Special Offer!Use code first15 and
Get 15% off your first order
Table of Contents
Primary research can be defined as any kind of research that the researcher conducts by himself or herself. Some examples of primary research are personal observations, conducting interviews, surveys, and ethnographic research. Primary research is often used to supplement secondary research or in the drafting of one’s project. It is of use in a variety of cases: conducting academic research, personal research and business research. This paper will examine the various sampling methods, the issues involved in obtaining primary data as well as the methods used in getting the data.
Researcher - people/individual who is conducting the research.
Respondents - people who the researchers obtain research information from.
Subgroups - divisions of a population.
Technique - method.
A researcher uses various techniques in order to generalize findings and generate statistics. Sampling can be defined as a process that involves selection of individuals from a large population and drawing conclusions that the sample is the actual representation of the population. The population under research is usually a group of people who possess certain characteristics. The actual population of sampling can be of any size and thus referred to as target population (Fowler 2013). In most cases, the entire population is difficult to sample that is why the researcher chooses to sample the part of the population, which is accessible. According to recent studies, the steps of evaluation of sample quality imply definition of sample fame in these 3 classes:
Characteristics of a good sample frame are comprehensiveness, the probability of selection and efficiency.
Probability sampling. Stuart (1984) defined probability sampling as a sampling technique where each element of the population has a nonzero chance of being selected. Individuals in a population are selected randomly or by chance (Gaillet & Eble 2015). Random sampling is used in order to eliminate biasness that might disarrange the entire research. The five most common types of sampling used in doing educational research are the following.
Simple random sampling. In case of simple random sampling, each member of a population has an equal chance of being selected for sampling. This type is considered the best way to obtain samples from a population. When one has been selected for this sampling, he/she has no other chance of being selected for the sample.
Stratified sampling. It is a technique where certain subgroups known as strata are selected for the sample in a proportion that exists in a population. In stratified sampling, the researcher identifies strata and draws a number of subjects from the stratum. The advantage associated with this type of sampling is that it allows the researcher to conduct research in different groups of the population; thus, the research is likely to be effective. Some researchers define it as a more refined type of sampling than simple random sampling.
Cluster sampling. It allows selection of groups of people or clusters other than individuals. It is the most effective type of sampling very large numbers of the population are involved. Once a cluster has been selected for a study by the researcher, all the individuals in the cluster are sampled (Fowler 2013).
Systematic sampling is the type of sampling that involves getting a sample by taking every fixed number of cases from the population. In this type of sampling, once a certain case has been selected, all the cases from the case are included in the sample.
According to studies conducted, a large sample size of a population has better representatives of a certain population. The most important characteristic of a population is the representatives but not the size. Although some differences might exist between a sample and the population, the difference is likely to be insignificant.
Systematic samples are usually easy to execute, construct and understand. It is effective for studies with a tight budget.
These methods provide researchers with a certain degree of control and sense, which is helpful in studies with strict parameters.
The systematic method assumes that the population size is available and can be approximated. If a researcher wants to study a certain population of, suppose, cows and he/she does not know the actual population of the cows, he/she cannot make conclusions. The researcher cannot select the interval size because there is no record of the size of population.
Second, if the population under research might be following a certain trail of a standardized pattern, there is a risk of obtaining very common results and cases.
A survey carried out during research in the systematic sampling can lead to skipping of some data in the research population; for example, if in a research on dogs, every dog put on a list was small and the others were large, systematic sampling will omit the large dogs.
In addition, there is a higher risk of manipulation because in systematic sampling, researchers can construct systems in a way to improve the likelihood of achieving their target result.
Researchers should be aware of the nature and size of a population before conducting research in the area to ensure that the results obtained are adequate and are not prone to errors of common output.
The researcher should also choose the research method effectively because certain methods are considered more effective than others, for example, the simple random sampling technique.
Incomplete data. Sometimes, people being sampled in order to provide data about a phenomenon can give incomplete records (Walliman 2010, p. 253). In cases of questionnaire administration, people actually send back questionnaires with a bunch of attempt questions. This might happen due to the fear to answer certain personal questions or reveal business information, which can lead to negative consequences.
Benefit from Our Service: Save 25% Along with the first order offer - 15% discount, you save extra 10% since we provide 300 words/page instead of 275 words/page
Lack of honesty. During research, the researcher might not achieve the desired result because sometimes, the interviewee may hide very critical information. This may result from unwanted results. Honesty is an important part of any research as information gathered might be used to describe a certain phenomenon.
Improper analyzation tools. The researcher might fail to use the wanted analyzation procedures thus making the research results inaccurate. The researcher should be sure of the types and tools of analyzation before conducting an analysis.
Failure of tests and experiments. In the process of realization of experiments involving two distinct groups of research, some of the instruments used may fail to provide the correct outcomes.
