Marketing Research Paper on Sampling: Elements of the Population (Research Paper Sample)

Student’s Name:
Student’s ID:
Tutor’s Name:
Table of Contents TOC \o "1-3" \h \z \u Introduction PAGEREF _Toc437569715 \h 3Sampling PAGEREF _Toc437569716 \h 3Probability Sampling PAGEREF _Toc437569717 \h 3Simple Random Sampling PAGEREF _Toc437569718 \h 3Complex Probability Sampling PAGEREF _Toc437569719 \h 3Non-Probability Sampling PAGEREF _Toc437569720 \h 4Convenience PAGEREF _Toc437569721 \h 4Purposive Sampling PAGEREF _Toc437569722 \h 4Snowball Sampling PAGEREF _Toc437569723 \h 5Questionnaire PAGEREF _Toc437569724 \h 5Types of Questions PAGEREF _Toc437569725 \h 5-Free Response Questions PAGEREF _Toc437569726 \h 5-Dichotomous Questions PAGEREF _Toc437569727 \h 5-Multiple Choice Questions PAGEREF _Toc437569728 \h 5-Checklist PAGEREF _Toc437569729 \h 5-Rating Questions PAGEREF _Toc437569730 \h 6-Ranking Questions PAGEREF _Toc437569731 \h 6Measures of Central Tendency & Dispersion PAGEREF _Toc437569732 \h 6Data Samples and Distribution PAGEREF _Toc437569733 \h 6Symmetric Distribution PAGEREF _Toc437569734 \h 6Asymmetric Distribution PAGEREF _Toc437569735 \h 7Interview PAGEREF _Toc437569736 \h 7Individual Depth Interview PAGEREF _Toc437569737 \h 7Group Interview PAGEREF _Toc437569738 \h 8Focus Group PAGEREF _Toc437569739 \h 8Conclusion PAGEREF _Toc437569740 \h 8References PAGEREF _Toc437569741 \h 9Appendices PAGEREF _Toc437569742 \h 10
Introduction
The descriptive assignment attempts to shed light on numerous processes and methods involved in conducting a meaningful research, moving from the initial steps of selecting the best samples based on various applicable sampling types to organizing the research tools like questionnaires and interviews and finally getting into the analysis phase with the data obtained so far.
Sampling
Sampling is choosing a few elements of the population that may be termed as representatives of the entire population. These elements are chosen from a listing of all the population elements called the sample frame.On the other hand, a population is the collection of all the elements under study. The study that entails all the elements is called a census.
The conditions for having a sample, rather than census, are outlined as follows
* A sample is feasible if the population from which the sample is drawn is large enough to be studied in entirety.
* When there is a close approximation among elements of the population, such that drawing one or the other doesn’t change the results by margins.
There are certain characteristics of a sample that add to itsvalidity. For a sample to qualify being considered one, it must have adequate no. of elements to offset any understating or overstating biases, thus making a sample more accurate and free from biases.
The next criterion is the degree to which the sample is likely to conclude almost the same results the population would, had it been studied. That is to say that the values obtained from that of the sample must not deviate significantly from the corresponding values of the population. The lesser the deviation among results from sample and population, the more precise the sample set and the little sampling error it carries (Cooper & Schindler, 2014)
Sampling has the following two types based on representation basis:
Probability Sampling
It’s a sampling type in which each element hasan established likelihood of being selected (Kuiper &Clippinger, 2012). It is technically superior form of sampling data and is typically used when the findings are to be generalized for the population. It is also established that this sampling type provides closer estimates, thus making the sample more precise. Its types include:
Simple Random Sampling
It is the sampling type in which elements are selected directly and no controls are imposed in elements selection. This sampling type weighs equal chance of selection for each element of the population. Hence, this unrestricted sampling type is most accurate and has the least bias, however, it’s expensive and time-taking exercise to implement (Cooper & Schindler, 2014).
Complex Probability Sampling
It is such type of sampling which is economically as well as statistically efficient, meaning that it achieves the desired precision in a lower cost. Such is not the case with simple random sampling. Also called restricted probability sampling, it further has four sampling designs which have been in use to curtail the inefficiencies of controls-free unrestricted sampling (Kuiper &Clippinger, 2012).
