groups
T-test
test that checks difference between means of 3 or more groups
ANOVA
(analysis of variance)
test checks difference between two or more percentages or proportions of categorical outcomes (not mean values)
chi-square (X2)
compares percentages or proportions
Parson’s correlation coefficient (r)
r is always between -1 and +1
closer the absolute value of r is to 1: the stronger the linear correlation between two variables
coefficient of determination = r2
