FA 3 Flashcards
epidemiology - Mean
Sum of values / total number of values
epidemiology - Median
Middle value of a list of data sorted from least to greatest
epidemiology - Mode
MC value
Measures of central tendency: Most affected by outliers (extreme values)?
mean
Least affected by outliers (extreme values)?
Mean mode or median?
mode
Measures of dispersion
- Standard deviation
2. Standard error of the mean
Standard deviation (SD or σ)
How much variability exists from the mean in a set of values
Standard error of a mean (SEM)
An estimate of how much variability exists between the sample mean and the true population mean
Standard deviation - standard error of the mean
SEM=σ/(n riza)
Normal distribution (proportion
1σ
2σ
3σ
68%
95%
99.7%
Nonnormal distributions
- Bimodal
- Positive screw
- Negative screw
Positive skew
Asymmetry with longer tail on right (peak at left)
Mean>median>mode
Negative skew
Asymmetry with longer tail on left (peak on right)
Mean is the smaller
statistical variance - definition and equation?
Variance is how far a set of numbers are spread out
variance = (standard deviation) in square
how to decrease SEM (standard error of the mean)
increases n
standard deviation vs precision
increased precision –> decreased standard deviation
Alternative H1 vs null (H0) hypothesis
alternative: Hypothesis of some difference or relationship
null: no difference or relationship
Outcomes of statistical hypothesis testing
- Correct results
a. Null b. Alternative - Incorrect results
a. Type I error (α) Type II error (β)
Outcomes of statistical hypothesis testing - correct results explain
- Stating that there is an effect or difference when one exists (null hypothesis rejected in favor of alternative hypothesis )
- Stating that there is not an effect or difference when none exists (null hypothesis not rejected)
Incorrect result - type I error
Stating that there is an effect or difference when none exists (null hypothesis incorrectly rejected in favor of alternative hypothesis)
FP ERROR
type I error….α?
It is the probability of making a type I error
type I error…..p?
It is judged against a preset α level of significance (usually 0,05). If p less than 0.05, then there is less than a 5% of chance that the data will show something that is not really there
Type II error
Stating that there is not an effect or difference when one exists (null hypothesis is not rejected when it is in fact false)
FN ERROR
Type II error …….β?
Β is the probability of making a type II error. Β is related to statistical power (1-β), which is the probability of rejecting the null hypotephesis when it is false
Increase statistical power and decrease β by
- Increase sample size
- Increase expected effect size
- Increase precision of measurement
confidence interval - definition
range of values in which a specified probability of the means of repeated samples would be expected to fall
confidence interval - equation
CI=mean +-Z (SEM)
confidence interval often used
95% CI (corresponding to p=0.05)
For the 95% CI:
Z?
Z=1.96
For the 99% CI:
Z?
Z=2.58
95% CI for a mean difference between variables
if it includes 0 then there is no significant difference and Ho is not rejected
95% for ODDS ratio or relative risk
IF it includes 1, Ho is not rejected
if the CI between 2 groups overlap
usually NO significant difference exists
statistical power (1-β)?
the probability of rejecting the null hypotephesis when it is false
Common statistical tests
- t-test
- ANOVA
- Chi-square (x^2)
T - test definition / example
Checks differences between MEANS OF 2 groups
- Comparing the mean blood pressure between men and women
ANOVA test - definition and example
Checks differences between means of 3 or more groups
- Comparing the mean blood pressure between members of 3 ethnic groups
Chi-square (x^2) test - definition and example
Checks differences between 2 or more PERCENTAGES OR PROPORTIONS of categorical outcomes (NOT MEANS)
- Comparing the percentage of members of 3 different ethnic groups who have essential hypertension
Meta-analysis?
a statistical procedure that integrates the results of several independent studies considered to be combinable
t-test vs ANOVA vs CHI-square according to action
t-test –> checks difference between means of 2 groups
ANOVA –> Checks differences between means of 3 or more groups
CHI-square –> Checks differences between 2 or more percentages or proportions of categorical outcomes (not mean values)
t test - types (explain)
independent (nonpaired) –>2 different groups of persons are sampled on one occasion (eg. one group with the drug A, and one group with the drug B)
dependent (paired) –> The same persons are sampled on 2 occasions (before and after the treatment)
ANOVA - types (explain)
- one way analysis –> 1 variable (eg. weight loss mean in 3 different programs)
- 2 way analysis –> 2 variables (eg. weight loss mean in 3 different programs and men vs women)
Pearson correlation coefficient (r): range
-1 …..+1
the closer the absolute value of r is to 1
the stronger the linear correlation between the 2 values
positive vs negative r value –>
positive correlation: as one variable increases, the other variable increases
negative correlation:as one variable increases, the other variable decreases
Coefficient of determination
r^2 (value that is usually reported)
ROC (receiver operating characteristic) - definition and explanation
is a graphic representation between sensitivity (y axis) and 1-specificity (FP rate) (x axis) for a diagnostic test
explanation –> the closer the curve is to the diagonia, the less discriminating ability of the test. The closer the curve to the y axis, the better discriminating ability of the test
variables - definition
a quantity that changes under different circumstances
variables - types and definitions
- independent variables –> characteristics that an experimenter can change (eg. amount of salt in a diet)
- dependent variables –> outcomes that reflect the experimental change (blood pressure under different salt regiments)
