Ch 11 Sampling
The Nature of Sampling Sampling Population Element Population Census Sampling frame
Why Sample? Greater accuracy Availability of elements Availability of elements Greater speed Sampling provides Sampling provides Lower cost
When Is a Census Appropriate? NecessaryFeasible
What Is a Valid Sample? AccuratePrecise
Types of Sampling Designs Element Selection ProbabilityNonprobability UnrestrictedSimple randomConvenience RestrictedComplex randomPurposive SystematicJudgment ClusterQuota StratifiedSnowball Double
When to Use Larger Sample Sizes? Desired precision Number of subgroups Number of subgroups Confidence level Population variance Small error range
What is Sample Sampling 1. Any group on which information is obtained 2. Usually is representative of a larger group called a population 3. Defining the population --- The group of interest to the researcher --- A group researcher would like to generalize --- Usually has at least one characteristic that sets it off from other populations --- The target population may not have the characteristics one would like to generalize from --- The accessible population is the population one could generalize from
Random Sampling Definition 1. Obtaining a rather accurate representative view of a larger group 2. Every individual in the population has an equal opportunity to be selected 3. When each member of the population does not have a chance of being selected because the researcher is looking for specific criteria it is an example of a nonrandom sample or purposeful sample
Random Sampling (Continued 1) Random Sampling Methods 1. Simple Random Sampling --- Every member of the population has an equal and independent chance of being selected --- This done by using a table of random numbers which is a large list of numbers that has no order or pattern --- The list is usually focused in the back of statistics books --- The purpose of random sampling is that if it is large enough it should produce a representative sample
Simple Random Advantages Easy to implement with random dialing Disadvantages Requires list of population elements Time consuming Uses larger sample sizes Produces larger errors High cost
Random Sampling (Continued 2) --- A random sample needs to be more than 20 to 30 individuals to be large enough to be representative a. Descriptive studies: 100 b. Correlational studies: 50 c. Experimental studies: 30 d. Causal-comparative: Stratified Random Sampling --- Strata-sub groups are selected for the sample in the same proportions as they exist in the population --- Selects a representative or equal percentage from each strata using the table of random numbers --- This will improve the likelihood that the key characteristics the researcher is wising to make generalizations about will be included proportionately in the sample
Stratified Advantages Control of sample size in strata Increased statistical efficiency Provides data to represent and analyze subgroups Enables use of different methods in strata Disadvantages Increased error will result if subgroups are selected at different rates Especially expensive if strata on population must be created High cost
Selecting a Stratified Sample 219 female students (60%) 146 male students (40%) In a population of 365 twelfth- grade American governments 66 female students (60%) From these, she randomly selects a stratified sample of: and 44 male students (40%) The researcher identifies two subgroups, or strata:
Cluster Advantages Provides an unbiased estimate of population parameters if properly done Economically more efficient than simple random Lowest cost per sample Easy to do without list Disadvantages Often lower statistical efficiency due to subgroups being homogeneous rather than heterogeneous Moderate cost
Random Sampling (Continued 3) 3. Cluster Sampling --- A randomized sample of group within the total population rather than individuals --- Everyone within the group will need to become part of the sample --- Used when individuals are difficult to randomize --- Can only make generalizations about the group and not the individuals within the group. This is a common error researchers make. 4. Two Stage Random Sampling --- Combines cluster sampling with individual sampling --- This would help to eliminate any problems with just cluster sampling --- One would randomly select a cluster and than randomly select individuals within the cluster
Stratified and Cluster Sampling Stratified Population divided into few subgroups Homogeneity within subgroups Heterogeneity between subgroups Choice of elements from within each subgroup Cluster Population divided into many subgroups Heterogeneity within subgroups Homogeneity between subgroups Random choice of subgroups
Nonprobability Samples Cost Feasibility Time No need to generalize Limited objectives
Nonprobability Sampling Methods Convenience Judgment Quota Snowball
Random Sampling (Continued 4) Non-Random Sampling 1. System Sampling --- Every individual in the population list is included in the sample --- Random Start-Draw a number from a hat and select every individual who is within that sampling interval 2. Convenience Sampling --- A group of individuals are available for the study --- Such samples cannot be considered representative --- Demographics and characteristics need to be discussed --- To improve validity of the sample more than one study should be used to overcome an one time occurace
Random Sampling (Continued 5) 3. Purposeful Sampling --- Researcher uses personal judgment about the sample based on the knowledge of the population and the specific purpose of the research Generalizing from a sample 1. Population generalizability --- The degree to which a sample represents the population of interest and can be generalizable to a large group --- A representative sample is the extent to which a sample is identical in all characteristics to the intended population
Random Sampling (Continued 6) ---- Any researcher who loses over 10 percent of the original ample would be well advised to acknowledge this limitation and qualify their conclusions accordingly ---- If the sample is less than expected the researcher should describe the sample as thoroughly as possible so that interested persons can judge for themselves to which any of the findings apply Ecological Generalizability ---- The degree to which the results of a study can be extended to other settings or conditions --- The environment and or the settings have to be the same