week_3.doc

Running head: RANDOM SAMPLING 1

RANDOM SAMPLING 2

Random sampling plan

Introduction

Sampling is a process in which a sample is taken from the large population which is known as a subset of population and estimates the uniqueness and individuality of the whole population. The sample can measure the independent characteristics of the whole population by defining and selecting the subset of individuals.

To draw a random sampling plan the following process is followed:

a) Determining the population and size

b) Selecting the target population

c) Identify the sampling element

d) Determining the sample

e) Applying the sampling methods

Simple random sampling

A sampling method is a process for selecting the elements of sample from the target population. A simple random sampling is a sampling method which has all the properties of a population, sample and equal and likely recurrence of all samples. Simple random sampling is widely used for the analysis of sample results with the aid of statistical methods (Black, 2011).

A simple random sample can be obtained with the help of various non statistical methods such as lottery, blind picking etc. Regardless of the sampling type used the first step is to recognize the target population and then determine the accessible population. Once the target population is identified and is accessed then the size of the sample is determined and the sample is selected.

Target Population

For the purpose of studying current decrease in holiday sales revenue due to a lack of gift card sales, the target population would be the customers. It is not possible to research all the customers and find the sample. All the customers may or may not be accessible for the purpose of drafting a sample and therefore a list of accessible family and offices, schools, colleges may be drawn from which the sample will be extracted.

Population and size

To draw a random sampling plan, the population from which the sample is taken is to be identified. Population can be correctly described as the one which includes all individuals, items or characteristics, the understanding or research of which is to be done. It is not possible to study the entire population and it includes a lot of money and time (Kalton, 1983).

Therefore the goal is to find a sample from the population which can be the basis of further studies. This goal of this random sample plan is to identify current decrease in holiday sales revenue due to a lack of gift card sales. Therefore for the purpose of finding sample, a population of target market is analyzed for drawing a sample and further understanding.

After determining the population, a sample size needs to be determined. Sample size is very relevant and is a very critical issue as inappropriate sample size may lead to inaccurate results. Also excessively large sample may waste the resources and time and a relatively smaller sample will not create desired results. The mean of the population may also be taken for finding the sample size.

Sampling element

Once the target population is identified and the population is determined then the sampling element is taken. Sampling element can be found with the help of any of the following methods:

· Data Mining

· Survey

· Observation

For the purpose of extracting a sample element, data mining can be done. In the process of data mining, historical data of sales and gift cards are extracted.

Sample size

Determining the size of the sample is a very critical task in itself and needs to be determined appropriately. The sample size should not be very large or very small. To calculate the sample size the following formula can be used:

Sample size = Z^2 * (p) * (1-P) / C^2

Where:

Z = Z value

p = Percentage of picking

c = confidence interval

For a Population size of 10000, the sample size with 95% level of confidence and 5% margin of error is calculated with the help of this formula as 370.

Method of random sampling

Method of random sampling used in this research is cluster sampling. In this type of sampling clusters are created which can be further subrogated into samples or another finer cluster to obtain a list of sample elements. To select a random sample of 370, the historical data of total sales and purchase of gift cards is taken week wise. Then from the cluster 500 months figure of total revenue and purchase of gift card, 370 random samples are taken for the study.

Conclusion

The data obtained with this sample plan would be subject to a margin of error of 5% with 95% confidence level. The data for this study can be collected using data mining of historical sales and purchase related data. This data can be protected either in physical format or electrical format. For both physical and electrical storage, appropriate precaution strategies need to be taken such as protection of data from theft, protection against virus in case of electrical storage etc.

References

Black, K. (2011). Business Statistics: For Contemporary Decision Making. John Wiley & Sons.

Kalton, G. (1983). Introduction to Survey Sampling. SAGE Publications.

Thompson, S.K. (2012). Sampling. John Wiley & Sons.

Appendix

Calculation of sample size:

Sample size = Z^2 * (p) * (1-P) / C^2 = 376

Where:

Z = Z value (At 95% confidence the value of Z is 1.96)

p = Percentage of picking (for this purpose 50% is assumed)

c = confidence interval (sample size is calculated at 5% margin of error)

N = 10000 (Population)