Probability & Non-Probability Samples - Coursework Example

NON-PROBABILITY SAMPLING IN MANAGEMENT RESEARCH Student’s Name Professor’s Name Institutional Affiliation Date Definition: Probability & Non Probability Samples Probability samples ascertain that each and every subject under study has an equal chance of being selected at any given moment in time (Teddlie & Yu, 2007). It possesses a distinguishable attribute that each and every unit within a given population possesses a fairly-known non-zero probability of being chosen within a given set of sample base. The capacity to allow each and every subject a chance of being selected is indeed a positive attribute as it eliminates the possibility of researchers being biased and depend solely on their own personal opinions and desires in coming up with outcomes (Teddlie & Yu, 2007). Non probability samples do not facilitate the given study outcomes to be generalised from the sample of a given population set (Smith, 1983). It is important to note that non probability sample; the researcher is expected to restrict their study outcomes to the individuals or elements that have been sampled only hence they are not provided with an opportunity to compute particular sampling statistics that avail information in regards to the precision set out in the end results. Main Benefits and Limitations of Using Non-Probability Samples in Management Research Benefits It is important to note that non probability sampling forms a valuable set of sampling methodologies that can be effectively used in management research that seeks to adhere to such aspects qualitative; mixed and qualitative research designs (Mays & Pope, 1995). In the course of adhering to qualitative designs, non probability sampling methodologies like purposive sampling allows management researchers with a rather stronger theoretical rationale behind their specific choice of units that would be included in a sample base. In convenience sampling, the management researcher is allowed the audacity to choose specific units that are deemed to be convenient to them since they do not necessarily call for pre-planning of the selection process (Mays & Pope, 1995). It is crucial to add that convenience sampling seeks to ensure that management researcher enjoys lots of ease evaluating the readily-available sources lists and accessibility of the subjects under study. Notably, such non probability sampling as quota sampling seeks to ensure that the management research attains high levels of ease in the course of executing the sampling research surveys (Mays & Pope, 1995). They are allowed to replace units or rather subjects under study in the event that a respondent is not willing and able to cooperate in providing information. It is significantly less expensive and fosters a speedy collection of informational data. Judgmental sampling is yet another form of non probability sampling techniques that presents a significant number of benefits to a management researcher at hand. Considering the fact that in this form of non-probability sampling method the researcher will select subjects or units under study in regards to the readily-available information, it cuts down the overall cost and time that could have been spend in the preparation of samples (Nielsen, 2010). More to this, the form allows researchers to involve definite and positive aspects that relate to stratification within the underlying sample. Nielsen (2010, p.304) argues that non probability sampling can be indeed useful in exploratory management research activities whereby its main objective rests with finding out whether specific operational issues or challenges actually exists in a much faster and affordable manner. It is especially applicable in the event that there is limited or even lack of research that supports a specific theory and sampling bias of particular non probability sampling is used as a tool to help the management researcher to find solutions needed (Tansey, 2007). In essence, non probability sampling is mostly adopted due to the fact that the procedures adopted in the course of selecting the subjects to be studied to be included in a sample is far much faster; easier and affordable in comparison to when the management researcher opts to use probability sampling. Limitations Despite the aforementioned benefits of non probability sampling for management researchers, those researchers that seeks to adhere to quantitative research design, non probability sampling methodology is considered to be an inferior alternative to probability sampling (Schillewaert, Langerak, & Duhamel, 1998). This is related to the fact that the units under study for purpose of being included into a specific sample are not chosen in a random manner. Due to this hitch, management researchers that use this technique are thus perceived to be feeling that they are being coerced to adopt non-probability sampling methodology due to the mere perception that they cannot effectively utilise probability sampling technique. Consequently, such non-probability sampling technique like convenience sampling lacks the basic aspect of accuracy. This is specifically because the probability of inclusion in this form of sampling methodology is not known for each of the subject under study hence none of the sampling precision statistics can be effectively and efficiently computed (Schreuder, Gregoire, & Weyer, 2001). Notwithstanding, quota sampling, faces the drawback of being biased given that the management researchers might opt to intentionally select non-threatening or an easy-to-approach subjects or even those that are easy to come across and contact. The fact that the quota sampling technique expects that there is extensive development of hypothetical model puts the management researcher in a dilemma to select specific and informational data. In fact, the validity of this hypothetical model required in quota sampling is certainly unreliable and poses a challenge to easily verify (Schreuder, Gregoire, & Weyer, 2001). Another major drawback that relates to quota sampling rests with the assumption that the management researcher has a hard time accessing comparable level of estimates of precision at any given moment in time. Judgmental sampling is a non probability sampling methodology that management particularly focuses on in conducting research surveys and fairly depends on past knowledge and research-based skills that is technically utilised in the process of selecting the subjects or even elements to be engaged in the sampling process (Ritchie, Lewis, & Elam, 2003). In this regards, the aspect of expert judgment that is focused on prior experiences technically means that most of the users of this model is from those management researchers that have had an opportunity to stay close to the events. Limiting the sample to only this type of researchers is a drawback since the subjects that are used in the projection cannot depend on it in case where the attributes of judgmental sampling does not conform to past history and experiences (Ritchie, Lewis, & Elam, 2003). Certainly, considering that it