Lit Review

Front 1

Populations

Back 1

All objects(subjects) of a particular kind in the universe (Ex all hypertiensive patients)

Front 2

Sample

Back 2

Portion of the population (must be a representiative of the population)

Front 3

Parameters

Back 3

Measurements that describe a population (Ex number of male hypertensive patients)

Front 4

Statistics

Back 4

Measurements that describe a sample (Ex ratio of fale to female hypertensive patients in a sample)

Front 5

Variable

Back 5

Charicteristic that is being observed of measured (data)

Front 6

Independant Variable

Back 6

Intervention (Ex antihypertinsive drugs)

Front 7

Dependant Variable

Back 7

Whats being measured(Ex blood pressure)

Front 8

Quantitative Discrete Data (numerical)

Back 8

has a limited set of values and assumes the value of a hole number

Front 9

Qualitative Discrete Date (categorical)

Back 9

Non-numerical and falls into specific catagories (Ex sex, eyecolor)

Front 10

Quantitative Continuous Data (numerical)

Back 10

This type of data can take on any value within a defined range (Ex hight, Weight, Blood glucose)

Front 11

Nominal Scale

Back 11

Scale where data that fits into a mutually exclusive catagories without any implied order of rank (Ex male or female)

Front 12

Orinal Scale

Back 12

Scale where data also fits into a catagory but there is an implied order of rank (Ex pain scale, tumor grading)

Front 13

Interval Scale

Back 13

Has no true 0 point (Ex. calendar, celcius scale)

Front 14

Ratio Scale

Back 14

Has a true 0 point (Ex. kalvin scale)

Front 15

Descriptive Statisitics

Back 15

Just describe the data in a sample

Front 16

Standard Error of the Mean (SEM)

Back 16

SEM is used to estimate the true mean of the population from which the sample was drawn (its used to calculate the confidence intervals) SEM=SD/root(n)

Front 17

Inferential Stats

Back 17

Determines if an observed diffrence in outcome measured is due to chance or real diffrence (hypothesis testing)

Front 18

Event Rate

Back 18

# of people experiencing event as proportion of # of people in the population

Front 19

Relative Risk

Back 19

Ratio of event rate in exposed to unexposed group

Front 20

Relative Risk Reduction RRR

Back 20

Difference in event rates between 2 groups as a proportion of the event rate in the untreated group (more impressive)

Front 21

Absolute Risk ARR

Back 21

Arithmetic difference between 2 rates (more important to look at)

Front 22

Number Needed to Treat NNT

Back 22

The number of patients that would have to be treated for 1 patent to benifite (NNT=100/ARR)

Front 23

Number Needed to Harm NNH

Back 23

The number of patients that would have to be treated for 1 patient to experience an adverse effect (NNH=100/Absolute Risk Increase)

Front 24

Type I (Alpha) Error

Back 24

Occurs when the null hypothesis is rejected when no true difference exists between study groups (Cause: Chance/sampling variation)

Front 25

Type II (Beta) Error

Back 25

Occurs when there is a failure to reject the null hypothesis when a true difference exists between study groups (Cause: Sample size too small)

Front 26

Study Power

Back 26

The power of a study is the ability of the study to detect a statistically significant difference between study groups if one dose exist and is related to sample size (Power= 1-Beta)

Front 27

Null Hypothesis

Back 27

The null hypothesis is the "no difference" hypothesis and is the basis for statistical analysis