4.6.3 Stratified sampling
Stratified sampling is distinguished by the two-step procedure it involves. In the first step the population is divided into mutually exclusive and collectively exhaustive sub-populations, which are called strata. In the second step, a simple random sample of elements is chosen independently from each group or strata. This technique is used when there is considerable diversity among the population elements. The major aim of this technique is to reduce cost without lose in precision. There are two types of stratified random sampling; (a) proportionate stratified sampling and (b) disproportionate stratified sampling. In proportionate stratified sampling, the sample size from each stratum is dependent on that stratum's size relative to the defined target population. Therefore, the larger strata are sampled more heavily using this method as they make up a larger percentage of the target population. On the other hand, in disproportionate stratified sampling, the sample selected from each stratum is independent of that stratum's proportion of the total defined target population. There are several advantages of stratified sampling including the assurance of representativeness, comparison between strata and understanding of each stratum as well as its unique characteristics. One of the major difficulty however, is to identify the correct stratifying variable.
Cluster sampling is quite similar to stratified sampling wherein in the first step the population is also divided into mutually exclusive and collectively exhaustive sub-populations, which are called clusters. Then a random sample of clusters is selected, based on probability random sampling such as simple random sampling. The major difference between stratified and cluster sampling is that in stratified sampling, all the subpopulations (strata) are selected for further sampling whereas in cluster sampling only a sample of subpopulations (clusters) is chosen. The objectives of these methods are also different. The objective of stratified sampling is to increase precision while cluster sampling strives to increase sampling efficiency by decreasing costs. Because one chooses a sample of subgroups with cluster sampling, it is desirable that each subgroup be a small scale model of the population. Thus, the subgroups (clusters) ideally should be formed to be as heterogeneous as possible. If all elements in each selected cluster are included in the sample, the procedure is called one-stage clustering. However, if a sample of elements is drawn probabilistically from each selected cluster, the procedure is called two-stage clustering. The most common form of cluster sampling is area sampling in which the clusters consists of geographical areas. There are several advantages of cluster sampling including the reduction in costs due to available data with regard to population groups (such as telephone directories and address lists) and feasibility of implementation. However, one of the major disadvantages of cluster sampling is the homogeneity among the selected cluster. Ideally each cluster should represent the population at large however, in reality it is quite difficult to achieve.
The selection of probability and nonprobability sampling is based on various considerations including, the nature of research, variability in population, statistical consideration, operational efficiency and sampling versus nonsampling errors. Nonprobability sampling is mainly used in product testing, name testing, advertising testing where researchers and managers want to have a rough idea of population reaction rather than a precise understanding. Ad depicted in figure 4.1 there are various types of nonprobability sampling including, convenience sampling, judgement sampling, quota sampling, snowball sampling.
As the name implies, in convenience sampling, the selection of the respondent sample is left entirely to the researcher. Many of the mall intercept studies (discussed in chapter 3 under survey methods) use convenience sampling. The researcher makes assumption that the target population is homogenous and the individuals interviewed are similar to the overall defined target population. This in itself leads to considerable sampling error as there is no way to judge the representativeness of the sample. Furthermore, the results generated are hard to generalize to a wider population. While it has a big disadvantages relating to sampling error, representativeness and generalizability, convenience sampling is least time consuming and least costly among all methods.
Judgement sampling, also known as purposive sampling is an extension to the convenience sampling. In this procedure, respondents are selected according to an experienced researcher's belief that they will meet the requirements of the study. This method also incorporates a great deal of sampling error since the researcher's judgement may be wrong however it tends to be used in industrial markets quite regularly when small well-defined populations are to be researched. For example, if a manager wishes to the satisfaction level among the key large-scale business customers judgement sampling will be highly appropriate. Same as convenience sampling, judgement sampling also has disadvantages relating to sampling error, representativeness of sample and generalizability however the costs and time involvement is considerably less.
Quota sampling is a procedure that restricts the selection of the sample by controlling the number of respondents by one or more criterion. The restriction generally involves quotas regarding respondents' demographic characteristics (e.g. age, race, income), specific attitudes (e.g. satisfaction level, quality consciousness), or specific behaviours (e.g. frequency of purchase, usage patterns). These quotas are assigned in a way that there remains similarity between quotas and populations with respect to the characteristics of interest. Quota sampling is also viewed as a two-stage restricted judgement sampling. In the first stage restricted categories are built as discussed above and in the second stage respondents are selected on the basis of convenience of judgement of the researcher. For example, if the researcher knows that 20% of the population is represented by the age group 18-25, then in the final sample s/he will try to make sure that of the total sample 20% of them represent the age group 18-25. This procedure is used quite frequently in marketing research as it is easier to manage in comparison to stratified random or cluster sampling. Quota sampling is often called as the most refined form of nonprobability sampling.49 It also reduces or eliminates selection bias on the part of field workers which is strongly present in convenience sampling. However, being a nonprobability method it has disadvantages in terms of representativeness and generalizability of findings to a larger population.
