# Biostatistics, Epidemiology and Developmental Mile stones .

version from 2015-12-03 19:33

## Section 1

What's precision (reliability)?the consistency and reproducibility of a test (reliability)/ the absence of a random variation in a test (increased precision->reduced standard deviation
What's accuracy (valididty)?the trueness of test measurement (validity) / the absence of systematic error or bias in a test (systematic error-reduces accuracy in a test)
External validity ishow finding is representative of the general population from which sample has been taken
Internal validity isdegree to which the sample is representative of the general population
Validity measureshow close measured value is to the true value
What are the 9 bias possible in a study?1) selection bias 2) recall bias 3) sampling bias 4)late-look bias 5)procedure bias 6)confounding bias 7)lead0time bias 8)observer-expectancy effect 9)hawthorne effect
What means bias in a study?occurs when there is systematic error (reduce the accuracy) or favor in a particular direction
What is a selection bias?non random assignmnent to participation in a study group (ex: bekson bias, loss to follow-up)

## Section 2

Positive skewed mean>median>mode
Negative skewed mean
Type one error ( alpha ) define there's effect or difference when none exists . you saw a difference that did not exist ( null hupothesis incorrectly rejected in favor of alternative hypothesis)
Type one error judge p value , p<0.05 there's 5% chance that data will show something is not really there , p<= 0.05 , reject null hypothesis. p>0.05 do not reject the h0
Type 2 error ( beta ) define you were blind for adifference that did exist, null hypothesis is not rejected when it is in fact false) ( 1-beta )
Increase power and decrease beta by inc sample size , inc expected effect size , inc precision of measurement
Type 2 error calculate beta ( type 1 error ) = 1-power
RR ( - 1.77) , CI ( 1.22-2.45)Statistically significant , inc Risk
RR ( -1.63) , CI ( 0.85 - 2.46 ) Not significant ( risk is the same )
RR ( 0.78) , CI ( 0.56 - 0.94 )Statistically significant , No Risk
t-TEST mean of 2 groups
ANOVA 3 or more groups
Chi-square 2 or more percentages
Pearson correlation coefficient 2 intervals . r always between -1 and +1 , + r - positive correlation . - r - negative correlation
Medicare Elderly
Medicaid Destitute
Autonomy Respect the patient as individual
Beneficence Special ethics from the doctor to act on the patient best interest
Nonmaleficence Do no harm
Justice treat persons fairly and equally