Chapter 11 Summary

Discuss the concept of sampling and list reasons for sampling.

Sampling can be thought of as taking a portion of the whole and studying that portion to make estimates about the whole. We use the fundamentals of sampling in many of our everyday activities (e.g., selecting a TV program to watch, test-driving a car before deciding whether to purchase it, determining if our food is too hot or if it needs some additional seasoning). The term target population is used to identify the complete group of elements (e.g., people or objects) that are specifically identified for investigation and determined by the specific objectives of the research project. The researcher is able to select sampling units from the target population and use the results obtained from the sample to make conclusions about the target population. It is essential that the sample be representative of the target population if it is to provide accurate estimates of population parameters.
Sampling is frequently used in marketing research projects as opposed to conducting a census because sampling can significantly reduce the amount of time and money required in data collection activities. In instances where the process of measurement in a study destroys or contaminates the elements being studies, sampling may be the only alternative.

Identify and explain the different roles that sampling plays in the overall information research process.

Sampling plays an important role when there are short time frames for gathering the needed information. These time frames are more likely to fit into the decision time frames of the users of the new information. Sampling plays a significant role in the process of identifying, developing, and understanding new marketing/business constructs as well as in developing the scale measurements used to actually collect raw primary data about people or objects. Decisions concerning the use of samples indirectly affect the process of designing questionnaires. Finally, sampling plays a significant role in decisions concerning the type of data analysis procedures that can be employed to statistically investigate the sample data statistics.

Identify the fundamental differences between probability and nonprobability sampling methods, and point out their strengths and weaknesses.

We discussed probability and nonprobability sampling methods. Probability sampling methods require that each elements in the target population be identifies and have a known, nonzero probability of being selected for the sample, so that statistical methods can be used to project sample results to the target population within a specified margin of error. Probability sampling methods produce unbiased estimates of the defined target population characteristic of interest and ensure the representativeness of the sample. However, probability sampling designs can be time-consuming and expensive compared to nonprobability sampling methods.
Nonprobability sampling does not allow for the use of statistical methods to determine the degree of representativeness to the defined target population. Rather, sampling units are selected based on the subjective judgment or intuitive knowledge of the researcher. The success of the research project will depend on the decisions made by the researcher because it is all but impossible to accurately generalize the sample data derived from a nonprobability-based method to all the elements making up the target population.

Types of probability sampling designs include simple random sampling (SRS), systematic random sampling (SYMRS), stratified random sampling (STRS), and cluster sampling. Four types of nonprobability sampling designs are convenience (or accidental) sampling, judgment (or purposive) sampling, quota sampling, and snowball sampling.

Discuss and calculate sampling distributions, standard errors, and confidence intervals and how they are used in assessing the accuracy of a sample.

A sampling distribution is the frequency distribution of a specific sample statistic or sample proportion that would result if we took repeated random samples of the same size. The central limit theorem from statistics tells us that there is a high probability that the mean of any random sample taken from a target population will closely approximate the actual population mean as the sample size increases. Formulas are used to compute the estimated standard error of a sample mean as the sample size increases. Formulas are used to compute the estimated standard error of a sample mean and the estimated standard error of a sample percentage. The estimated standard error or gives us an indication of how far the sample data results lie from their respective actual population parameters.

Confidence intervals are based on the researcher's desired level of confidence and within a calculated degree of sampling error for which estimates of the true value of the population parameter could be expected to fall.

Identify the criteria involved in determining the appropriate sample design for a given research project.

The sampling design for a given research project is chosen after considering several factors: (1) the research objectives, (2) degree of accuracy required, (3) availability of resources, (4) time frames, (5) advanced knowledge of the target population, (6) geographic scope of the project (e.g., local or national), and (7) perceived data analysis needs.

Discuss the factors that must be considered when determining sample size.

The researcher must consider several factors when determining the appropriate sample size to use for a given study. The amount of time and money available often affect this decision. In general, the larger the sample, the greater the amount of resources required to collect raw data. Three factors that are of primary importance in the determination of sample size are (1) the variability of the population characteristic under consideration, (2) the level of confidence desired in the estimate (CL), and (3) the degree of precision desired in estimating the population characteristic. The greater the variability of the characteristic under investigation, the higher the level of confidence required; and the more precise the required sample results, the larger the necessary sample size.

Discuss the methods of calculating appropriate sample sizes.

Statistical formulas are used to determine the required sample size in probability sampling. Sample sizes for nonprobability sampling designs are determined using subjective methods such as industry standards, past studies, or the intuitive judgments on the part of the researcher. The size of the defined target population does not affect the size of the required sample unless the population is small relative to the sample size. Sample sizes are not the same as usable observations for data analysis. Having fewer observations than desired will affect the accuracy of the data. Researchers must therefore consider reachable rates, overall incidence rates, and expected completions rated on the number of prospective respondent contacts necessary to ensure sample accuracy.

Identify and explain the steps involved in developing a sampling plan, and design a variety of different sampling plans.

In the appendix to this chapter, we will briefly summarize the seven steps that should be included in the development of a sampling plan: (1) define the target population, (2) select the data collection method, (3) identify the sampling frames needed, (4) select the appropriate sampling method, (5) determine the necessary sample sizes and overall contact rates, (6) create an operating plan for selecting sampling units, and (7) execute the operational plan.