4.7.4 Snowball sampling
In snowball sampling, an initial group of respondents is selected, usually at random. After being interviewed however, these respondents are asked to identify others who belong to the target population of interest. Subsequent respondents are then selected on the basis of referral. Therefore, this procedure is also called referral sampling. Snowball sampling is used in researcher situations where defined target population is rare and unique and compiling a complete list of sampling units is a nearly impossible task.50 For example, in the case of the earlier discussed example of the manager of brand X of washing machine, if s/he wanted to study the owners of the second hand washing machines it will be very difficult to identify the owners of such washing machines and therefore, snowball sampling may provide a way forward. If traditional probability of nonprobability methods were used for such a study, they will take too much time and incur high costs. The main underlying logic of this method is that rare groups of people tend to form their own unique social circles.51 While there are several disadvantages in using this procedure as it is a nonprobability technique. However, on the other hand it is a good procedure for identifying and selecting hard-to-reach, unique target populations at a reasonable cost and time.
As discussed above, both probability and nonprobability sampling techniques have their own advantages and disadvantages. Overall, it depends on various factors to choose the most appropriate sampling technique. A researcher has to consider the research objectives first as to do they call for qualitative or quantitative research. Secondly, available resources should be kept in mind including the time frame available for conducting the researcher and making the findings available. The knowledge regarding the target population as well as the scope or research also is important in selecting the right kind of sampling technique. Researcher should also focus on the need for statistical analysis and degree of accuracy required with regard to the research and the expected outcomes. On the basis of these parameters a researcher can identify an appropriate sampling technique.
This chapter focused on one of the most important research issue in marketing research, sampling. As detailed in the chapter sampling is quite a common phenomenon in our decision making process. Before delving deeply into the sampling process one must be aware of several basic constructs involved with sampling namely; population, target population, elements, sampling unit and sampling frame. Determining the final sample size for research involves various qualitative and quantitative considerations.
There are two basic techniques of selecting sample; probability sampling techniques and nonprobability sampling techniques. Probability sampling techniques are more robust in comparison to nonprobability sampling. Findings based on nonprobability are hard to generalize to a wider population.
Probability sampling is sub-divided into simple random sampling, systematic sampling, stratified sampling and cluster sampling. While being robust probability sampling techniques are resource intensive in terms of cost and time involved. Nonprobability sampling is subdivided into convenience sampling, judgement sampling, quota sampling and snowball sampling. Nonprobability sampling techniques are less costly and less time consuming however they have problems relating to selection bias also.
Selecting an appropriate sampling technique depends on various factors such as research objectives, available resources, knowledge of target population and scope of research, degree of accuracy and statistical analysis required for result interpretation.
Employees at FOSS Analytical A/S are living proof of the company value - First - using new inventions to make dedicated solutions for our customers. With sharp minds and cross functional teamwork, we constantly strive to develop new unique products -Would you like to join our team?
FOSS works diligently with innovation and development as basis for its growth. It is reflected in the fact that more than 200 of the 1200 employees in FOSS work with Research & Development in Scandinavia and USA. Engineers at FOSS work in production, development and marketing, within a wide range of different fields, i.e. Chemistry, Electronics, Mechanics, Software, Optics, Microbiology, Chemometrics.
A challenging job in an international and innovative company that is leading in its field You will get the opportunity to work with the most advanced technology together with highly skilled colleagues.
Read more about FOSS at www.foss.dk - or go directly to our student site www.foss.dk/sharpminds where you can learn more about your possibilities of working together with us on projects, your thesis etc.
The Family owned FOSS group is the world leader as supplier of dedicated, high-tech analytical solutions which measure and control the quality and production of agricultural, food, pharmaceutical and chemical products. Main activities are initiated from Denmark, Sweden and USA with headquarters domiciled in Hillerod, DK. The products are marketed globally by 23 sales companies and an extensive net of distributors. In line with the corevalue to be 'First', the company intends to expand its market position.
Dedicated Analytical Solutions
Was this article helpful?
There are people all over the world trying to find ways to make money online. From stay at home moms looking to make a few extra dollars to college students and entrepreneurs, the allure of making your own hours and working from home or from the local coffee shop is very appealing.