stats unit 1: ch 1-3

individual

person, animal, or thing described in a set of data values

variable

attribute that can take different values for different individuals

catergorical variable

values that are taken as labels

quantitative variable

values that are quantities

population

entire group who’s info is desired

census

collection of data from every individual in the population

sample

subset of individuals in a population from which data is collected

observational study

set observation: no attempt to influence response

experiment

imposes treatments on individuals, then measures response to treatment

convenience sample

individuals are easy to reach

bias

systematic tendency where methods used to produce data produces an inaccurate/biased depiction of reality

voluntary response sample

people choose to be in a sample by responding to general invitation

random sample

individuals in population are selected by chance process

under coverage

members are less likely or cannot be chosen for sample (invalid)

nonresponse

individual cant be contacted or refuses to participate in sample

response bias

consistent inaccurate responses

simple random sample (n)

sample chosen where every group of n individuals have an equal chance of selection

• label

• randomize

• select

strata

shares characteristics, associated with variables being measured

stratified random sampling

a simple random sample is chose from each stratum and combined into one overall sample

cluster

individuals are near each other in a population

cluster random sampling

clusters randomly chosen, the every individual in these clusters are selected

systematic random sample

sample from ordered arrangement, randomly selecting one of the first k individuals then choosing every kth individual after

response variable

measures outcome of study

explanatory variable

can help explain or predict changes

confounding

two or more variables are associated and effect the response variable

treatment

condition applied to individuals in experiment

experimental unit

object or person where treatment is random assigned

subject

human experimental unit

placebo

treatment has no active ingredient

control group

provides baseline to compare effects of treatment

placebo effect

subjects responding in favor of treatment, even a placebo

double blind

subjects or those who interact with them who measure response don’t know what treatment their receiving

single blind

either subject or people who measure the response don’t know which treatment a subject is receiving

frequency table

shows number of individuals having each data value

bar/pie chart

shows each category as a bar or sector of circle

two way table

summarizes relationship between two catergorical variables for some group of individuals

dotplot

shows each data value as a dot above its location on a number line

roughly symmetrical

right and left sides of data is nearly mirrored

skewed data

one side of plot is clearly longer

• left side tail longer: skewed left

• right side tale longer: skewed right

approximately uniform

all dot plots have roughly the same heights

unimodal

single peaked

bimodal

double peaked

stemplots

stem: left digit

leaf: final digit

median

middle point in numerical data

mode

most common data poinr

histogram

shows each interval of a bar, height of bar determines frequencies of each interval

scatterplot

shows relationship between to quantitative variables measured on same individuals

mean

sum of all data points divided by count

range

difference between largest and smallest datapoint

standard deviation

calculate mean, the subtract mean from each data point, find average of answers then take square root of average

interquartile range

first quartile: median of data values less than median (Q1)

third quartile: median of data values greater than median (Q3)

IQR= Q3 - Q1

five number summary

min, Q1, median Q3, max

outliers

low: Q1 - 1.5 x IQR

high Q1 + 1.5 x IQR

box plot

visual representation of five number summary

percentile

percentage of values in a distribution that are less than that individual’s value

z score

value - mean/std dev