Stats exam 1 vocab

data

observations collected from field notes, surveys, and experiments

statistics

the study of how to best collect, analyze, and draw conclusions from data

statistical investigation process (6)

identify question, design study & collect relevant data, describe data, statistically analyze data, make inferences, consider limitations

statistic

a single number summarizing a sample of data

observational unit

a single person or object being observed

variables

characteristics of observational units

data set

a way to organize data where rows correspond to observational units & columns correspond to variables

quantitative

variable type that can take on numerical values, where it's sensible to ad, subtract, or take averages of those values

discrete

quantitative; responses can only take certain numerical values with spaces between numbers; can be counted

continuous

quantitative; responses can take on infinetly many possible values; must be measured not counted

categorial

variable where responses are categories; qualitative

population

target group of interest, usually too large to collect data for every observational unit in it; census

sample

a subset of the observational units in the population

parameter

number summarizing the population

anecdotal evidence

haphazard way to collect data, where observational units are probably not actually representative of the population

bias

when a sample is skewed toward a particular interest, making it not representative of the population

sample bias

when the method used to select a sample makes it not representative of the population

convinience sample

a type of nonrepresentative sample where easily accessible indiciduals are more likely to be included

voluntary response sample

a type of nonrepresentative sample where individuals choose themselves whether or not to be included in the sample

non-response bias

when too many people refuse to provide information in a survey and it's unclear whether the respondents area good representation of the population

response bias

when individuals provide inaccurate information

observational study

a way of collecting data that doesn't directly interfere with the observational units

prospective study

type of observational study that identifies individuals and collects infomration as events unfold

retrospective study

type of observational study that collects data after events have taken place

confounding variable

a varialbe that wasn't included in the study but is associated with both the variables of interest; lurking variable

simple random sampling

each observational unit in the population has an equal chance of being included

stratified sampling

population is divided into groups of similar observational units called strata, then individuals are selected using a second sampling method from each stratum

cluster sampling

obervations are in groups with a lot of variability, called clusters; entire clusters are randomly selected and combined to create the sample

multi-stage sampling

similar to cluster sampling, except simple random samples are taken within each selected cluster

systematic sampling

a frame is not available but oversvationsl units are encountered in a sequential manner; a random number is generated for the first observational unit to include and after that every ?th unit is also included

frequency table

a table for a single variable, showing the number of observational units associated with particular values of the variable

proportion

percentage, written as a decimal value

relative frequency table

A frequency table showing proportions instead of numbers

bar chart/graph

categories of variable are on one axis & counts/proportions are on the other axis

distribution frequency table

the possible values of the variable along with how often each occurs

common issues with misleading graphs

a non-zero starting point, 2-dimensional pictures to depict a single variable, making graphs 3-dimensional, two separate vertical scales

success

the outcome or category of interest

p^

number of successes/n

n

sample size, the number of observational units in the sample

point estimate

a value computed from sample data that can be used to estimate the corresponding value in the population

hypothesis test

a statistical technique used to evaluate the completing claims using data

null hypothesis

H0; the first claim which takes a stance of no difference or effect

alternative hypothesis

HA; the second claim which is concluded if we reject the null

randomization

mimics what would happen based only on random fluctuation

simulation

repeat randomization to represent the outcomes associated with all the observational units in a sample, under the assumption of the null hypothesis

p-value

probability of observing data at least as favorable to the alternative hypothesis as our current data set, if the null hypotheses were true

test statistic

a summery statistic of the data used to compute the p-value and evaluate the hypotheses

statistically significant

results that produce a p-value that is less than a previously set threshold

significance level

a, the threshold for determining whether a p-value is small

a (alpha)

0

type 1 error

rejecting the null hypothesis when H0 is actually true

type 2 error

failing to reject the null hypothesis when HA is actually true

type 1 error is more serious

smaller significance level

type 2 error is more serious

bigger significance level

confirmation bias

looking for data that supports our ideas

one sided test

a test using hypotheses that explore only one direction of possibilities; HA contains < or >

two sided test

test using hypotheses that include both directions of possibilities; HA contains x=

null distribution

a smooth looking distribution representing proportions due to chance

p

the population proportion of successes

null value (P0)

the reference value for the parameter in H0