Chapter 12 Pwpt Flashcards
What is sampling?
Process of selecting representative units (subjects) of a population
Description of sampling strategy is usually found in the _______ _________.
Methods section
___________ and ___________ __________ must be described by the researcher
Exclusion and inclusion criteria
What is important to consider about sampling?
How does the sampling strategy limit the potential effect of extraneous variables or bias to explain the results of the study?
A sample should be _______________ because it has ______________________________.
Representative; the same key characteristics as the entire population
What is the purpose of sampling?
To be more efficient; it is non cost effective, or even feasible to study the whole population
Two types of sampling are _______ or ________ and _________ or _________.
Nonprobablility or purposive sampling
Probability or random sampling
Non probability (purposive) sampling is where?
Inclusion in study sample is NOT random
Less generalizable
Less representative
Types of nonprobability (purposive) sampling are?
Convenience
Quota
Matching
Networking sampling (snowball sampling)
Probability (random) sampling does what?
Uses randomization to include include in sample
More generalizable
More representative
Types of probability (random) sampling are?
Simple random sampling
Stratified random sampling
Cluster sampling
Remember that random sampling DOES NOT equal
Random assignment to group
Sampling threats of internal validity
Selection effect - bias in selecting
Attrition and mortality- loss of subjects
Sampling threats of external validity
Generalizability to other populations or settings
Critical appraisal of nonprobability samples
What motivated some of the people to participate and others not to participate (self-selection)?
What kind of data would have been obtained if nonparticipants had also responded?
How representative are the people who did participate in relation to the population of interest?
What kind of confidence can you have in the evidence provided by the findings, based on the sampling strategy?
Sample size must consider?
Type of design and sample procedure
Formula used for estimating sample size (power analysis)
Heterogeneity of attributes investigated
How often the phenomenon occurs in the population
Projected cost
The power analysis is used to determine the…?
Number for each group of a study
To perform a power analysis, what needs to be determined?
Effect size, level of significance, how much power of detection is desired
As we increase power to avoid a ___________, we get an increased chance of making a __________.
Type 1 error; type 2 error
How much power of detection is desired? Set to avoid a type 2 error. What is a type 2 error?
When the results are not statistically significant, when there is a real relationship between variables or a difference between groups
Level of significance: set by the researcher to avoid a type 1 error. What is a type one error?
When the statistical analysis determines that there is a difference between groups or correlation between variables when there is not.
When might an entire population most likely be used in a research study?
When the population size is very small
What type of sampling includes using the internet and social networking to locate samples that are otherwise difficult or impossible to locate?
Snowball sampling