ap stats - experiments exam

population

entire group of individuals we want infor about

census

collects data from every individual in population

sample

subset of individuals we actually collect data from

convenience sample

choosing individuals who are easy to reach

bias

using a method that favors some outcomes over others

voluntary response sample

people who choose themselves by responding to general invitations

random sample

using a chance process (rolling a die) to determine who in the population is included in the sample

simple randomized sample (SRS)

of size n is chosen in such a way that every group of n individuals in the population has an equal chance to be selected as a sample

stratified random sample

1. classify population into groups of similar individuals (strata)


1. individuals within stratum are similar
2. choose a separate SRS in each stratum and combine these SRS to form the sample


1. differences across strata

cluster sample

1. classify population into groups of individuals located near each other (clusters)
2. choose an SRS of the clusters with all individuals in clusters included in the sample


1. each cluster represents variability in the population

undercoverage

when some members of the population cannot be chosen in a sample (i.e. not everyone is in a phone book)

nonresponse bias

an individual chosen for the sample cannot be contacted or refuses to participate

* ONLY AFTER SAMPLE SIZE HAS BEEN CHOSEN
* no nonresponse bias in voluntary random sample because every individual has opted to take part

response bias

systematic pattern of incorrect responses in a sample survey

observational study

i.e. sample surveys

* observes individuals
* measures variables of interest
* does not attempt to influence the responses

experiment

* deliberately imposes some treatment on individuals to measure their responses
* ONLY EXPERIMENTS: CAUSE AND EFFECT

confounding

two variables are associated in such a way that their effects on a response variable cannot be distinguished from each other (explanatory variable)

treatments (experiments)

if experience has many explanatory variables, a treatment is a combination of specific values of these variables

experimental units

smallest collection of individuals to which treatments are applied (i.e. humans, subjects)

factors

explanatory variables

level

each treatment is formed by combining a specific value

HOW TO EXPERIMENT WELL

1. include a control that does not receive x variable so that we have something to compare with
2. two groups (caffeine and noncaffeine) are as similar as possible and are treated in exactly the same way (except treatment)


1. randomization
2. replication
3. control (not control group)

principles of experimental group

1. COMPARISON: compares 2 or more groups; doesn’t necessarily mean control group
2. RANDOM ASSIGNMENT: use chance to assign experimental units to treatments--helps create roughly equivalent groups of experimental units by balancing the effects of other variables among the treatment groups
3. CONTROL: not control group--keeps variables the same
4. REPLICATION: enough times

completely randomized design

treatments are assigned to all the experimental units completely by chance

* some may include control group that receives inactive treatment or baseline treatment
* groups don’t have to be same size--ideally, they should be tho

block

group or experimental units that are known before the experiment to be similar in some way that is expected to affect the response to the treatments

randomized block design

random assignment of an experimental units to treatments

matched pairs design

randomized blocked experiment in which each block consists of a matching pair of similar experimental units

placebo

substance or other kind of treatment that looks just like a regular treatment or medicine, but it is not--usually an inactive “look-alike” treatment or substance (not a medicine)

the placebo effect

favorable response to a dummy treatment; mant patterns respond favorably to any treatment, even placebo, perhaps bc they trust the doctor

double blind experiment

neither the subject nor those who interact with them and measure the response variable know which treatment a subject received

single blind experiment

1. individuals interacting with the subjects and measuring the response variable don’t know which subjects are receiving the active medication or treatment and which are not
2. the subjects are unaware of which treatments they are receiving but the people interacting with them and measuring the response variable do know

inferencing in experiments

* if individuals were randomly selected, you can make inferences about population
* if individuals are randomly assigned to groups, you can make inferences about cause in effect