Sampling Methods
Sampling method can help researchers choose a sample from the target population. Sampling methods eliminate the sample bias and reduce cost of gathering samples. Nevada National Bank case study uses the sampling methods to analyze users purchasing credit cards in Los Vegas, Nevada. Firat and Fettahoglu (2011) states, “Behavioral finance focus on the systematic financial market implications of psychological decision processes” (p.153). Nevada National Bank case study use sampling methods to identify credit cards’ users for purchase reasons, even though there are privacy and security issues may impact concerns for users.
Population and Sampling Frame
Population allows researchers to gather observation at an individual phenomenon. Population examples include the number of items, heights, weights and outcomes of subjects. For example, the population in this case would the number of users purchasing credit cards in the Nevada National Bank. Another words, 400,000 users are using credit cards and set of all accounts in transactions. Sampling frame offers list of items or population designed in sample. A good illustration is the Nevada National Bank would have the list of users using the credit cards for purchase purpose only. The problem is that there were common errors, such as sampling errors, billing errors and dispute errors. Another good example concerning this case that because 400,000 credit cards has issued in the entire Nevada state, which lead to the sampling errors. This situation causes privacy and security issues relating the credit cards. Iacobucci, D., & Churchhill, Jr. G.A. (2010) states, “the total population is considered to be cases that conform to some designated applications” (p.282).
Simple Random Sample
Simple Random Sample incorporates different factors to complete methods. Simple Random Sample example is 100 users’ information appeared out of 400,000 users’ information in the Nevada state. The types of information available are the name, address, phone number, income and education. Mainly the researchers focus on providing names and phone numbers only. This case also indicates the population out of 400,000 indicates each group of users’ credit card and information gather around the Nevada state. Here are steps to design the simple random sample in this case. It is important to analyze the members involved in the population. Next is to categorize each personal on item either by name or title. Then in this case choose the sample size of the population. Finally, it is good to develop a table of random numbers including other information will be a benefit to this case.
Stratified Sampling
Stratified sampling consisted of the population arranged in small groups. For example, this sampling divides non-credit card and users including age, gender and industry. Another good illustration arranged these groups as purchasers and non-purchasers. These groups can arrange in either one or more characteristics of the population. A third example is to offer surveys to gather information into determine which groups using more than others. The fifth example found in the correlation between credit cards usages versus socioeconomics on features of credit cards. If researchers use this sampling, there are three steps to using for this case study. It is important analyze and measure the strata. Then, the next step is to get the total sample size. The final step is to deliver all samples in the strata. Clarke and Courchane (2005) indicates, “Using stratified sampling are used to designated number of credit cards from each of the strata” (p.6).
Cluster Sampling
Cluster sampling does the same thing as the stratified sampling, but except it divides groups into clusters instead of small groups. For instance, researchers wanted to arrange 400,000 credit card users and arranged by 108 cities offering credit cards, and then these participants arranged in about 3,703 users per city. In order to conduct cluster sampling, here is the process that would be useful in this case. First step is to solve the sum of the credit card users. Then, list the random number, such as 1.2….400,000. The third step is to change numbered and repeat to select other users. All credit cards selected accounts.
Conclusion
In conclusion, Nevada National Bank case study use sampling methods to identify credit cards’ users for purchase reasons, even though there are privacy and security issues may impact concerns for users. The reason cluster sampling is the best method to use because simple and systematic random sampling can never do in emergency-affected populations and only economically justified when reducing cost to overcome the loss. By using a cluster sampling for any case study depends on some factors including characteristics of the population on objectives, effect on the sample size for cluster sampling. This sampling is more feasible in populations but does not require complete lists. The downside of using cluster sampling decreased precision, need a large sample in order to conduct this sampling methods in the case study. Another downside of this sampling is calculations of confidence intervals, and p value is more complicated in case studies.
0 comments