Sampling refers to the research technique or method that involves selecting part of the population with a view of developing solutions or drawing conclusions on the entire population. It involves the selection of a portion of the total for investigative purposes. A sampling frame consists of a predetermined number of units from the total population (Burns & Bush, 2008). An ideal sampling frame is that which is able to identify a unit or an element of the population only once. Such a sampling frame is a rarity in practical research because of the existence of defects within the sampling frame. Some common defects within the sampling frame include inadequate frames, inaccurate frames and incomplete frames. Instances of incomplete frame occur when some aspects of the population’s unit which are legitimate are omitted. Contrastingly, inaccurate frames occur when some aspects of the population’s unit are listed inaccurately (Hair, 2008). In some instances, the listen population attributes are actually non-existent. Inadequate frame occurs when all the units of the population are not included in the sample. Defects in the sampling frame often leads to errors and one of such an error is referred to as the sampling frame error.
The sampling frame error is defined as the difference between the results that would have been arrived at from a conclusive and comprehensive enumeration of the entire population and those that have been obtained from the samples. It can also be defined as the difference between the population as depicted by the sampling frame and the population defined by the researcher. Sampling errors normally occur due to a number of random factors. They mostly occur as a result of reducing the sample size and hence the sample ends up representing only a part of the population. A common example of an occurrence of a sampling frame error is in the use of a mailing list generated from a random mass email in a corporate email inbox that is used to generate a list of email addresses does not accurately represent the population of the organization’s potential customers due to new email addresses that have not been included in the mailing list, obsolete email addresses and unlisted email addresses.
A sampling frame error is calculated by subtracting the statistically estimated amount from the actual amount from the sample. The first step in calculating the sampling error is to compute the percentage of the statistical measure that is being taken. For instance, if you would like to know what percentage of employees in a given organization smoke a particular brand of cigarette, the first step would be to take a random sample, say n, the sample size equals to 40. Have the 40 employees to fill out a survey and compute the percentage number of employees who smoke the particular brand of cigarette that is under survey. If 10 employees smoke the particular brand of cigarette, then the percentage becomes 25%.
Every employee in the organization is then let to participate in the survey and the percentage of employees who smoke the brand of cigarette is computed. For example, if 20 out of 200 employees smoke, then the computed percentage becomes 10%. The sampling frame error thus becomes 25 (estimated) – 10 (actual) = 15. The units of a sampling frame error are given by any absolute value (Proctor, 2006). If a sampling frame error calculation arrives at a negative unit of calculation such a unit is squared to make it positive. This is then referred to as the squared error. The closer to zero the result is, then the smaller the sampling frame error and the more reliable the estimation was. In real life however, the actual value is not used confidence intervals and standard errors are used for interpretation (Churchill, 2011).
References.
Burns, A. C., & Bush, R. F. (2008). Basic marketing research: Using Microsoft Excel data analysis. Upper Saddle River, N.J: Pearson Prentice Hall.
Churchill, G. A. (2011). Basic marketing research. Fort Worth: Dryden Press.
Hair, J. F. (2008). Essentials of marketing research. Boston: McGraw-Hill/Higher Education.
Proctor, T. (2006). Essentials of marketing research. Harlow: Financial Times Prentice Hall.
