Chapter 4-Collecting Data

Census

Collects data from every individual in the population

Sample

A subset of individuals selected from the population to gather information about the whole

Sample Survey

A study that collects data from a sample to draw conclusions about the population

Voluntary Response Sampling

Individuals choose themselves by responding to an invitation; often leads to strong or biased opinions

Random Sampling

Using chance to select a sample

Simple Random Sample (SRS)

Every possible sample of a given size has an equal chance of being selected

Sampling Without Replacement

Once an individual is selected, it is not put back into the pool for possible reselection

Stratified Random Sampling

Dividing the population into similar groups (strata) and taking an SRS from each group

Strata

Non-overlapping groups of individuals that are similar in some way important to the study

Cluster Sampling

Dividing the population into clusters, randomly selecting clusters, and including all individuals in chosen clusters

Systematic Random Sampling

Selecting every kth individual after randomly choosing a starting point

Undercoverage

Occurs when some people in the population cannot or are less likely to be chosen

Nonresponse

Occurs when selected individuals cannot be contacted or refuse to respond

Confounding

When two variables’ effects on a response variable cannot be distinguished from each other

Retrospective Study

An observational study that looks back at past data

Prospective Study

An observational study that follows subjects into the future

Experimental Units

The individuals or objects to which treatments are applied

Treatments

The conditions applied to experimental units in an experiment

Levels (of a factor)

The specific values of the explanatory variables used to form treatments

Comparison

The experimental design principle of comparing two or more treatments

Random Assignment

Using chance to assign experimental units to treatments

Replication

Using enough experimental units so differences between groups can be distinguished from chance variation

Completely Randomized Design

An experimental design where all experimental units are assigned to treatments completely at random

Block

A group of experimental units that are similar in a way that may affect responses

Randomized Block Design

An experiment that randomly assigns treatments within each block of similar units

Matched Pairs Design

A type of randomized block design comparing two treatments using matched pairs of similar units, or by giving both treatments to each subject in random order

Sampling variability

different random samples of the same size from the same population produce different estimates