is a non-probability sampling model; the level and direction that relates to sampling errors that is mostly introduced by the management researcher is not reliably measurable while the underlying statistics that measure the overall precision of the estimations are not easily computable. The Appropriate Areas where Non Probability Sampling Can be Used Effectively For most cases, non probability sampling is mostly adopted for both commercial and academic research purposes since its underlying procedures that are utilised in the selection process of subjects that should be included in a given sample is far much faster, easier and affordable in comparison to when opting to go with its counterpart; probability sampling. In fact, this basically true with the case of convenience sampling (Ritchie, Lewis, & Elam, 2003). For those students that are conducting their dissertations within the undergraduate and master’s level like in the event of practicalities will often opt to go for non-probability sampling approach. For those researchers, whether academic or commercial, that prefer a quantitative research design; the adoption of non-probability sampling is extensively perceived as being a rather inferior alternative to its immediate counterpart probability sampling model. However, regardless of the fact that there can be a possibility of engaging probability sampling methodology, the use of non-probability sampling is inevitable since it avails a viable alternative that can be effectively applied and used to get imminent outcome (Barendregt, Van der Poel, & Van de Mheen, 2005). In these regards, non probability is perceived to be appropriate since it makes significant strides to ensure that research adhering to quantitave research design cannot be merely abandoned due to the fact that they can fail to meet the criterion set out for probability sampling techniques or when the cost of meeting such a criterion is deemed to be expensive and time consuming so that there cannot be readily-available sponsors. In the event that researchers opt to go for qualitative research designs, the uses of non probability sampling model like purposive sampling methodology is able to avail researchers with a rather stronger theoretical reasons for their immediate selection of subjects or even cases that should be involved within the specific samples (Barendregt, Van der Poel, & Van de Mheen, 2005). As opposed to utilising probabilistic techniques like random selection in the course of generating a given sample, academicians engage non-probability sampling procedures that entail the clear and concise application of subjective judgements while still drawing on theoretical aspects that involves academic literature as well as practical rationales to effectively attain positive results (Arber, 2001). It is crucial to note that commercial researchers engage non probability sampling for purposes of determining the intricacies that are involved in the sample under study. It is not appropriate to use this sampling technique to attain the aspect of objectivity especially in the course of sample selection process or in cases where there are efforts to formulation generalisations as in the case of statistical inferences. In truth, while it can be a researchers intention to formulate generalisations in regards to the sample under study, it is important that this aspect be viewed more of being a secondary consideration lest it triggers elements of bias and limitation to transferability phenomenon (Arber, 2001). In addition to this, non probability sampling technique is mostly considered appropriate in exploratory research especially in the event that its overall objectives lie in establishing whether an issue is present in a more affordable way (Arber, 2001). For instance, it is possible that academicians might choose to select solely those subjects that are included in the populations sample that are deemed to be portraying the issue that is expected to be explored altogether. In the event that a researcher ascertains the inexistence of a problem in a given biased population sample then there is a high likelihood that it cannot be presented in an unbiased population sample. In this way, researchers can then cut down on unnecessary time consumption and costs that are incurred in the course of evaluating the potential issues that could possibly exist at any given moment in time (Arber, 2001). In overall, in the process of making decisions on whether or not to use non probability sampling; researchers in both academic and commercial fields should be able to establish the influence that would emanate from their specific choice of research strategies and thereby go ahead to adopt appropriate form of non-probability sampling technique for that matter. Potential Issues and Implications of Using Non-Probability Samples for Research Design As noted earlier on within the discussion, the implication of choosing non probability sampling as opposed to probability sampling ascertained that the former is inferior alternative in research design and process (Arber, 2001). Of particular interest to note, despite the fact that non probability sampling methodology can avail valuable sets of information; its underlying level of results can never be generalized to a larger set of population and, also the applications of statistics cannot postulate the reliability of the end outcomes that have been computed. In this regard, the sampling technique suffers from aspects of inaccuracies and constant bias especially in the case of judgmental and quota non probability sampling forms. References List Arber, S., 2001. Designing samples. Researching social life, vol.2, pp.58-82. Barendregt, C., Van der Poel, A. & Van de Mheen, D., 2005. Tracing selection effects in three non-probability samples. European Addiction Research, vol.11, no.3, pp.124-131 Mays, N. & Pope, C., 1995. Rigour and qualitative research. BMJ: British Medical Journal, vol.311, no.6997, p.109 Nielsen, S. 2010. Top management team diversity: A review of theories and methodologies. International Journal of Management Reviews, vol.12, no.3, pp.301-316 Ritchie, J., Lewis, J. & Elam, G., 2003. Designing and selecting samples. Qualitative research practice: A guide for social science students and researchers, vol.2, pp.111-145 Smith TFM 1983, "On the validity of Inferences from Non-random Samples", Journal of the Royal Statistical Society, vol. 146, pp 394-403 Schillewaert, N., Langerak, F. & Duhamel, T., 1998. Non-probability sampling for WWW surveys: a comparison of methods. International Journal of Market Research, vol.40, no.4, p.307 Schreuder, H.T., Gregoire, T.G. & Weyer, J.P., 2001. For what applications can probability and non-probability sampling be used? Environmental Monitoring and Assessment, vol.66, no.3, pp.281-291 Tansey, O., 2007. Process tracing and elite interviewing: a case for non-probability sampling. PS: Political Science & Politics, vol.40, no.4, pp.765-772 Teddlie, C & Yu F. 2007. Mixed Methods Sampling: A Typology with Examples. Journal of Mixed Methods Research, vol.1, no.1, pp. 77-100 Read More