Ch 11- Populations and Samples
Probability Sampling
Uses random sampling procedures Selects sample from elements or members of population Types - Simple - Stratified - Cluster - Systematic
Nursing Research Studies
Usually limited to small convenience samples Generalizations to total population difficult Small sample sizes warrant replication studies. Similar results from replication help with generalization
Random Sampling or Random Assignment
Random sampling: Each subject has equal probability of being included. Random assignment: Procedure to ensure that each subject has equal chance
Sample Selection
Representation of the population Method for getting the sample Sample size for the study
Simple Random Sampling
- Type of probability sampling - Importance of this sampling: Equal chance of selection and Independent chance of selection
Disadvantages of Stratified Random Sampling
Requires accurate knowledge of population May be costly to prepare stratified lists Statistics are more complicated
Approaches for Stratified Random Sampling
Proportional stratified sampling: Determine sampling fraction for each stratum Ensure that this stratum is equal Proportion in total population Disproportional stratified sampling: Determine stratum is represented Used when strata are very unequal Note the key word disproportional
Nonprobability Sampling Procedures
Advantages: Time Money Disadvantages: Nonrandom Not able to generalize findings
Research Data
Based on voluntary responses from subjects Biased sample occurs if subjects do not participate
Sampling Bias
Bias when samples are not carefully selected All nonprobability sampling methods have it. May occur in probability sampling methods - Subjects decide not to participate when chosen. - Final sample is now not representative of population
Disadvantages of Cluster Random Sampling
Causes a larger sampling error Requires each member assignment of population to cluster Uses a more complicated statistic analysis
Convenience Sampling
Chooses the most readily available subject or object Does not guarantee that the subject or object is typical of the population
Disadvantages of Simple Random Sampling
Complete listing of population is necessary It is time consuming to use.
Population
Complete set of persons or objects Common characteristic Of interest to the researcher Goal of sampling in quantitative research is to be able to make generalizations about the population from which the sample came from
Target Population
Definition: Entire group of people or objects People or objects meet designated set of criteria. Generalization of the findings
Accessible Population
Definition: Group of people or objects Researcher has access to them Must meet the desginated criteria of interest to the researcher
Advantages of Systematic Random Sampling
Easy to draw sample Economical Time-saving technique
Sample Terms
Element - Single member of a population Sampling frame - Listing of all elements of a population - Elements or members of a population are selected from a sampling frame
Tell whether the following statement is true or false: Researchers usually sample from the target population.
False Researchers usually sample from the accessible population but should identify the target population to which they want to generalize their results.
Tell whether the following statement is true or false: The aggregate of cases in which a researcher is interested is called a sample.
False The aggregate of cases in which a researcher is interested is called a population. A sample is selection of a portion of the population to represent the entire population.
Power Analysis
Helps to determine sample size May prevent type II error Helps to detect statistical significance Low power; type II error high External funding sources require it. Helps determine the optimum sample size
Steps for Simple Random Sampling
Identify the accessible population or list of elements Choose the method for getting the sample Note how easy it is through this example - Names of elements on slips of paper - Papers are placed into a hat. - Individual draws a slip of paper. - Individual continues until sample number is met
Sampling Error
Random fluctuations in data Not under the control of the researcher Chance variations occur when sample is chosen
Advantages of Stratified Random Sampling
Increases probability of being representative Ensures adequate number of cases for strata
Random Selection
Key word in sample selection Every subject has an equal chance
Cluster Random Stratified Sampling
Large groups or clusters, not people, are selected from population. Simple, stratified or systematic random sampling may be used during each phase of sampling
Advantages of Simple Random Sampling
Little knowledge of population is needed. Most unbiased of probability method Easy to analyze data and compute errors
Classification of Research Studies
Longitudinal Cross-sectional
Longitudinal Versus Cross-Sectional Studies
Longitudinal studies: Accurate means of studying changes over time Studies take a long time to perform. Cross-sectional studies: Less expensive Take less time Easier to conduct
Larger Sample Sizes
Many uncontrolled variables are present. Small differences are expected in members. Population must be divided into subgroups. Dropout rate among subjects is expected to be high. Statistical tests are used that require a minimum sample size
Sample Size
No simple rules Qualitative studies use much smaller samples than quantitative studies. Factors to consider for sample sizes in quantitative studies: - Homogeneity of population - Degree of precision desired by the researcher - Type of sampling procedure that is used Central limit theorem Sampling distribution of the mean
Probability Sampling
Researcher hopes that the variables of interest in the population will be present in the sample in the same proportion as would be found in the total population No guarantee Helps with inferential statistics with greater confidence Probability sampling allows researcher to estimate the chance that any given population element will be included in the sample Gives the ability to generalize the findings
Nonprobability Sampling
Sample elements are chosen nonrandomly. Produces biased sample Each element of the population may not be included in the sample. Restricts generalizations made about study findings
Volunteers As Subjects
Sample selection varies. - Subjects volunteer for a study. - Researcher approaches subjects
Disadvantages of Systematic Random Sampling
Samples may be biased. After first sample is chosen, no longer equal chance
Advantages of Cluster Random Sampling
Saves time and money Arrangements made with small number sampling units Characteristics of clusters or population can be estimated
Longitudinal Research Study
Subjects are followed over time. A cohort study is one example. Subjects are studied based on: - Similar age group - Similar background Data are gathered: - Same subjects - Several times - Tells influence of time
Cross-Sectional Study
Subjects checked at one point in time Data collected from groups of people Data may represent differences in: - Ages - Time periods - Developmental states - Important considerations
Sample
Subset of a population Sample represents the population If selected properly, the researcher is able to make claims about the population based on data from the sample alone
Tell whether the following statement is true or false: Probability sampling involves random selection of elements.
True Probability sampling involves random selection of elements.
Snowball Sampling
Type of convenience sampling method Study subjects recruit other potential subjects. May be called network sampling May find people reluctant to volunteer
Purposive Sampling
Type of nonprobability sampling Researcher uses personal judgment in subject selection. Each subject chosen is considered representative of population. Many qualitative studies use this technique.
Quota Sampling
Type of nonprobability sampling Researcher selects sample to reflect characteristics. - Examples of stratum Age Gender Educational background Number of elements in each stratum Number is in proportion to size of total population
Systematic Random Sampling
Type of probability sampling Every kth element is selected. Process: - Obtain a listing of population - Determine the sample size - Determine the sampling interval (k = N/n) - Select random starting point - Select every kth element
Stratified Random Sampling
Type of probability sampling Population is divided into subgroups or strata Simple random sample taken from each strata
Populations and Samples
Unable to study whole populations So...we sample from the population Quantitative research prefers populations Qualitative research focuses on individuals
Research Studies
Use voluntary subjects Follow the ethics of research - Subjects must voluntarily agree - Subjects may refuse to participate
Threefold Randomization Process
Used for experimental studies Helps represent the ideal study procedure Steps to ensure the process: - Subjects randomly selected from population - Subjects randomly assigned to groups - Experimental treatments randomly assigned to groups
Nonprobability Sampling
Uses nonrandom sampling procedures Selects sample from elements or members of population Types - Convenience - Quota - Purposive