Chapter 5: Sampling Methods

Chapter 5: Sampling Methods New Vocabulary: Probability sampling: techniques in which the probability of selecting each subject is known (randomized) Non-probability sampling: methods in which subjects do not have the same chance of being selected for participation (not randomized) Sampling error: differences between the sample and population due to chance Sampling bias: a systematic error made in the selection of a nonrandom sample Cluster sampling: probability of using a group or unit rather than an individual Convenience sampling: non-probability sampling that collects data only from available groups Quota sampling: non-probability sampling in which you select the proportions from different subgroups without random sampling Simple random sampling: probability sampling in which every subject in a population has the same chance of being selected Stratified sampling: probability sampling divided the population into subsamples according to a characteristic of interest and then randomly selects from those subgroups Systematic sampling: probability sampling involving the selection of subjects according to a standardized rule Sampling Methods -​ The population is the whole group that you intend to research -​ A sample should be representative of this population, and should reflect the characteristics of the population, such as if 50% of the population are males over the age of 60, your sample should also have 50% being males over 60 -​ The sampling method you use consists of the process of selecting the subjects from the population you are studying -​ Can be probability sampling -​ Or non probability sampling Probability Sampling -​ Probability sampling consists of techniques in which the probability of selecting each subject is known -​ This is typically not feasible in large populations as it takes a lot of resources -​ Examples of probability sampling are: -​ Simple random sampling: in which every subject in a population has an equal chance of being selected. -​ This is the most preferred method of sampling, although it may not be plausible with large populations -​ Stratified sampling: divides the population into subsamples according to a characteristic of interest and then randomly selects the sample from these subgroups -​ This ensures the representativeness of the characteristic -​ Cluster sampling: randomly selects a group or unit rather than an individual. It is used when it is difficult to find a list of the entire population -​ Cluster sampling improves efficiency and reduces cost, but it may increase the variance as compared to other methods. Sampling Error vs. Sampling Bias -​ Sampling error is due to chance, and cannot be prevented. It is when there is differences between the sample and the population that is out of your control -​ Sampling bias is a systematic error made in the sample selection that results in a nonrandom sample Sampling Distributions -​ A sampling distribution consists of all the possible values of a statistic from all the possible samples of a given population -​ If your sample size is large enough (usually more than 30), the distribution will be normally distributed, even when the population is not. -​ When a population is not distributed normally, you may need to use other methods to analyze it Nonprobability Sampling -​ Nonprobability sampling consists of methods in which subjects do not have the same chance of being selected for participation, and it is not randomized. -​ Examples of nonprobability sampling are either quantitative or qualitative: -​ Convenience sampling: collecting data from readily available groups -​ Quota sampling: you select the proportions of the sample for different subgroups like stratified sampling, but it is not randomized -​ Eg. after deciding on the proportions of the sample you collect subjects continuously until you have 50 day-shift subjects, 30 evening-shift subjects, and 20-night shift subjects Inclusion and Exclusion Criteria -​ Inclusion criteria makes up the list of characteristics a subject must have to be eligible to participate in your study (or your target population) -​ Exclusion criteria are criteria or characteristics that eliminate a subject from being eligible to participate in your study Sample Size -​ The larger your sample size, the more likely you are to find a difference that really exists -​ This may be referred to as the “power” of the study -​ Larger samples have greater power, whereas smaller samples have less power