Use of Non-probability Samples in Management Research - Essay Example

Student Name: Tutor: Title: Use of Non-probability Samples in Management Research Course: Introduction Use of samples makes it possible for research to be manageable. When a researcher picks a sample they select a number of cases from the social whole for the purpose of observation or enumeration. Sampling enables researchers to save money and time. When sampling is done correctly it does not prevent the making of generalizations concerning the population where the sample was taken from. Poor sampling can have serious consequences since the inferences that will be made will be wrong (Zea, 2010). Sampling refers to the procedure of selecting units from the population of interest in order to study the sample and make fairly generalized results of the population where the sample was taken from. The study population refers to the population where the sample is obtained from. In many cases the population is always huge that trying to study the entire population is usually impossible or impractical. Consequently, sample subjects provide a researcher with a manageable as well as representative subset of the population. In non-probability sampling the researcher has no way of specifying the probability of every item inclusion in the sample and there is no any assurance that every item has a chance of being selected (Wilcox, 2010). When the set of units in the population has no any chance of being included in the sample, this means that the definition of the population has to be restricted. Where the traits of the subjects are unknown hence the precise nature of the population is not distinguishable. 1. Non-probability samples and probability samples A sample is a representative or a subset of a larger group referred to as the population. Nonprobability samples do not have random selection. In nonprobability sampling the subjects are selected to be part of the sample in non-random ways. In nonprobability sampling there is the assumption that there is an equal or even distribution of the characteristics within the population. Probability sampling involves the use of random selection of the sample. The person conducting the research has the belief that any sample will be representative and the results will be accurate (Bryman, 2008). The researcher is not interested in getting a sample that is representative of the population. In random selection a researcher has to define the procedure or process that makes sure that the different subjects within the population possess equal chances of being selected. Probability sampling is applied where a researcher is looking for a strong correspondence between the sample drawn and the research population. Stronger correspondence shows greater level of confidence that variations, patterns and trends are the true representative of the research population. It is not possible to measure Reliability in nonprobability sampling. Data quality can only be attained by comparing the survey results with the information that is available concerning the population under study. There is no assurance that estimates will attain an acceptable degree of error (Bryman, 2008). In many occasions researchers are reluctant to apply nonprobability sampling since there is no way of measuring the resulting sample precision. The common types of nonprobability sampling include quota sampling, judgment sampling, volunteer sampling, and convenience or haphazard sampling. 2. Main benefits and limitations of non-probability samples In any kind of research, accurate random sampling is not always easy to achieve. Most researchers are limited with time, workforce and money and following these limitations, it is nearly impossible to sample the entire population randomly and it is usually necessary to use another sampling technique which is the non-probability sampling technique. Nonprobability sample is not as a result of randomized processes of selection (Henn, Weinstein & Foard, 2006). Nonprobability sampling can be used in the demonstration that a specific trait exists within the population. It is used to confirm a certain trait in the population. Nonprobability sampling can also be applied where the researcher wants to carry out a pilot, qualitative or exploratory study. This kind of sampling is important where randomization is impossible such when the population is almost limitless (Valliant & Dever, 2011). Nonprobability sampling can be used where the researcher does not aim to get results that will not be applied to create generalizations with regard to the whole population. Nonprobability sampling is also significant where the researcher got limited budget, workforce as well as time. This kind of sampling can be used in a particular study that can be carried using randomized, probability sampling (Wilcox, 2010). Nonprobability sampling is also helpful when the descriptive comments concerning a sample are desired. Nonprobability sampling is quick, convenient and inexpensive. There are incidents like in applied social research where it is impractical and unfeasible to carry out probability sampling. Majority of non-sampling techniques need some effort as well as organization to complete, but others such as convenience sampling are carried out casually and do not require a plan of action that is formal. The limitation of nonprobability sampling technique is that unknown part of the entire population was not part of the sample (Bryman, 2008). This means that the sample may or may not represent the population being targeted accurately. Consequently the outcome of the research cannot be applied in generalizations concerning the entire population. 3. When and where non-probability samples are most suitable Nonprobability samples are used both in both professional and academic research. There are occasions when nonprobability samples are more suitable. If there is no time to engage in direct fieldwork, it is important to use nonprobability samples to save time and resources. Both professional and academic researchers use nonprobability samples because of the time constraint and resource constraint (Gray et al, 2007). Non-probability sampling techniques aims to attain a sample that can give the most useful insights that can be gained by researchers into the study being focused on. There are occasions that it may be possible or desirable to get findings that are generalizable to a population. In some occasions it is significant to get insights concerning a group or a particular phenomenon that there is little data available to come up with a sampling frame that is rigorous (Zea, 2010). Consequently a survey can be applied in understanding the experiences, attitudes and practices of a particular group of population. Alternatively it may be important to carry an in-depth research on a smaller number of participants or cases in