MISUNDERSTOOD AND MISUSED SAMPLING MISUNDERSTOOD AND MISUSED
FACTORS AFFECTING SAMPLE SIZE OBJECTIVE OF RESEARCH DESCRIPTION INFERENCE HOMOGENEITY OF POPULATION SIZE OF POPULATION MARGIN OF ERROR
SAMPLING TERMS SAMPLE – some part of a “whole” ELEMENT – that unit about which information is collected and which provides the basis for analysis POPULATION – the theoretically specified aggregate of elements REPRESENTATIVENESS – the extent to which the sample “mirrors” the population EPSEM – Equal Probability of Selection Method
Sampling Terms (cont) SAMPLING UNIT – that element or set of elements considered for selection in some stage of sampling SAMPLING FRAME – the actual list of sampling units from which the sample, or some stage of the sample, is selected OBSERVATION UNIT – (unit of data collection) is an element or aggregation of elements from which information is collected SAMPLE SIZE – the number of elements selected
TYPES OF SAMPLES NON-PROBABILITY PROBABILITY
NON-PROBABILITY SAMPLES CONVENIENCE - procedure of obtaining those sampling units/elements most conveniently available Judgment – an experienced researcher selects the sample based on appropriate characteristics of the sample Quota – ensures that various subgroups of a population SNOBALL – initial respondents are selected by some method and then additional respondents are obtained from information provided by the initial respondents
Why Probability Samples? Typically more representative than other types of samples – bias Permit the researcher to estimate the accuracy or representativeness of the sample Saves time/money
Sampling Error Biased Selection – misses and/or over represents categories of elements Chance Variability – a sample deviates from the population value as a result of chance – increasingly problematic as sample size decreases
Stages in Selection of a Sample Define the Target Population Select a Sampling Frame Determine Sampling Method Plan Procedure for selecting elements Estimate Sampling Size * Draw Sample Conduct Field Word Check Sample against the Population or Sampling Frame * * If probability sample
Probability Sampling Simple Random – technique which assures that each selected element in the population has an equal chance of being included in the sample Systematic – an initial starting point is selected by a random process and then every nth numbered element in the frame is selected Stratified – random subsamples are drawn from within each stratum. The sub samples may be proportional or disproportional to the number of elements in each stratum
Systematic Sample Distance between elements = SAMPLING INTERVAL = K e.g. we want a sample of 144 = n, where N = 1300 N/n or 1300/144 = 9.02 this then is the Sampling Interval K = 9 Using a random start, every 9th element would be selected Sampling Ratio = proportion of population to be selected (N/n) where n = the desired sample size N/n e.g. N = 1000 and n = 100 Sampling Ratio = 1000/100 Sampling Ratio = 1/10th or as per above 1/9th A random sample of 100 or 144 elements would be selected
Probability Sampling (cont) Cluster – large clusters of elements, not individual elements, are selected in the first stage of sampling Area – Cluster sampling when the cluster consist of a geographical area Multistage Area – Cluster sampling that involves a combination of two or more probability sampling techniques