Karteikarten
Chapter 2: Sampling
Karten in diesem Stapel (28)
- 1Samplinga representative of the population that is a group or collection
- 2Biasexists wen some individuals are systematically favored over other
(three types) - 3Response Biasexists when the respondents give inaccurate information (seen in cases where questions involve legal or social issues) or if the interviewer influences the subject to respond in a certain way to the wording of the questions
- 4Non response Biasexists when individuals chosen for the sample cannot be contacted or fail or refuse or respond
- 5Selection Biasexists when the selection of individuals, groups, or data is not properly randomized, resulting in a sample that is not representative of the entire population
- 6Simple random Samplinga list of all possible individuals in the population is made and 'n' subjects are chosen in a way that every set of 'n' subjects has an equal chance of being selected
- 7Stratatwo or more groups of similar subjects naturally divided from the population
- 8Stratified random samplinga simple random samples that are chosen from stratum groups which are combined to give a complete sample
- 9Multistage random samplingthe population is divided into several clusters which are further divided into smaller clusters. Simple random sampling is used to choose several of the clusters and sub-clusters. Then, for each of the smallest sub-clusters, simple random sampling is used to select a group of individuals that will be measured
- 10Haphazard samplinga method that involves selecting a sample by some convenient mechanism that does not involve randomization
- ex. mall surveys - 11Volunteer response samplinga method that involves sampling people to volunteer to be a part of a study
ex. internet surveys - 12Experimental unitsthe subjects (individuals, units) on which the measurements are made
- 13Factorexplanatory variable that causes a change in the response
- 14Levelsspecific values of a factor
- 15Comparisonall experiments must involve a comparison between two or more groups
- 16Treatment group(s)the group of experimental units who receive the treatment (there can be multiple)
- 17Control groupthe group of the experimental units who do not receive the treatment (multiple control groups is not required)
- 18Placeboa fake treatment that is given to the experimental units in the control group which prevents them from knowing if they received the treatment or not
- 19Replicationeach group should contain more than one experimental unit, which should reduce the chance variation in the results
- 20Randomizationthe experimental units are randomly chosen to be in the experiment then randomly assigned to the treatment or control groups
- 21Blindingthe experimental units do not know what they have been assigned
- 22Double-blindingthe experimental units and those conducting the experiment do not know which group the experimental units have been assigned
- 23Confoundingthe existence of a factor other than the treatment that makes treatment and control groups different
- 24Observational studyrandomly select experimental units to be in the experiment; units are only observed
- 25Controlled experimenta designed procedure in which experimental units are randomly selected to be in the experiment and treatment of experimental and control groups is assigned
- 26Blocka group of individuals that are known prior to the experiment to be similar in some way that is expected to affect the response to the treatments
- 27Block designfor each block, subjects are randomly assigned to the treatments
- 28Matched pair designa design when two treatments are compared using a pair of subjects that are a close match
Informationen
- Beschreibung
- Data analysis process, bias, sampling
- Erstellt von
- Erin Vales
- Aufrufe
- 7
- Bewertung