"For XXIAO only"
karmam1
week_4sampling_and_data_collection_plan.docx|
Running head: SAMPLING AND DATA COLLECTION PLAN |
1
|
|
Sampling and Data Collection Plan |
2
|
Sampling and Data Collection Plan
Karima Mayer
University of Phoenix
Gibran Rezavi
June 8, 2015
Population and size
The sample population constitute of a variation of different persons with arying characteristics and demographics entirely. The sample size will be drawn from population of 27,377 people from different age groups and who are from different classes in society. They were from different classes of socio-economic life; upper, middle or lower class in society. These variables include tax and the number of drunk driving accidents. Since not all accidents are drunk related, it is significant to comprehend the information within the sample is more widespread than a target group of drunk drivers who consume Samuel and Adams beers. The sample population is to reflect an overall idea of how all of the independent variable, alcohol tax rate will come into play versus the dependent variable of the number of accidents due to drunk drivers.
Sampling element
The sampling element for Samuel Adams Brewery is collected through data mining and with the help of its activity monitoring device, the +FuelBand. The marketing strategy of Samuel Adams Brewery to collect relevant customer data has now been totally shifted to technological resources. Along with other gadgets including iOS and Android apps. Samuel Adams Brewery already uses the data it collects from Samuel Adams Brewery+ Systems to design products and builds its brand strategy. The company plans to turn its data mining venture into intimate, highly personalized marketing strategy. Having now discussed the method of collecting data from Samuel Adams Brewery users, it is also worth mentioning how this data will be stored and how will it be protected. The data collected through data mining will be stored into a centralized database management system whose access will be granted to all onshore and offshore units through an interface depicting customer needs and expectations module.
The sample design involves a simple random sampling, which comprises a random choosing of the individuals within the data set. To keep the idea widespread, and reflect the overview of the number of accidents due to drunk drivers, random sampling is most likely the best form of design about the entire population. To be specific stratified random sampling will be used. They are factors that divide the population into sub-populations, and the measurement of interest is expected to vary among the different sub-populations. It has to be accounted for a sample from the population is selected to obtain a sample that is representative of the population (Black, 2011). Hence, because random sampling is simply because the topic is much too generalized to the population. The sample size of 95 people does not reflect a proportionate sampling (5% of the population size). In light of that fact, it is harmless to accept that this is not an appropriate sample size to round up a conclusion of what the larger picture may look like.
In conjunction with the random sampling and appropriate sample size, it is also important to recognize the validity and reliability of the measures of the data set. Reliability defines the measure as to the extent of how unbiased the test is and also the consistency of the measures. Reliability ensures a stable foundation for the test to be performed and calculated accordingly. The data given within the sample set not bias by any means; therefore, the reliability of the measures is high due to the nature of the survey. For example, if another 50 people, who drink Samuel and Adams beers were chosen for another data set, the results would be nearly the same in whole with another set of varying statistics. Now, if the survey conducted included questions such as, “will there be a correlation in drunken driving accidents based on alcohol tax rates,” then it would be safe to assume that specific variable contains a bias to the answer. Validity goes right along with reliability in this situation because the results in the data set are pure facts (tax rate and the number of accidents).
References
Black, K. (2011). Business Statistics: For Contemporary Decision Making. Business & Economics, 228-230. Sekaran, U., & Bougie, R. (2010). Research Methods for Business: A Skill Building Approach. New York: John Wiley & Sons.
