Unit 4: Collecting Data Flashcards
Population
The entire group of individuals that the information is supposed to represent.
Sample
The subset of the population used to gather information.
Sample Survey
A study that asks questions of a sample in the hope of learning about the entire population.
Convenience Sample
Sampling in which individuals are chosen based on how easy they are to reach.
Bias
Using a method that statistically favors certain outcomes.
Voluntary Response Sample
Sampling in which individuals choose themselves by responding to a general appeal. Ex: A receipt’s survey at the bottom.
Simple Random Sample (SRS)
Sampling in which each individual from a population has an equal chance of being in the selected sample via random picking.
Stratified Random Sample
Sampling in which the population is divided into groups of individuals that have similar characteristics (called strata), then SRS is drawn from each stratum & combined to make the sample.
Cluster Sample
Groups of individuals that mimic the characteristics of the entire population (called clusters).
Systematic Sample
A sample in which individuals are chosen systematically from a sampling frame. The first number must be random. Ex: Every 10th person.
Multi-Stage Sampling
Sampling in which you combine several different types of sampling methods.
Inference
Drawing conclusions about a population based on the results from a sample.
Margin of Error
How far we expect estimates to vary from the truth, at most.
Sampling Errors
Errors that are the result of the sampling process.
Sampling Frame
A list of individuals from which the sample is drawn.
Undercoverage
The inability to be able to represent the entire population due to certain groups being unable to be represented. Ex: Sampling via phone numbers excludes people that do not own phones.
Nonsampling Errors
Errors that result after the sample is selected.
Nonresponse
An individual chosen for a sample can’t be contacted or refuses to participate.
Response Bias
A systematic pattern of incorrect responses in a sample survey. Similar to Desirability Bias.
Observational Study
A study in which members of a survey are surveyed without being affected in any way.
Experiment
A controlled study conducted to test a hypothesis by manipulating one or more independent variables and observing the effects on one or more dependent variables.
Lurking Variable
Variables that are not among the explanatory or response variables in a study but may influence the response variable.
Confounding
Refers to two variables that are associated in such a way that effects on a response variable cannot be distinguished from each other.
Treatments
A specific condition applied to individuals in an experiment.
Experimental Units
Smallest collection of individuals to which treatments are applied.
Subjects
When the experimental units are people, they are referred to as subjects.
Factors
Explanatory variables in an experiment
Levels
The specific values of each of the factors.
Random Assignment
Experimental units that are assigned to treatment groups at random.
Completely Randomized Design
Treatments are assigned to all experimental units completely by chance.
Control
To ensure that the only systematic difference between treatment groups is the treatment itself.
Replication
Refers to using enough experimental units so that differences in the effect of the treatments can be distinguished from chance differences between the groups.
Placebo Effect
The response to a dummy/fake treatment.
Single Blind Experiment
An experiment in which only the subject doesn’t know which treatment they received OR the subject knows but not the response variable.
Double Blind Experiment
An experiment in which neither the subject nor the experimenter know what group/treatment they’re receiving.
Statistically Significant
If the results from an experiment are statistically significant, it means that an observed effect is so large that is would rarely occur by chance.
Randomized Block Selection
A group of similar experimental units, similar to stratified sampling.
Matched Pairs
Compares two treatments by matching pairs of similar experimental units. Often times, this is accomplished by testing two treatments on the same subjects.
