Chapter 1 Flashcards
Observation Unit
An object on which a measurement is taken
Target Population
The complete collection of elements we want to study
Sample
Subset of the population
Sampled population
Collection of all possible observation units that might have been chosen in a sample
Sampling unit
Unit that can be selected for a sample
Sampling Frame
A list of sampling units
Bias
Systematic error in sampling
Selection bias
Occurs when the target population doesn’t coincide with the sampled population
OR
When some population units are sampled at a different rate than intended by the investigator
Selection bias examples (6)
Judgement
Convenience
Self-selected sample
Under-coverage
Over-coverage
Non-response
Questionnaire Design (9)
Test the questions before taking the survey
Keep it simple and clear
Define your terms
Relate your questions to the concept of interest
Decide whether to use open or closed questions
Pay attention to response option scales
Avoid double negatives
Ask only one concept per question
Attention to question order (sensitive questions at end)
Sampling Errors
Margin of error
Non-sampling Errors
Selection Bias
Measurement Error
Non-sampling Errors
Selection Bias
Measurement Error
Convenience Bias
sample easy, convienent
Judgment Bias
guessing by estimation
Interpreting ambiguity
Self-selection Bias
Individuals that choose to answer are part of sample
Under-coverage
doesn’t include all observation units
Over coverage
include some not in sample
Non-response
When not all in sample respond
Assumptions of Simple Probability Samples
Sampled population same as target population
Sampling frame is complete
No non-response or missing data
No measurement error
Systematic Sampling what is K
N/n
This is the pattern at which values are chosen from the sample
Why do we choose a stratified sample?
Want to be protected from the possibility of a really bad sample
Want data of known precision from subgroups of the population
May be more convenient to administer and result in lower cost for survey
Often gives more precise (lower variance) for population means
When constructing strata we want?
Statum means to be as different as possible
Variation within strata is small compared to the variability of the population
