Descriptive Statistics and Interpretation

Vincent Nguyen

August 14, 2014

Dr. Raj Singh

In statistics it is not possible to do obtain information from the whole population due to different factors. Some of the factors include the cost involved, unwillingness of some of individuals of the population to participate, accessibility among others. Due to these facts a small sample within the large population is taken for research purpose. The researcher must understand all the characteristics of the population .In our previous research project we had looked at Base Thrainsson Company which had lost more than 40 million credit cards and other customer information and we wanted to know whether this affected customers’ loyalty to the company. Due to cost involved we settled for population size of 100 individuals and a sample size of 50 though the required sample size can be calculated from sample size.

Population

In statistics population is a big set of objects with related characteristics. It’s a complete set of items or individuals that have at least one common property and can be subjected to statistical analysis (Weir & Cockerham, 1984). There is no a formulae of getting population size hence its got from estimation.

Size

Base Thrainsson has a large population and sampling the whole population will be a very expensive process that’s why for study purposes we choose a smaller population of about 100 individuals.

The target population is the customers of Base Thainsson Company and due to complexities of obtaining data from all concerned customers we just use 100 customer size of the population. The cost of obtaining data is also another factor to consider.

Sampling element

Obtaining data for statistical data is done through several ways. Some of the ways include mining, survey and observation. For our research we thought survey is the best way of obtaining data though it has some challenges. We designed interview question to help us in obtaining required information from customers. The question included the following

What are your names?

· Do you think Base Thrainsson is careless with clients’ information security?

· For how long have you been a customer of Base Thrainsson Company?

· Which method do you use for your shopping (cash/credit card)?

· How was your confidence on the company affected by the breach of security?

· Currently do you shop at Base Thrainsson?

Sample size

Sample size is a subset of a larger set that is popular, which is used for the calculation of estimates and that of the population. Sample size is the number of observation which is used to represent a given population with similar characteristics (Bearden, Sharma & Teel, 1982). Sample size is very important in obtaining accurate data from a population. There is a formula for calculating sample size when you have the standard deviation.

Sampling can be done in different ways including simple, systematic, stratified, cluster and convenience sampling .For this research simple random sampling was used so that every customer in the population had an equal opportunity to be selected. In simple random sampling we have two cases either a person can be selected either once or can be selected and returned to the selection pull. In these two cases the probability of selecting a person can be selected using different formulae as shown below.

Case 1

P= where n is the sample size and N is the population sizes.

Case 2

Sample size calculation

Calculate the sample size using a 95% confidence level, and a 5% margin of error

Z value of 95 confidence level is 1.96

We will pick our to be 0.5

E for 95% confidence level is 0.05

=385

New sample size= where p is the population size

To achieve reliability a large sample size would be required. When we have large sample size it increases the precision of the data obtained.

To ensure that subjects used in the research are protected anonymity is used. Also top secrecy will be maintained to ensure no linkage of information.

The method used to obtain data from the Base Thrainsson’s customers through conducting interview. Also observation method was used.

The data collected will be stored in a protected file in a computer so that unauthorized people don’t get access to it .Authentication will be required to ensure that only the authorized can view the data.

References

Bearden, W. O., Sharma, S., & Teel, J. E. (1982). Sample size effects on chi square and other statistics used in evaluating causal models. Journal of Marketing Research, 425-430.