AP Statistics: Unit 1
Treatment
A specific condition applied to individuals in an experiment
Statistically Significant
An observed effect so large that it would rarely occur by chance
Simple Random Sample (SRS)
Chosen in a way to where all individuals have an equal chance to be selected (Ex. 20 people out of 200 having an equal chance to be selected is an SRS)
Sample Survey
Collects data from a sample
Census
Collects data from every individual in a population
Matched Pairs Design
Compares two treatments by matching pairs of similar experimental units
Completely Randomized Design
Experimental units have an equal chance of receiving any treatment from experiments (Chance does not guarantee a completely randomized design)
Block
Group of experimental units known to be similar (Will also react similarly to treatments)
Clusters
Groups of individuals similar to each other (within a population) (Individuals sampled in a cluster aren't necessarily similar to each other; ideally, a cluster mimics a population on a smaller scale)
Strata
Groups of similar individuals (Chosen based on knowledge prior to taking sample)
Explanatory Variable
Help explain or predict changes in a response variable
Subjects
Human experimental units
Experiment
Imposes some treatment on individuals to measure their responses (Only source of fully convincing data)
Scope of Inference (For Random Samples)
Inference about population
Control (Principles of Experimental Design)
Keep other variables that might affect the response the same for all groups (Control is important because 1: You will be able to view the difference in response variables 2: It will reduce variability in the response variable) (Provides a baseline for comparing the effects of other treatments)
Random Sampling
Letting individuals in a population be chosen by chance (Uses a chance process to determine which members of a population are included in a sample)
Response Variable
Measures an outcome of a study
Double-Blind Experiment
Neither a subject nor researcher know who receives a treatment
Nonresponse vs Voluntary Response
Nonresponse occurs after a sample has been selected, individuals in a voluntary response opt to take part
Observational Study
Observes individuals and measures variables of interest but does not attempt to influence the responses (Poor way to gauge the effect that variables have)
Confounding
Occurs when 2+ correlated variables have an indistinguishable effect on response variables
Nonresponse (Sampling Issue)
Occurs when an individual cannot be contacted or refuses to participate (Can result in bias; Individuals cannot give data)
Undercoverage (Sampling Issue)
Occurs when members of a population cannot be chosen in a sample (Can result in bias; Not everyone will be able to give their data)
Randomized Block Design
Random assignment of experimental units to treatments is carried out separately within each block (Allows researcher to account for variation in the response that is due to the blocking variable. This makes it easier to determine if one treatment is more effective than another)
Experiment vs Observational Study
Researchers make no changes during observational studies, only during experiments
Placebo Effect
Response to a dummy treatment
Stratified Random Sample
Samples individuals by classifying the population into groups of similar individuals (called strata)
Cluster Sampling
Samples individuals into groups based on proximity (called clusters) (An SRS is then conducted using the clusters)
Convenience Sample (Bad Sampling)
Sampling from individuals who are easy to reach (Can result in bias; favors some outcomes over others meaning sample may not be representative)
Voluntary Response Sample (Bad Sampling)
Sampling from individuals who chose to respond (Can result in bias; voluntary sampling tends to attract people who feel strongly about an issue)
Experimental Units
Smallest collection of individuals to which treatments are applied
Strata vs Clusters
Sratas sample similar individuals, clusters sample individuals close to each other
Sample
Subset of population
Population
The entire group of individuals we want information about
Inference
The process of drawing conclusions about a population on the basis of sample data
Comparative Experiment (Principles of Experimental Design)
Use a design that compares two or more treatments (Produces data that can give good evidence for a cause-and-effect relationship between explanatory and response variables)
Replication (Principles of Experimental Design)
Using enough experimental units in each group so that differences in the effects can be distinguished (Control groups provide a baseline for comparing the effects of other treatments)
Bias
When data from a study is consistently underestimated or overestimated
Random Assignment (Principles of Experimental Design)
When experimental unitss are assigned treatments using a chance process
Single-Blind Experiment
When subjects don't know who is in the group receiving treatment and what they are receiving (Only researcher knows)