order to get more complex insights. This will comprise of studying events or people who may not be a representative of the wider population, but who are important to be studied since they represent critical, exceptional or intense illustrations of a specific phenomenon that the researcher is interested in. there might be some sources of data that the researcher can access in a manner that is intensely rich, and which could not be accessed by other researchers. The findings may not be straightforwardly generalizable with regard to the wider population, but have the potential of giving valuable insights. The fundamental purpose is to be able to give something theoretically with regard to the findings from a specific sample that has been studied in a manner that gives insights or raises questions concerning other contexts or cases (Henn, Weinstein & Foard, 2006). Nonprobability samples are mostly used in surveys to find out more about a certain characteristic in the population that the research is familiar with. One main reason why model-based techniques are not frequently applied in surveys is because coming up with suitable models and going ahead to test their assumptions is difficult as well as time-consuming hence requiring adequate statistical expertise and experience (Terhanian & Bremer, 2012). Assumptions have to be evaluated for all the key estimates and a model that will work well for some estimates may not be appropriate for others. Attaining the simplicity exhibited in the course of probability sampling methods for getting multiple estimates is very difficult when it comes to nonprobability sampling techniques. 4. Implications of applying non-probability samples for presentation of findings and research design. An implication of these studies that are based on nonprobability samples have to be designed in a manner that is consciously informed either through theoretical debates or establishment of new theory by applying an approach like grounded theory. The use of members of extended family by Robert Orsi in the study of the twentieth-century American Catholicism is a perfect example of a theoretically useful study that applies a nonprobability sample. He documented his findings in tge book titled Between Heaven and Earth (Orsi, 2005). Whereas the researcher use of his own family data history raises methodological questions, it implies that the complexity and duration of his contact with his people in the research would not be obtained by another researcher. Orsi uses members of his family to give insights into the American history and specific aspects of the population like religion and disability. The findings have broader insights about the American population and are important. When a nonprobability sample is picked and studied in ways that are theoretically thoughtful then it should give insights contexts and lives within similar cultural and social dimensions or where comparable theoretical means is being applied (Strauss, 2009). Problems may come up when studies apply nonprobability samples without having any relationship to any theoretical concepts that enable relating of the sample to the wider social processes and findings are reported with the indication that they are generalizable. Nonprobability samples cannot give generalizable findings, but ways of observing the data that may be useful in other cases or contexts. A range of various methods are used by researchers in nonprobability sampling. They are sometimes referred to as purposive sampling techniques since the researcher internationally chooses the content and size of the sample for the goal of maximizing learning from the population sample (Terhanian & Bremer, 2012) Nonprobability sampling decision in practice are likely to be impacted on by a variety of judgments by the researcher with regard to the cooperation as well as quality of data that can be accessed from the various sources and relevance of data to important and theoretically informed questions that are being addressed. There is no correct sample when it comes to nonprobability sampling. It is appropriate to be inclined towards the richness of the data in addressing the key questions raised by the researcher. It is advisable that during nonprobability studies the researchers have to continue to develop the sample until saturation is achieved in the data (Wilcox, 2010). In some instances this can lead to the researcher focusing more on certain themes within the sample. Nonprobability sampling is useful where time and resources are limited. Nonprobability sampling calls on the researcher to critically think about the implications of the accessible data. Conclusion Understanding procedures for sampling is important for learning the conventional art of what it is to be a social researcher. Nonprobability samples are important in conducting surveys that give useful insights into a characteristic or trait within the wider population that the researcher is familiar with. This kind of approach provides insight into the wider population without using probability sampling. Nonprobability samples can be used where the researcher has limited time and resources to cover the research. The characteristic being investigated within the population has to be strong and easily accessible for research. The researcher has to be versed with the population sample above every other person. Nonprobability samples provide an insight into the wider population. Nonprobability samples are convenient and less time consuming. Limited resources in terms of money, time and workforce make it unattainable to randomly sample the whole population hence requiring the use of nonprobability sampling technique. References Bryman, A. 2008, Social Research Methods, 3rd edition, Oxford University Press, Oxford. Gray, P. S. et al. 2007, The Research Imagination: An Introduction to Qualitative and Quantitative Methods, Cambridge University Press, Cambridge. Henn, M., Weinstein, M. & Foard, N. 2006, A short ıntroduction to social research, Sage, London. Orsi, R. 2005, Between Heaven and Earth: The Religious Worlds People Make and the Scholars Who Study Them, Princeton University Press, Princeton, NJ. Strauss, M.A. 2009, Validity of Cross-National Research Using Unrepresentative Convenience Samples, Survey Practice 43(3). Terhanian, G., & Bremer, J. 2012, A Smarter Way to Select Respondents for Surveys? International Journal of Market Research 54(6):751–780. Valliant, R., & Dever, J.A. 2011. Estimating Propensity Adjustments for Volunteer Web-Surveys, Sociological Methods & Research 40(1):105–137. Wilcox, R. R. 2010, Fundamentals of modern statistical methods: Substantially improving power and accuracy (2nd ed.), Springer, London. Zea, M.C. 2010, Reaction to the Special Issue on Centralizing the Experiences of LGB People of Color in Counseling Psychology, Counseling Psychologist 38(3):425–33. Read More