Establishing and understanding a sample frame is one of the first and most important steps in the development of a sampling methodology and process. A sample frame is “the list of elements from which the sample is actually drawn” (Cooper & Schindler, 2014, p. 347). This means that it is a list of the entire population that one is interested in studying and from which the sample will be drawn. For example, if one was interested in knowing about NCAA basketball players eating habits they would select a sample of all NCAA basketball players to study, which would make all NCAA basketball players the sample frame. Below, five sampling methodologies will be discussed in light of how they could be used to study new hires skill needs in the Accounting field and a description of the sample frame will be offered.
Simple Random Sampling
In a simple random sample, all members of the population have an equal chance of being selected for sampling. To determine the employers’ skill needs in new hires in the accounting field using a simple random sample would be a daunting task, because one would need to know the skills that all successful newly hired accountants possess. Therefore, to perform this study the sample frame from which the simple random sample would be drawn would be a listing of all newly hired accountants that have been successful in their workplace. This listing of individuals would have to include each individual’s certifications and college degree, in order to satisfy the need to know their skill set.
This methodology of sampling was used in the Bible, to take a census of the entire population. “Take a census of all the congregation of the people of Israel, by clans, by fathers’ houses, according to the number of names, every male, head by head” (Numbers 1:2, English Standard Version).
For the use of systematic sampling, the sample frame itself would be fairly similar to that which was used in a simple random sample. However, the method by which a sample is drawn from the sample frame would be far different when using a systematic sampling method. While it is still necessary to have information from the entire population to draw a sample, the population could be defined differently from a systematic sample. Rather than using each individual in the population to define the population, one could acquire all job postings in the industry and form a list of all required skills, degrees, and certifications. From this list, a systematic approach of randomizing the order of the list and then selecting every sixth item to sample could be used. The researcher would take each selected item and then measure the success of each person who filled a job posting with this requirement.
According to Teddlie and Yu, “stratified sampling occurs when the researcher divides the population into subgroups such that each unit belongs to a single stratum and then selects units from those strata” (2007, p. 79). Considering this definition, to determine the skill that employers need for new hires in accounting, the population of all new hires would need to be broken down into subgroups. For this sample frame, the most informative and useful strata would be company size of hire. Meaning that knowing the skills needs of companies based on their size is the most informative sample frame, in a stratified sampling methodology.
Cluster sampling occurs when not an individual, but a natural group formed in the population is used for sampling purposes (Teddlie & Yu, 2007, p. 79). To determine what the sample frame would be for this study, one must understand what clusters for naturally in the population. Since accounting jobs are needed nationwide, clustering could be done based on state or region to for the sample frame in this situation.
In double sampling, the researcher uses data from a previously surveyed sample to select a subsample for further research. Since there are many different types and levels of accounting jobs, it would be interesting to understand the skill needs of employers for the different levels of accounting jobs. To do this, the simple random sample in which skills were defined by certifications and degrees could be used. One could take this information and only perform further sampling on the individuals who were in the top 25 percent of certifications and degrees to better understand this population.
Cooper, D. R., & Schindler, P. S. (2014). Business research methods (12th ed.). New York, NY: McGraw-Hill. ISBN: 9780073521503.
Teddlie, C., & Yu, F. (2007). Mixed Methods Sampling: A Typology With Examples. Journal of Mixed Methods Research, 1(77), 77-100. Retrieved December 6, 2016, from http://sociologyofeurope.unifi.it/upload/sub/documenti/Teddlie – Mixed Methods Sampling – A Typology With Examples.pdf