data
info that comes from observation, counts, measurements, or responses
statistics
science of collecting, organizing, analyzing and interpreting data in order to make decisions
populationsample
types of data sets
population
all outcomes, responses, measurements or counts of interest
when a population isn't practical
when do we use a sample
sample
subset of population
to make reasonable guesses
what do we use a sample for
Parameter and Statistic
types of sample
parameter
numerical data abt a population
statistic
numerical data abt a sample
descriptive and inferential
branches of statistics
descriptive statistics
describes or presents data
inferential statistics
makes educated guesses abt a population
qualitative and quantitative
two types of data
qualitative
attributes, labels, and numbers that act as labels
hair color, species, jersey number, SSN
examples of qualitative data
Quantitative data
numerical in measurements
age, number of pets, time and temp
examples of quantitative data
4
how many levels of measurement are there
nominalordinalintervalratio
levels of measurement
nominal ordinal
what levels of measurement are qualitative
intervalratio
what levels of measurement are quantitative
nominal
a non meaningful list
ordinal
list that has meaningful order
interval
time and temp zero means 0 (its still a number)
ratio
measures weight zero does not exist
observationexperiment simulation survey
methods of data collection
observation
observe and measure characteristics of a sample hands off
experiment
treatment is applied to a sample group and responses are measured hands on
simulation
uses mathematical or physical models to reproduce desired conditions
crash test danger
when are simulations used
survey
ask ppl questions
census sample
types of survey
census
survey of entire population
sampling
survey of a sample
control randomization replication
3 elements of a good experiment
cofounding variable
experiment cant tell difference betweenterm-37 the effects of different factors on a variable
placebo effect
subject reacts favorably bc they think they're being given a treatment but it was a placebo
Hawthorne effect
ppl change their behavior for an experiment
blinding doubt blinding
how to get rid of bias
blinding
subjects dont know which group they are in
double blinding
neither the subjects nor the experimenter knows which group is which
randomization
The process of randomly assigning subjects to different treatment groups
completely randomized design
subjects are assigned groups through random selection
randomization block design
divide group w similar characteristics then randomly assign subjects to treatments and control groups
matched pair designs
subjects are paired accordingly to similarities then one is placed in treatment and other in control group
replication
repetition of an experiment using a large group of subjects
data is more reliablemore confident
benefits of using a large sample size
bias
a systematic error introduced into sampling or testing by selecting or encouraging one outcome or answer over others
biased question biased sample
2 types of bias
bias question
What can I get you for dessert this evening?
bias samples
sample that doesnt rep a whole population
Sampling Techniques
Simple random samplestratisfiedcluster systematicconvenience
simple random sample
every member of a population is assigned a number, then numbers are randomly,ly selected
stratified
population is divided into a strata them random samples are taken from each strata few from each
strata
subset of a population
grades
example of strata
cluster
population is divided into strata but then randomly select some groups and choose everyone all from few
systematic
each member of a population is assigned a number then members are chosen at random intervals every 10th, 5th, 3rd
convenience
compromised of available members of the population, select people the are conveinent