biostat 212 final

observational unit

one row of the spreadsheet

variable

characteristic of interest gathered from each observational unit

population

entire group we have interest in learning more about

parameter

numeric value that describes some characteristic about the population

statistic

numeric value that describes some characteristic about a sample

nominal data

data that falls into non-numeric categories that don't have any ordering

ordinal data

data that falls into categories that have some sort of meaningful order but NOT numeric

discrete data

data that follows a numeric scale, but only takes limited value

continuous data

data that is numeric and measurable, and can be taken in any range

right skew

majority of data is to the left, positive skew

left skew

majority of data is to the right, negative skew

mode

most common data value

mean

average

median

midpoint

q1

median of lower half of data

q2

median of the entire set

q3

median of the upper half of the data

5 number summary

min, Q1, median(q2), Q3, max

range

distance from max to min

inter-quartile range

distance from q1 to q3

mean absolute deviation

the average distance between each data value and the mean

standard deviation

typical deviation from the mean

proportion

number of cases that fit a category of interest of out of the total

p

parameter, proportion of population in a category

p hat

statistic, proportion of sample in a category

boxplot

graphical representation of 5 number summary

univariate

data that describes a single characteristic of the population

multivariate

relationship between more than two variables

jitter plot

shows all data points with black bar at sample mean

absolute risk reduction

difference of risks

absolute risk

proportion of cases out of the total

effectiveness

percent of individuals that avoid infection by taking part in intervention

relative risk

ratio of risk under one condition to another condition

response variable

variable that we have an interest in explaining or understanding better, y axis

predictor variable

a variable whose scores are used to estimate scores of a criterion variable, x axis

stratification

analytical process of breaking up one variable into subgroups based on the value of another variable

causality

the relationship between cause and effect

mediating variable

a variable that helps explain the relationship between two other variables

confounding variable

a factor other than the factor being studied that might influence a study's results

scatterplot

a graphed cluster of dots, each of which represents the values of two variables

correlation coefficient

a statistical index of the relationship between two things (from -1 to +1)

interpolation

predicting y based on an x value within the range of x values we have info for

extrapolation

predicting y baed on an x value outside of the range of x values we have info for

simpsons paradox

An association between two variables that holds for each individual value of a third variable can be changed or even reversed when the data for all values of the third variable are combined.

statistical noise

Variables influencing the outcomes of a process in unpredictable ways.

residual variance

variability due to unknown or uncontrolled variables; error term

coeffecient of determination

(r2) measure the % of total variation in the response variable that is explained by the least-squares regression line; must be in between or equal to 0 and 1; r2=0 no explanatory value; r2=1 line explains 100% of the variation

probability

likelihood of observing particular outcomes from the same known population

statistics

describing observed data and making claims about a population

sample sapce

the collection of all possible outcomes

outcome

A possible result of a probability experiment

event

particular situation that we want to investigate

complement

every possible outcome that is not already included in the observed event

compound event

An event made up of two or more simple events.

disjoint event

both events cannot be true at the same time

union of events

at least one of the events takes places

U

intersection of events

both events take place

U upside down

independent event

one event has no effect of the probability of the other event

dependent event

probability of one event changes based on knowing the outcome of the other

Conditional Probability

probability of one event given known info about another event

theoretical probabilities

known probabilities associated with a population of random process

empirical probability

estimated probabilities based on how many of each outcome we oberved

positive predictive value

probability of a true result given the test result is positive

negative predictor value

probability of a true result given that the test result is negative

sensitivity

probability of a true result given the person has the condition of interest

specificity

probability of a true result given the person does not have the condition of interest

random variable

a variable whose value is a numerical outcome of a random phenomenon

discrete random variable

a random variable that can take one of a finite number of distinct outcomes

continuous random variable

a random variable that may assume any numerical value in an interval or collection of intervals

expected value

mean of a random variable

density curve

Graph of a continuous probability distribution

area under curve is probability

normal distribution

describes a symmetrical, bell shaped curve that shows the distribution of many physical and psychological attributes

z score

number of standard deviations a data value is away from the mean

discrete uniform distribution

all outcomes are equally likely to occur

continuous uniform distribution

all equally-sized outcome ranges are equally likely to occur

population distribution

the entire distribution of a particular random variable

sample distribution

distribution of measurements collected from our sample

converge to the population distribution in shape

convergence in distribution

the larger your sample, the more your sample distribution will converge toward the population distribution in shape

convergence of an estimator

the larger your sample, the more your sample mean will converge toward the population mean in value

law of large numbers

the sample mean from results obtained from a large number of trials should be close to the expected value and tend closer to the expected value as more trials are performed

sampling distribution

the parameter your statistic is estimating

converges to a singular point

central limit theorem

the distribution of sample averages tends to be normal regardless of the shape of the process distribution

standard error of the sample mean

expected error of the sample mean

statistical inference

a claim about a population based only on sample data

hypothesis test

method of inference that examines whether a candidate parameter is plausible or not

null hypothesis

the candidate parameter is true, there is no change

alternative hypothesis

the parameter is some alternative value, at least some change occurs

directional hypothesis

States the direction of the difference or relationship

p value

measures the compatibility between our sample results and the null hypothesis

reject Ho

if p-value is less than alpha

fail to reject Ho

if p-value is greater than alpha

type 1 error

incorrectly rejecting the null hypothesis

type 2 error

incorrectly failing to reject the null hypothesis

weak evidence

p value is greater than 10%

modest evidence

p value is 5-10%