Question

5. The data below reports the US credit card industry’s market share data for 2015. You are interested in understanding credit card usage online during the holiday season. Credit Card Market Share (%) Visa 53.9 MasterCard 28.9 American Express 13.2 Discover 4.0 Using your knowledge of sampling methods, comment on how you might select a sample for such a study. Also comment on how you would actually collect the data. What sample size would you use? Be sure to include specific reasons why this method is appropriate and appropriate details for your sampling process.

Answer 1

Sampling method

Before answering the details we need to understand as to what may a sample is?

A sample is a selection of data from the data set (called as population) chosen in such a way that they represent the total population as good as possible.

Thus ,using a correct survey sample size is crucial .A sample that is too big will lead to the waste of precious resources such as time and money, while a sample that is too small will not allow you to gain reliable insights.

So, how large should your sample be? This depends largely on how accurate we want our survey data to be. In other words, how closely we want our results to match those of the entire population.

We need to keep two things in mind.

There are two measures that affect the accurateness of the data

First of all there

• is the margin of error (or confidence intervals). In short, this is the positive and negative deviation that we allow on our survey results for the sample. . Suppose we set our margin of error on 5% for the given set of credit cards.. Now from data we can see that 53.9% of our respondent population like VISA , a 5% margin of error means that we can be ‘sure’ that between 48.9% (53.9%-5) and 58.9% (53.9%+5) of the sample actually likes VISA.
• Second there is the confidence level. This tells you how often the percentage of the population that likes VISA (for example) actually lies within the boundaries of the margin of error.

Once we have decided how accurate we want our sample data to be, we can start calculating how many sample data we will actually need.

With the help of the sample size estimator table shown below we can calculate the sample size .

To start with we can assume the overall population to be indexed at 1000.

For a confidence level of 95% and with margin of error of 5% for 1000 respondent population, I should choose 278 respondents randomly from the population of 1000 for my samplling.

In order to actually do the sampling practically, we also need ot know the response rate i.e the rate of positive feddback to questions asked . Thus if we assume a response rate of say 50% then we need to take about 560 people for sampling so that we actually get about 278 repsonses at 50% response rate.