CJ 4980
Sampling Methods
2 of them: Non-probability and probability.
Sampling frame
A list of all elements or other units containing the elements in a population
Cluster Sampling
A multistage sampling technique in which natural groups are first selected Then, members of each selected group are sub-sampled afterward Used when it's not possible to create a list of all elements that compose the target population This method is efficient, but less accurate Rule of thumb: Maximize the number of clusters selected while decreasing the number of elements in each cluster Why? More clusters represent more of the population Clusters tend to be homogeneous, so fewer elements are needed to represent the group
Experiment
A process of observation, to be carried out in a situation expressly brought about for that purpose At a basic level, experiments involve: Taking action Observing the consequences of that action
Simple Random Sampling
Basic method that assumes that the researcher has a sampling frame With sampling frame is in hand, the researchers assigns a single number to each element on the list Using a table of random numbers, or a computer program (or some other method) elements are selected to be included in the sample
Internal Validity Instrumentation
Changes in the measurement process lead to changes in results (e.g., pre-test vs. post-test)
Internal Validity Selection Bias
Comparisons of groups are meaningless unless the groups are comparable at the start of the experiment
Non probability methods
Convenience sampling Purposive sampling Snowball sampling Quota sampling These sampling methods are based on the availability of subjects, or the researcher's judgment
Internal Validity History
External events may occur during the course of the experiment
Internal Validity Statistical Regression
If subjects start the experiment with extreme scores, and appear to improve, it may be that they improved because they regressed (had nowhere else to go) Example: math scores
Internal Validity Experimental Mortality
In longitudinal designs, participants can drop out as the experiment progresses This, of course, affects the composition of groups and therefore the comparisons of the groups
Internal Validity Maturation
People are constantly growing and changing and these changes can affect the results in long-term studies
Probability sampling
Samples are selected in accordance with probability theory, typically involving random selection Allows for generalization to a larger population Likelihood of selection is based on probability and is known
Non probability
Samples are selected in some way not suggested by probability theory Therefore, the likelihood of selection is unknown
Stratified Sampling
Stratification means grouping of the units composing a population into homogenous groups (strata) before sampling The sample is then drawn from those groups The purpose is to gain greater representativeness and therefore decrease sampling error It is a modification of simple random sampling and systematic sampling
Systematic Sampling
Systematic sampling A type of probability sampling in which every kth unit in a list is selected for inclusion in the sample You should select the first element at random Sampling interval -The distance between elements selected from a population in the sample Sampling ratio The proportion of elements in the population that are selected to be in a sample LOOK IN SLIDE SHOW TOO
Internal Validity Demoralization
The control group feels deprived and gives up on the experiment This drives their 'scores' down
Population
The entire set of individuals or other entities to which study findings are to be generalized
Sampling Error
The error associated with studying a sample instead of a whole population Every sample has some error associated with it! Since a sample is not an exact representation of the population, error results We can estimate the degree to be expected through mathematical formulas Reduced by: (1) larger samples and (2) homogeneous populations
Confidence Level
The estimated probability that a population parameter lies within a given confidence interval Example: on a normal curve, we are 68% confident that our sample estimate is within one standard error of the parameter
Elements
The individual members of the population whose characteristics are to be measured
Internal Validity
The possibility that conclusions may not accurately reflect what went on in the experiment.
Internal Validity Testing
The process of testing and retesting can influence results, especially if participants are assessed many times
Representativeness
The quality of a sample of having the same distribution of characteristics as the population from which it was selected
Confidence Interval
The range of values within which a population parameter is estimated to lie Knowing this allows us to know how big of a sample we need (see Appendix E)
Sample Statistics
The summary description of a variable in the sample, used to estimate a population parameter Example: estimates of income of neighborhood residents from the sample
Population Parameter
The summary statistic of a given variable in the population Example: average income of neighborhood residents
Generalizability
There are two different meanings of generalizability: Can the findings from a sample of the population be generalized to the population from which the sample was selected? Can the findings from a study of one population be generalized to another, somewhat different population? Sample generalizability depends on sample quality, which is determined by the amount of sampling error Sampling error The difference between the characteristics of a sample and the characteristics of the population from which it was selected The larger the sampling error, the less representative the sample—and thus the less generalizable the findings
Classical Experiment
True experiments must have at least three things: Two groups (in the simplest case, an experimental and a control group) Variation in the independent variable before assessment of change in the dependent variable (time order) Random assignment to the comparison groups Results in greater confidence in the validity of causal conclusions than is possible in other research designs In the natural and social sciences, classical experiments involve three major pairs of components: Independent and dependent variables Experimental and control groups Pre-testing and post-testing
Sampling Theory
focuses on the generalizability of descriptive findings to the population from which the sample was drawn It also considers whether statements can be generalized from one population to another
Purposive Sample
involve selecting the sample on the basis of knowledge of the population, its elements, or the purpose of the study Judgmental sampling because cases are selected on the basis of the researcher's judgment about which ones will be the most useful or representative You may want to study a small subset of a larger population in which many members of the subset are easily identified Example: Dungeons & Dragons players
Snowball Samples
involves asking current participants to refer or suggest additional people to participate in the study Accidental sampling "Snowballing" - process of accumulation as each located subject suggests other subjects Questionable representativeness and generalizability Example: Wright & Decker's Burglars on the Job Studies of the homeless or other hard to reach populations
Convenience Samples
rely on available subjects Example: Stopping people on a street corner to ask them what they think about a given topic Considered risky because representativeness cannot be established Compromises generalizability of the results When to use: When the researcher is interested in people passing the sampling point at some specified time When something better isn't feasible
Quota Samples
select units into the sample on the basis of their characteristics in the population This way, the sample has the same distribution of characteristics assumed to exist in the population being studied Helps with representativeness of characteristics (e.g., gender, race, age, class, etc.) Example: Gallop Presidential poll (1936) vs. Reader's Digest Poll