-
The entire set of subjects that are of interest to the researcher
Population
-
A carefully selected subset of the population that refelcts the composition of that population
Sample
-
A statistical value that indicates differences in results found in the sample compared to the population from which the sample was drawn
Sampling error
-
In quantitative study the sampling strategy is aimed at maximizing the potential for generalization or the ability to apply the findings to larger groups
-
A condition that occurs when subjects are selected for a study or assigned in groups in a way that is not impartial. This may pose a threat to the validity of the study. Controlled almost exclusively by a sound sampling strategy.
Selection bias
-
Having an adequate number of subjects provides the study with power
-
The ability to detect effects
Power
-
The first step of a sampling strategy is:
To clearly define the population of interest
-
The potential participants who meet the definition of the populatino and are accessible to the researcher
Sampling frame
-
The selection strategy involves making decisions about how subjects will be recruited, selected, and if appropirate, assigned to groups.
-
The goal of the selection strategy is to:
- Prevent bias
- Support validity of the study
- Enhance credibility of the results
-
A homogenous sample is one in which the subjects are very similar in characteristics, and it makes generalizing to other populations difficult
-
In qualitative research, the goal of selection strategy is credibility rather than generalizability so selection methods are purposive
-
Guidelines for choosing subjects with a predetermined set of characteristics that include major factors important to the research question
Inclusion criteria
-
Characteristics that eliminate a potential subject from the study.
Exclusion criteria
-
A technique used in qualitative reserarch in which the subjects are selected because they possess certain characteristics that enhance the credibility of the study
Purposive selection
-
Quantitative samples are best when they are selected and assigned to groups randomly
-
A sampling process used in quantitative research in which every member of the available population has an equal probability of being selected for the sample
Probability or random sampling
-
The only way to be sure a sample represents a population is if it incorporates two essential criteria:
- Each member of the pop has an equal probability of selection for the sample
- Each subject selection is an independent event
-
A condition that occurs when the selection of one subject has no influence on selection of other subjects; each member of the population has exactly the same chance of being in the sample
Independence
-
Samples are structured so that important characteristics are evenly distributed across all groups
Stratified random sampling
-
Sample in which the first subject is drawn randomly and the remaining subjects are selected at predetermined intervals
Systematic random sampling
-
Researcher randomly selects entire groups and then randomly selects subjects from only those groups
Cluster random sampling
-
A nonprobability method of selecting a sample that includes subjects who are available conveniently to the researcher
Convenience sampling
-
Sampling is used when a table of random numbers is used to select subjects rom the sampling frame.
Simple random sampling
-
A nonprobability sampling method that relies on referrals from the initial subjects to recruit additional subjects
Snowball or referral sampling
-
The definition of the major entity or subject that will be analyzed in the study
Unit of analysis
-
The standard for a sample size in a qualitative study is:
The achievement of saturation
-
Saturation has been achieved when the researcher concludes that responses are repetitive, and no new info is being generated.
-
An analysis that indicates how large a sample is needed to adequately detect a difference in the outcome variable
Power
-
Big effects are easier to see in the data, just as a large object is easier to see than a small one
-
This error occurs when there is a difference between groups but the researcher does not detect it. (The intervention works but researchers do not detect it)
Type II error
-
The ability to generalize the findings from a research study to other populations, places, and situations.
External validity
-
The link between finding knowledge through research and using knowledge in practice
External validity
-
A type of external validity where the findings can be generalized and applied to other settings
Ecological validity
-
A type of external validity where the findings can be generalized and applied to other subjects
Population validity
-
Subjects who have a personal interest in the study are more likely to complete it. A combination of personal enthusiasm and nurturing by the researcher is often necessary to keep subjects in a study.
-
Samples with at least 80% power are desirable
-
If statistical significance was achieved, the sample had sufficient power
|
|