The above mentioned problems in research can be avoided by the provision of easy questions; questions provided to the interviewee should not be personal, and a questionnaire should have multiple answers in order to make it easy for the people to fill in the form. The researcher should use good and effective analyzation tools in order to gain the required information (White & McBurney 2012, p. 307). Finally, the research should be conducted in an environment where it is not doomed to failure.
According to the case study objectives, the population sample includes all the people of Singapore Orchard Road. It involves a large population of shoppers.
Random sampling method: this was the method used by the student. A sample is termed as unplanned if each person among the inhabitants being sampled possesses the same probability of being incorporated in the sample (Thompson 2012, p. 204). Random selection is the foundation of every good sampling technique. Choice of a sample entails identification of all the individuals from whom the selection will be made. Then the researcher should make a list of the study population from whom to select the sample. The researcher may also allot a number to every character and then pick particular numbers referring to a random sample. The opening point for assortment should be randomly chosen. The student used this kind of sampling to select random people.
This sampling technique was effective and in line with the objective set. It was effective because the student needed to collect information from random people and got all the firsthand information.
The sample is descriptive of the whole population because usually in business, the trend of customer does not change. Through interviews, people are able to give the most quality and clear information. The information gained from the sample is likely to be the truth and thus represent the entire population. The researcher did not consider one gender or age but sampled the whole population. The information was not gained from only one part of the population, which allows assuming that the quality of the research results is high.
Try our Top 30 writers
Benefit from the incredible opportunity at a very reasonable price!Order only for $10.95
Choice of Sampling Method. The selection of a certain sampling technique is principally influenced by the practical consideration. In field work, it is almost impossible to choose a sample that is entirely random. To check a bias that is always present in every sample, the investigator must have some basic knowledge of the study population (Valliant, Dever & Kreuter 2013, p. 48).
Quota sampling. This is a sampling technique where the sample under research has equal proportions of individuals and the rest of the population who have known characteristics. It is implemented in steps, and the first step is a division of the population into subgroups. Then, the researcher identifies the proportions of the subgroups in the sampled population; the proportion is applied in the sampling process (Valliant, Dever & Kreuter 2013, p. 50). Finally, the researcher picks subjects from the subgroups, with key consideration directed towards the proportions identified. Researchers like this kind of sampling because it allows them to sample only a small proportion of a subgroup, which is aimed at determining the characteristics of the subgroup. It also allows the researcher to determine relationships between different subgroups.
Try our VIP support
Benefit from the incredible opportunity at a very reasonable price!Order only for $9.99
Snowball sampling. It is a non-probability method used in a case where the desired characteristics of a sample are rare. It is difficult and costly to locate respondents in this kind of sampling. The student could have employed this kind of sampling to gain information about the buying habits of the people. Although it may be suitable, it may contain certain constraints that are difficult to deal with as the required results can turn out to be unacceptable because researcher does not sample the actual population.
Systematic Sampling. Here, every nth is chosen from the list. This method has a major advantage compared to the simple random sample. The beginning point for assortment should be randomly selected (Thompson 2012, p. 215). This is an alternative method, which can be used by the student to replace random sampling.
Multi-stage sampling method. In a survey covering a large area, the student can use this method in order to gain samples from stage to stage. The method may cause an error if the variable attribute, or disease is itself clustered in the population
Stratified Sampling. Whenever a population is unevenly distributed in relation to sex, age, etc., a stratified sample is used. It is a sampling method where the researcher divides the entire population of research into subgroups or strata; the researcher then selects random subjects from each stratum for investigation. The student could have used this method to gather research information from the groups characterized by age, gender and education level. One of its advantages consists in the fact that the method ensures that every subgroup data in every portion of the population is represented. In addition, information gained by the researcher is even and equal. In stratified sampling, the researcher can obtain data even from small and inaccessible subgroups of the population.
Convenience sampling. It is a procedure used by the researcher due to its convenience and availability. Subjects of the study are sampled because of ease of recruitment and involvement in the study. This kind of sampling is preferable because it is fast, not expensive, easy and the respondents are usually easily accessible (Valliant, Dever & Kreuter 2013, p. 40). The main disadvantage of this sampling technique is that the sample does not represent all the parts of the population, which is inconvenient and gives ground for criticism. Using this sampling technique has the disadvantage of getting inaccurate results as a consequence of systematic bias. This is a state where there is a constant change between the outcomes that the investigator obtains from the whole population and the results from a sample. Another major drawback of this technique is that outcomes achieved are not a representative of the entire population, so they cannot be used to describe the whole population.
Non-probability sampling. This is a kind of sampling technique where the individuals in the population do not have an equal chance of being selected. Researchers have many constraints such as finances, time, and workforce (Valliant, Dever & Kreuter 2013, p. 53). This does not allow them to study the part of the population they are interested in, so they use the easiest and reasonable way.
Sampling techniques differ according to the kind of study being done. Researchers might prefer the most efficient method while others can choose in favor of the cost effective one. As it has been mentioned, when obtaining a sample size, the researcher should consider all the aspects.