Statistics S1

bimodal

2 peaks

negative

Standard dev. can never be ____

five number summary, because it includes the IQR

When interpreting skewed data, it is best to use ________ to describe it

density curve

a curve with an area beneath it of exactly 1 that describes the overall pattern of a distribution

observational study

a sample survey is a type of ____

sample

a subject of the population

placebo

a treatment w/o any affect or active ingredient

lurking variable

a variable that has an important effect on the relationship among variables in a study but isn't one of the explanatory variables being studied

population

al individuals in which we are interested in

association

as x increases, does y increase or decrease? --> used for nonlinear data

common response

both explanatory and response variable are responding to some lurking variable

mean, median, min, max

changes when you add/subtract data

mean median, min, max, standard dev, range, IQR (everything)

changes when you multiply or divide data

observational study

collecting data w/o interfering in any way

comparative observational study

compare 2 or more groups to determine effect differences

normal curve

describes a distribution that's symmetric, unimodal, bell shaped

correlation

direction, strength, form, shows relationship between varaibles

standard dev, range, IQR

doesn't change when you add or subtract data

extrapolation

estimating a value that lies outside of the data set

interpolation

estimating a value that lies within the data set

standardized value (z-score)

explains how far from the mean a certain score is in units of standard deviation

explanatory variable

explains or causes changes in the response variable, x value

sampling variability

if you take different surveys, there are going to be slightly varying results

in the direction of the skew

in a skewed distribution, the mean moves ....

subject

individuals in an experiment

least-squares regression line

line that makes the sum of the squares of vertical distance data points from the line as small as possible

range

max-min

central tendencies

mean, median, "middle", peaks, modes

standard deviation

measure of spread using actual data values

quantitative variable

measurements w numbers and units , makes sense to do math

response variable

measures on outcome or result of a study, y value

five-number summary

minimum, Q1, med, Q3, max

r

number that describes correlation, decimal number between -1 and 1

median and IQR

of the following measures: median, mean, IQR, and standard deviation, which are not affected by outliers

unimodal

one peak

mean and standard deviation

parameters that describe density curves

categorical variable

responses that group individuals into a category

distribution of a variable

says what value the variable takes and how often it takes these values, can be displayed w a bar graph or pie chart

gaps

separations between groups or points

scatter plot

shows relationship between 2 sets of data w/ 2 variables graphed as ordered pairs on a coordinate plane

regression analysis

stat process for estimating the relationship between variables

sample survey

study a sample

experiment

study in which we apply a "treatment" or otherwise affect the results

census

study of the entire population

variable

the characteristic of the individuals that the data describe

residual

the difference between an observed value and a value predicted

isn't

the mean ______ resistant to extreme values

is

the median _____ resistant to extreme values

counts

the number of times each response occurs

individuals

the people, animals, objects, etc. that the data describe

rates

the percent or proportion of all responses it represents

1. ask a question 2. produce data 3. analyze data 4. interpret results

the statistical problem-solving process

outlier

value that's much higher/lower than the rest of the data

pie chart and bar graph

what are two good ways to measure categorical data

response

what happens to the subjects in response to the treatment

treatment

whatever is done y the experiment that's different

bivariate data

when you measure 2 variables at the same time