Business Stat

Front 1

respondents

Back 1

those who answer a survey

Front 2

subjects/participants

Back 2

people in an experiment

Front 3

experimental units

Back 3

other inanimate objects, animals, plants, etc

Front 4

variables

Back 4

characteristics recorded about each individual or case

Front 5

identifier variable

Back 5

unique kind of categorical variable. assigned to each individual or item in a group. do not have units.

Front 6

interval scale vs ratio scale

Back 6

interval=has no defined value for 0
ratio=has a defined value for 0

Front 7

biased

Back 7

surveys that over or under emphasize some characteristics of the population. the summary characteristics of a sample differ from the corresponding characteristics of the population it is trying to represent

Front 8

sampling error

Back 8

sample to sample differences

Front 9

census

Back 9

sample the includes the entire population

Front 10

parameters

Back 10

key numbers in the model

Front 11

statistic

Back 11

any summary found from the data

Front 12

sample statistic

Back 12

when statistics is matched with the parameters they estimate

Front 13

sampling frame

Back 13

list of individuals from which the sample will be drawn

Front 14

sampling variability

Back 14

sample to sample differences

Front 15

stratified sampling

Back 15

use srs within each stratum (homogenous group) and combine the results at the end

Front 16

clusters

Back 16

parts that represent the whole population

Front 17

cluster sampling

Back 17

performing a census within one or a few clusters

Front 18

nonresponse bias

Back 18

when individuals dont respond and share certain characteritstics

Front 19

undercoverage

Back 19

some portion of the population is not sampled at all or has a much smaller representation in the sample than it does in the population

Front 20

frequency table

Back 20

organizes data by recording the totals for each category

Front 21

relative frequency table

Back 21

displays the percentages of the counts

Front 22

categorical data condition

Back 22

bar charts and pie charts

Front 23

histogram

Back 23

like a bar chart, but used to display quantitative data

Front 24

uniform distribution

Back 24

doesn't appear to have any modes and all the bars are approximately the same height

Front 25

skewed

Back 25

if one tail stretches out further than the other tail

Front 26

when to use median

Back 26

if the distribution is skewed, contains gaps or outliers. is resistant because it is not affected by outliers, gaps, etc.

Front 27

range

Back 27

measure of spread

Front 28

quartiles

Back 28

frame the middle 50% of the data

Front 29

5 number summary

Back 29

reports the median, quartiles and extremes

Front 30

z-score

Back 30

standardized value from the mean

Front 31

time series plot

Back 31

display of values against time

Front 32

sample space

Back 32

collection of all possible outcomes

Front 33

probability

Back 33

its likelihood, the long-run relative frequency of an event

Front 34

independence

Back 34

the outcome of one trial does not influence the outcome of another

Front 35

addition rule

Back 35

add two probablities of events together if they are disjoint. this gives us the probablity that either occurs

Front 36

discrete random variable

Back 36

if we can write out all of the possible outcomes (such as number of students enrolled in a class)

Front 37

continuous random variable

Back 37

if it can take on any outcome between 2 variables (such as GPA)

Front 38

bernoulli trial

Back 38

only success or failure, p and q for each trial, trials are independent

Front 39

binomial model

Back 39

predicts the number of successes in a series of bernoulli trials

Front 40

probability density function

Back 40

shows the distribution of probabilities

Front 41

empirical rule

Back 41

68-95-99.7

Front 42

simulation

Back 42

use a computer to imitate drawing random samples from some population of values over and over

Front 43

sampling distribution

Back 43

distribution of proportions over many independent samples from the same population

Front 44

central limit theorem

Back 44

the larger the sample, the better the approximation will be. sampling distribution of the mean becomes normal as the sample size grows

Front 45

p-value

Back 45

probability of seeing the observed result or something even less likely than the null hypothesis. reject the null hypothesis with a low p-value

Front 46

type I error

Back 46

false positive. you reject the null hypothesis, but you really should not have rejected it.

Front 47

type II error

Back 47

false negative. the null hypothesis really is false, but you say that it is true.

Front 48

how to reduce errors

Back 48

increase the sample size!

Front 49

paired data

Back 49

before and after measurements of some property

Front 50

scatter plots

Back 50

ideal way to picture associations between two quantitative variables