Chapter 2 Flashcards
Define observational study.
If the investigator observes characteristics of a subset of the members of one or more existing populations.
What is the goal of an observational study?
To draw conclusions about the corresponding population or about differences between two or more populations.
Define experiment
When an investigator observes how a response variable behaves when the researcher manipulates one or more of explanatory variables.
What is the goal of an experiment?
To determine the effect of the manipulated factors on the response variable.
What conclusions is it impossible to draw from an observational study?
Cause and effect conclusions
Define confounding variable.
A variable that is both related to both group membership and the response variable of interest in a research study.
What is bias in sampling?
The tendency for samples to differ from the corresponding population in some systematic way.
What are the most common types of bias?
Selection bias, measurement or response bias, and no response bias.
When does selection bias occur?
When the sample is selected and excludes part of the population of interest.
When does measurement or response bias occur?
When the method of observation tends to produce values that systematically differ from the true value in some way.
When does no response bias occur?
When responses are not obtained from all individuals selected for inclusion in the sample.
Define Simple Random Sample.
Every individual in every subset of the population has an equal chance of getting picked.
Give an example of SRS.
Putting the names of everyone in a class in a hat. Could result with all boys or all girls.
Define sampling frame.
A list of the objects or individuals in the population.
What is sampling with replacement?
When each item that is picked is put back into the population with a chance of being picked again.
What is sampling without replacement?
When an item selected isn’t placed back into the population.
Define stratified sampling.
The population is divided into a set of no overlapping subgroups.
Define strata.
Subgroups
Define cluster.
Non overlapping subgroups
Define cluster sampling.
When the tester divides the population into subgroups, then randomly chooses a subgroup and tests all the subjects in that subgroup.
Define systematic sampling.
Picking every nth number from a list.
What is convenience sampling?
When you use an easily available or convenient group to form a sample.
Define voluntary sampling.
When individuals volunteer to be in the test.
Define experiment.
A planned intervention undertaken t observe the effects of one or more variables. Must include a treatment.
Define design.
The overall plan for conducting an experiment.
Define extraneous factor.
A factor that’s not of interest in the current study but is thought to effect the response.
Define blocking.
Groups that are similar with respect to the blocking factors.
Define confounded factors.
When the effect of two factors can’t be distinguished in the response.
Define randomization.
Random assignment to experiment groups
Define replication.
The design strategy of making multiple observations for each experimental condition.
Define direct control.
Holding extraneous factors constant so that this effects are not confounded with those of the experimental conditions.
Define placebo.
Something that is identical to the treatment, but not the treatment.
Ex.
A sugar pill instead of a drug.
Define double blind
When neither the subjects nor the individuals who measure the response know which group is teeatment.
Define single blind.
When only the subjects don’t know of they are in the treatment group or not.
Define experimental unit.
The smallest unit to which the treatment can be applied.
Define survey.
A voluntary encounter between strangers where an interviewer asks the interviewee questions.
What are the respondent’s tasks?
- Comprehension
- Retrieval from memory
- Reporting the response