Non-Probability Samples Management in Research - Essay Example

Customer’s Name : Customer’s Course : Tutor’s Name : December 6, 2012 : Non probability samples management in research A sample is a chosen item from a certain population. This chosen item is used for further research on the subject matter. In sampling, it is essential to make a good description of the individuals who may be left out during the sampling process or the being or having the biggest representation. In coming up with the sample the researcher needs to consider the sample size in order to employ the best technique. The size of the sample should increase as the variation in the individual item increase. The greater the degree of accuracy desired, the larger should be the sample as it will give the adequate sample for the study sample however cannot represent the sample even if it is carefully selected. This gives the difference between the researched and the true value obtained. As the sample increase and so does the error increase in case the sample has an error. Errors during gathering process are also what results in the errors present in the sample. Well defined concepts need to be drawn to help in arriving at valid conclusion but in management research other cases; a bit of biasness is tolerated (Maholtra & David, 2009). Probability sampling is therefore a mathematical procedure of identifying and selecting from individual observations in a set of population in order to yield a vital knowledge for the purpose of making a conclusion based on the findings. Probability sampling is a situation whereby there are samples which are collected together but they do not give the persons a present in a population a chance of being identified and selected to be used as sample (Reynolds, 84). In this method (non-probability sampling), the samples to be used are taken and selected on the grounds of how accessible they are and also depending on the judgment of the researcher in question. In cases of unequal concentration of units under study, the non-probability sampling is used (Mallhotra, 20). It makes the sampling procedure simple and relatively easier for the researcher. The work is also efficient. Tracking down specific individuals may prove to be cumbersome and therefore that credits the use of this technique. The researchers also have the benefit of using foregone conclusions by other researchers to narrow down and come to comprehensive findings. The benefits in non-probability sampling include: the first one is that it is more accurate due to the fact that the researcher is targeting a specific group under research and not the entire population. The answers from the specific sample will answer for the rest of the population. Dealing with a specific group is less tedious as the researcher will be able to know how to handle his or her samples in that particular case. One will know what it takes to get more and accurate information (Baines, 47). The research will apply the best techniques to come up with substantial data that will help in making quality conclusions. The second benefit is that, it is cheaper compared to probability sampling. This is so because when the research knows the subject sample, the researcher will apply the cost effective methods and techniques to make the research more economic. The researcher will also be able to apply the appropriate equipment which will usually not require much capital to procure. One may for example hire, lend or use any other means that requires less amount of money. The researcher will look for the equipment and method that is less costly in terms of revenue. The third benefit is that, it is of much assistance when it comes to dispersed population. In this case, one does not require going all over the location of the entire population in order to come up with the accurate data. One is only supposed to get the sample from the present population and use it. The researcher will go for available sample which will give the required data in the same way as the whole population. In most cases, dispersed population is tedious to understand and having an overview of it will call for sacrifice. Overview of the sample is vital in understanding the sample (Buckingham, 2011). The fourth advantage is that, it helps in the situations where the frame is not available. Frame lays the bases or it helps in strategizing the research. An ideal research requires a good plan to lay the guideline for the study. In the many researches done, a frame helps in general understanding of the population and therefore know what tools to apply, the manpower to apply and also the required time to cover the whole population. Without it, most researchers are left hanging without the knowledge of where to start. The technique also requires less skill for one to employ during the research. It does not also require formal experience. The findings are also based on personnel judgment and therefore, it can be changed by the researcher in order to suit the situation at hand (Yeager, 740). Some of the limitations of non-probability sampling are; that the selected sample may not represent the population under study. This is disadvantageous as the results of the research may be generalized as they may represent some facts about the population which are not accurate of the entire population. The non-researched sample may represent or have a different view on the subject matter. Failure to have its views denies it the participation which could have given different views yet essential to the study being made. A research always aims at gathering the best from the selected sample; in case of a situation of unrepresented sample, some facts remain hidden to the researcher. This therefore discredits the conclusion drawn so far from the previous sample. The second limitation is that, the conclusions drawn will highly depend on the researcher’s perception. This will be the case because conclusions which are drawn will depend on the researchers understanding of the selected population which may be not proficient, the researcher’s knowledge or the point of view (Churchill, 58). The sample can be biased due to deliberate selection. This is based on personal judgment where the researcher makes own conclusion with consideration o there people’s judgment. It can be brought about by substitution where one uses a different sample from what was supposed to be used as it is not available. The researcher uses the obtained sample as an alternative. These are a move to minimize wastage of money, time and manpower hours. Bias is also brought about by incomplete coverage of the required area of study. One can select a given sample to study it but, at the end of it all, the one doing the study fails to attend to the sample as required. This therefore makes the conclusion arrived at not to be competent enough. The research may also do haphazard collection of data thereby not arriving at a good conclusion. Human beings have a