1 / 13

Sampling

Sampling. Planned ways of selecting study participants. Sampling. Today we will discuss:. Purpose of sampling Types of samples Nonprobability Convenience Purposive Quota Snowball Probability Elements Terms Sampling error Central Limit Theorem and Confidence Intervals

lara-hull
Download Presentation

Sampling

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Sampling Planned ways of selecting study participants

  2. Sampling Today we will discuss: • Purpose of sampling • Types of samples • Nonprobability • Convenience • Purposive • Quota • Snowball • Probability • Elements • Terms • Sampling error • Central Limit Theorem and Confidence Intervals • Types of Probability Samples • GSS

  3. Basic Terms • Sampling  • Refers to planned ways of selecting units from the larger population. • Population • the collection of all elements (either known or unknown) from which a sample is drawn. In a probability sample this includes all of the elements in the sampling frame. • Sample  • A set of selected units (subjects) for study drawn according to some principle of sampling. • Unit  • the elements in a sample • Sampling Frame  • A list of all the elements in a population from which a probability sample may be drawn

  4. Purpose of Sampling • Representativeness • That the smaller sample that is studied represents something else. Since probability samples use scientific methods to select from the population by (possible universe of subjects), they represent the larger whole. • Selecting the subjects for the study in such a wasy that they will not bias the findings. • This is done in experiments through randomizing subjects into control vs. experimental groups. • Inference/Generalization • When the researcher selects subjects according to rules of probability (know the probability of each subject being selected into the sample), then can generalize or make inferences from the findings to the population as a whole. • Making Inference is primary goal of probability sampling.

  5. Nonprobability Samples • Refers to selection of subjects/units in a study that do not follow the rules of probability. • Used when, for whatever reason: • no sampling frame can be constructed. Cannot get a list of the elements of the population of interest. • Rational for using: • The best sample that can be designed for your study. • Commonly used in exploratory studies and qualitative studies.

  6. Types of Nonprobability Samples • Convenience Samples • A sample that is composed of subjects that are available and willing to participate. • Important considerations • Should be representative of the range of potential subjects in the population. • Should not include friends or relatives. • Purposive or Judgmental Samples • The subjects selected seem to have the most common characteristics that meets the studies needs. • This type of study figures out where people like those needed for the study are likely to be found and then tries to study them.

  7. Types of Nonprobability Samples • Quota Samples • Attempts are made to select subjects for subsamples from clearly defined groups. • Steps • Define groups of interest are defined • The size of each group is specified. • Individuals are selected to fill the quota for each subsample wherever they can be found. • Why not a probability sample? • No sampling frame is available or established from which subjects are selected • Establishing the subgroup sizes gives no information about the probability of each subject to be included in the study.

  8. Types of Nonprobability Samples • Snowball Samples • Find a few subjects who have the major quality/characteristic you seek for your study. Include these individuals in your study. • Ask these individuals for names of others whom they would they could refer and who posses that quality/characteristic of interest. • Repeat Again and again until have a sample of the desired size. • Population that is difficult to identify and gain access to.

  9. Elements of Probability Studies • Sampling frame • A list of all units in the population • Sample size • The number of units that will be selected into the study from the population. • Depends on the • homogeneity/heterogeneity of the population • number of variables in study • size of population – sampling ratio  the proportion of the population that will be selected = n/N • Goal is to reduce standard error • A statistical method showing how closely a sample statistic represents its population parameter. • Sampling Method: SRS, Systematic, Stratified, Multistage,PPS. • Weights to correct for oversampling of subgroups.

  10. Simple and Systematic Random Samples • Each and every element has an equal probability of being included in the sample. • Steps for creating a SRS • Develop a list of all members in your sample frame. • List should not be ordered (except maybe alpha) • Number consecutively each element of the list • Decide on our sample size • Select Sample Members/Subjects • For SRSRandomly select members to be included in sample (Eg: random number lists and start anywhere) • For Systematic start with random unit and then pick every nth unit (Based on sampling interval= population/sample size) until have desired number of subjects.

  11. Stratified Samples • Sampling technique in which a sample frame is divided into strata (groups) and the sample is drawn from each strata using SRS or systematic sampling techniques. • Purpose is to assure that all groups of interest (Strata) are included in the sample in proportion to their representation in the population.

  12. Steps for creating a Stratified Random Sample • Develop a sampling frame • Decide sample size • Determine the percentage of people in each strata of interest in the population • Sort the list of the sampling frame units into separate groups for every combination of strata • Determine how many units will need from each strata to end up with same proportions in sample as in population  calculate sampling ratio=n/N • Randomly select from each strata until get number of subjects needed for each strata. • Use SRS • Use systematic sampling– easier method

  13. Multistage Probability Sample • Sampling is done in a series of stages beginning with a heterogeneous cluster (organizations or geographical areas) and later stages in which members of the larger clusters are selected. In every stage random methods of selection must be employed. • Select Primary Sampling Unit – larger heterogeneous cluster • Select Secondary Sampling Unit – select members from PSU • The General Social Survey • Uses a multistage area probability sample. • Each stage is based on principles of probability proportional to size (PPS) • A method used to select strata proportional to their size within a cluster • See p 162

More Related