Systematic Sampling
It is such a sampling type in which every nth element is drawn from a sample frame after initiating with selection of a random element. The nth element is obtained by dividing the sample size into population size. This nth element is often called the ‘skip interval’ (Kuiper &Clippinger, 2012).
Stratified Random Sampling
This sampling type classifies the whole population based on a single factor to form strata. It is then followed by selecting randomly the elements from the strata. The subject may decide to choose from Proportionate Sampling, in which the proportionate percentage of elements is drawn from the respective stratum, or the Disproportionate sampling where in the elements are drawn based on the judgment of the researcher (Cooper & Schindler, 2014).
Cluster Sampling
This sampling type requires dividing the whole population into groups called clusters that have homogeneity all across. These clusters have diversity among elements in general. Once defined, the researcher selects a few clusters to analyze all of its elements (Cooper & Schindler, 2014).
Area Sampling
An important type of cluster sampling in which the study is focused on population that can be identified based on geographical area (Cooper & Schindler, 2014).
Double Sampling / Multiphase Sampling
In this sampling type, the data is collected initially using any of the four techniques discussed above. The information so found is examined to figure out if further study can be conducted on a subsample among the sample. This process is only applicable if the relative cost to forming a subsample is lesser than that of the cost associated with forming a fresh sample (Cooper & Schindler, 2014).
Non-Probability Sampling
It is a subjective sampling type which has no known probability for selection of any population element. The selection is entirely based on the choice of an individual and this makes the credibility of the sample debatable. Such samples usually carry strong bias and an error range cannot be estimated for such. However, at times, this sampling type makes sense. The methods for non-probability sampling as discussed below
Convenience
Convenience sampling is an unrestricted sampling type in which the research can opt to choose freely whoever is approachable. This easiest sampling type is the least reliable.
Purposive Sampling
A sampling where elements have no known probability of being selected and where certain protocols are applied in implementation of the sampling process is called Purposive sampling. Two of its types include:
Judgment Sampling
A sampling type in which the researcher cheery picks the individuals to seek the information that is limited among only a few individuals in the society. This is primarily applicable when the research is carried out on certain special or unique tasks (Cooper & Schindler, 2014).
QuotaSampling
A type of purposive sampling in which a sample is selected based on the ratio of representation each class has in the total population. So a population with 80-20 representation of management and janitorial level staff in a business environment will come into taking same representation (80-20%) when sampling on quota basis is considered (Cooper & Schindler, 2014).
Snowball Sampling
A non-probability sampling type which works more like a referral system in which the subject is selected based on certain traits and is then asked to assist in figuring out more subjects with the same traits. This is useful when obtaining subject is difficult owing to their scarcity (Snowball Sampling, 2009)
Questionnaire
A research tool where in questions in series are asked from the respondents to gather information on the subject matter. Therefore, the questions must be designed and arranged in a manner that serves the inquisitive nature of the questionnaire over a particular domain. Following are the varied types of questions a questionnaire may incorporate
Types of Questions
* Free Response Questions
Also called open-ended questions, they mainly provide the leisure to answer the given question in own words, without having to limit the answer to certain available options (Questionnaire Design 2009)
* Dichotomous Questions
This type of questions usually has options representing the two extreme poles. However, when one of the two extreme poles is likely to be rejected or not taken by the respondents, it is wise to have one moderate option coupled with an extreme one to enable equal selection opportunity for each option and eradicating any chances for errors (Cooper & Schindler, 2014).
* Multiple Choice Questions
It’s a question type in which more than two options are available while only a single option can be chosen from the alternatives. It is supposedly a wise choice to make when the questionnaire preparer has a sound view of all possible options a question may entail. However, at times the options may be mutually exclusive while the respondent seeks otherwise (Cooper & Schindler, 2014).
* Checklist
It’s a question type in which the respondent has leisure to choose multiple options from the given array of choices without having to showcase any preference among the choices being made.
* Rating Questions
...