tendency of deviating from the selected sample. If this happens therefore, the data collected will not be a true representative sample. The researcher can also fail to adequate interviewing. The research can be done in haste and therefore it will be incomplete and also it will be misleading as its conclusion done will not be based on fact-findings but it is done on incomplete data from the sample (Reynolds, 85). This type of sampling is mostly used when one knows something about the population under study. This means therefore that, the research should have some prior knowledge in mind about the sample that is currently under study. This knowledge will help the one doing the study to distribute the survey in the most appropriate percentages of the population .The is also used when and where it is only a limited number of persons who are in possession of the required traits for the study. In this case, only few selected persons have them and therefore not enough sample to use as probabilistic methods. Sample under study is thereby small and not adequate to use. Non - probabilistic methods are accommodative enough to be used in such cases. It is not possible to use non - probabilistic methods in the cases where the researcher knows the presence of adequate reliable or a certain organ the researcher knows has enough data for the study or capable of bringing together a good representative sample. The methods are also used in the study of a certain trait in a group or subgroup in a certain population. The researcher in this case may be interested in one single trait in whole population. One goes ahead to take only a certain subgroup. It can also be used in the study of the relationship between subgroups. It allows the researcher to study the relationship between traits in subgroups under study. Some of the methods of non-probability are used due to their accessibility to the user. A sample can be readily available to one doing the study and therefore the researcher decides to use the method. Such cases do simplify the work of the researcher as one does not incur extra expenses of obtaining samples from somewhere else (Bradley, 390). It also saves time required to strategize the sample. They also allow the researcher to get basic data about the particular sample under study and also the trends without unnecessary complications. It assists or it is vital in documenting that a certain trait or quality of a substance occurs in a certain specific sample. It helps in detection of correlation among diverse and different phenomena in certain en environments. The sampling methods are also useful in the description how a sample is different from another that is ideal and randomly selected. The techniques are also useful when one is carrying out the preliminary studies. These are studies carried out prior to a certain study that will be carried out later. Their aim in most cases is to assist one to know what the whole sample plus the study being carried out encompasses. The researcher will therefore be able to know all the areas that requires attention where the study (actual) will be carried out. It gives the outline of the whole research to be carried out. It is a form of briefing to the one carrying out the research. The researcher will come up with the specific equipment required in the study and hire the appropriate personnel or manpower required. After this is done, the researcher can also be able to do the probability sampling using the data. They are also used in the cases of time limited situation to assist as they do not required much time to be wasted in planning how the research will be done, in situations of limited budget where not much money is set aside to do the specific study. The technique is applicable in the areas where, the particular population study is limitless. This means that, the size of what to be studied is so large to accommodate the researcher and his own scope. This discredits the probability sampling leaving the one doing the study with the option of dwelling on the non-probabilistic techniques. In sampling technique, randomizing is the first step but however in other cases it becomes impossible to randomize. This is where the concept of non-probability technique is applied. Research which is not an aimed at generalized to the entire population under study. This kind of research is specifically aimed at giving the results for the specific group only (Thorpe, 57). What is composed in the research is just but what is all about the target group only. Although there is no generalization in this type of method, the samples used can still provide important information. The findings cannot be used as source of reference but are vital when it comes to exploration purpose. For example, the owner of a club may invite some regular customers to come and sample a new brand the owner wants to introduce. The owner inquiries from them what they like and they dislike (Baines, 52). If the owner finds out that a certain portion of young men do prefer the brand, the owner can decide on the times in the course of the week when the young men do visit and provide with the brand. Results from non-probability samples may be or have limitations; however they should not be discredited. The results so given can give information that will provide direction, lead a person to a good research base, enable the researcher to evaluate and see a certain problem in varied angles, and also discover modified and new ways of dealing with the problem under study. Moreover, it can help one to still gather vital information in the shortest time possible. This depends on whether one evaluates the findings through first understanding which groups were left out during the sampling and also with the use of previously made findings by other researchers in order to find out whether the finding so arrived was also got by other researchers. Works cited Baines, Paul. Introducing marketing Research. Chichester; John Wiley, 2002. Print. Buckingham, Alan. Business Research Methods London: Oxford University press 2011. Print. Bradley, Nigel. Marketing Research Tools and Techniques. Oxford, OUP Oxford. 2006. Print. Churchill, Gilbert. Basic Marketing Research London: The Dryden press. 2006. Print. Mallhotra, Naresh. Marketing research: an applied orientation. London: Pearson Education, 2010. Print. Maholtra, Naresh & David, Birks. Marketing research: an applied approach. Harlow; Prentice Hall, 2009. Print. Oakshott, Les. Essential Quantitative Methods for Businesss, Management and research. 2009. Print. Reynolds, Nina, Theoretical Justification of sampling in international Marketing Research. Journal of International Business Studies (2003) 34, 80–89. doi:10.1057/palgrave.jibs.8400000. Yeager, David. Comparing the accuracy of RDD Telephone Surveys and Internet Surveys Conducted with probability and non-probability samples. VoL.75, 709-747. Thorpe, Jackson Management Research. London: Sage Lee 2008. Print. Read More