Unit 3 test Flashcards
What is an experiment?
Different conditions (treatments) are imposed upon subjects
What may help to predict a change in the response variable?
Explanatory Variable
What is used to measure the outcome of a study?
Response Variable
Simple Random Sampling
Every group of a given size has an equal chance of being chosen
Advantages of SRS
Unbiased, easy to explain
Disadvantages of SRS
In certain cases, it can be difficult to implement and it may not be as precise as other methods
Advantages of Cluster Random Sampling
Unbiased and easy to perform, practical and affordable
Ideal situation for Cluster Sampling
You want it to be heterogeneous within groups, groups are similar to each other
Stratified Random Sampling
population is divided into strata, based on a similar characteristic. SRS within each stratum is taken. Selected individuals are combined into larger samples.
Advantages of Stratified Sampling
It represents the population by drawing some from each layer, reducing variability in the results that could arise because of differences among layers.
Bias
Systematic tendency to favour certain responses over others.
Examples of Selection/ Sampling Bias
Response Bias
Examples of Survey Bias
Voluntary Response Bias, Nonresponse Bias, Convenience Sampling, Undercoverage Bias,
What is Voluntary Response Bias?
When a sample is comprised entirely of volunteers or people who choose to participate.
Why is Voluntary Response Bias is bad?
Asking people to be part of a sample will usually attract people with strong opinions, detract people who either don’t care or have something to hide which no longer accurately reflects the truth of the overall population.
What is Convenience Bias?
One that is selected quickly out of ease for the researcher.
Why is Convenience Sampling bad?
These types of samples introduce potential for bias because they do not use chance to select individuals.
What is Undercoverage Bias?
Certain groups were excluded from the sample and when part of the population has a reduced chance.
What is Response Bias?
Any problem in the data collection process that would result in data that is untruthful or incorrect.
What is Non-response Bias?
Individuals chosen from the sample for whom data cannot be obtained and may differ from those for whom data can be obtained.
Factor
The explanatory variable to be manipulated
Levels
various settings for a factor (specific values experimenter chooses for factor)
Treatment
Actual Assignments given to subjects
Experimental Units
The “unit” receiving the assigned treatment and the “unit” measured at the end.
Subjects
If experimental units are human
Retrospective
Historical Data
Benefits of Random Assignment
Helps to balance the effects of uncontrolled variables so that the differences in responses can be attributed to the treatment and nothing else.
Benefits of Blocking
By blocking, we isolate the variability attributable to the difference between blocks so that we can see the difference caused by treatments more closely.
Statistically Significant
A difference in the response variable so large that it cannot be attributed to natural sampling variability
Purpose of Control Group
Assigned to a baseline treatment level, typically either default treatment (provides a basis for comparison).
