35 terms

# AP Statistics (POD 2)

Ch. 2 - Collecting Data Sensibly. An Introduction to Statistics and Data Analysis
###### PLAY
Observational Study
Investigator observes characteristics of a subset an existing population(s) to draw conclusions relating to the differences between two or more populations.
Survey
Interviewer seeks information from a respondent by engaging in a special type of conversation
Experiment
Investigator observes how a response variable behaves when other explanatory variables (aka factors) are manipulated; factors should be determined by random assignment.
Factors
Another name for the variables that researchers manipulate to create experimental conditions.
Confounding Variable
A variable whose effect on the response variable cannot be distinguished from the effects of the treatment; if one is present it is impossible to draw cause and effect conclusions
Random Sampling
Technique that allows every member of a population an equal probability of being selected for the sample; allows generalizations from a sample to a population.
Census
Data is collected on every member of a population
Selection Bias
A sample systematically excludes part of the population of interest; also called undercoverage
Measurement Bias
Observed values systematically differ from the actual values; may occur due to improper calibration or leading surveying questions or techniques; also known as response bias
Nonresponse Bias
Responses are not obtained from all individuals selected for inclusion in a sample; limited when the response rate is high and/or serious efforts to follow up with non-respondents are used
Simple Random Sample (SRS)
Sample chosen using a method that ensures that each different possible SAMPLE of the desired SIZE has an equal chance of being chosen
Sampling Frame
A list that is used to select a random sample; typically a random number table or generator is used to select the sample.
Sampling with Replacement
Individuals can potentially be selected more than one time since they are returned to the pool of potential subjects after each selection
Sampling without Replacement
Individuals can only be selected once since they are not returned to the pool of potential subjects after each selection
Stratified Sampling
A method of dividing a population into non-overlapping homogenous subgroups and then selecting a SRS from EACH subgroup
Cluster Sampling
A method dividing a population into non-overlapping heterogeneous subgroups and then random selecting entire subgroups; used to save time and money
Systematic Sampling
A procedure that selects one of the first k individuals at random and then every k-th individual in the sequence is included in the sample; list of items must contain no repeating patterns
Voluntary Response Sampling
Sample suffers from selection bias because only people with strong opinions choose to participate in the survey
Convenience Sampling
Sample suffers from selection bias because subjects are chosen without regard to whether they are representative of a population
Experimental Condition
Any particular combination of values for the explanatory variables; aka a treatment
Design
The overall plan for conducting an experiment; should limit confounding variables and extraneous factors to achieve conclusive results
Extraneous Factor
A variable that is not of interest in the current study but is thought to affect the response variable.
Blocking
Filters out extraneous factors , by creating subgroups that are homogenous with respect to those factors, and then randomly assigns subjects from each subgroup to all treatment and/or control groups
Confounded
When two or more factors effects on the response varaible cannot be distinguished from one another.
Blocks
Homogenous subgroups used to ensure that all experimental groups are similar in regards to one or more extraneous factor(s)
Randomization
Used to ensure extraneous factors that can not be controlled through blocking or direct control are equally likely to occur in any experimental group
Replication
The design strategy of making multiple observations for each experimental condition.
Control Group
The group in an experiment that does not receive the experimental treatment; receives either the current standard treatment or no treatment at all
Placebo
An object that is identical in appearance, taste, feel etc. to the treatment received by the treatment group, except that it contains no active ingredients.
Placebo effect
Experimental results caused by expectations alone; any effect on behavior caused by the administration of an inert condition, which is assumed to be an active agent
Single Blind
An experiment in which subjects do not know the treatment that they have received but the individuals measuring the response do know or occasionally vice-versa.
Double Blind
An experiment where neither the subjects or the individuals who measure the response know which treatment was received.
Experimental Unit
The smallest unit to which the treatments (which are assigned at random) are applied.
Strata
Homogenous subgroups used to ensure that all characteristics of a population are represented by a random sample
Direct Control
Filters out extraneous factors by ensuring that experimental conditions are constant for all subjects in regard to these factors

#### Flickr Creative Commons Images

Some images used in this set are licensed under the Creative Commons through Flickr.com.
Click to see the original works with their full license.