# Purpose of calculating quartiles, percentiles and correlation coefficient in a business context

purpose of calculating quartiles, percentiles and correlation coefficient of a study in a business context

Did you know that we write custom assignments? We have experts in each specific subject area with vast experience. Get a complete answer and find out more about our writing services.

These three measures are all part of statistical analysis and are invaluable to business in many ways.

Quartiles

Quartiles are points which divide an ordered data set into four equally sized parts. The first quartile is halfway (in number of data points, rather than value) between the smallest value and the median, the second is on the median, and the last is halfway between the median and the maximum value.

Quartiles can quickly be used to provide structure to a dataset and identify groups. This can be used to set targets, for example, targeting the bottom quartile of customer satisfaction scores.

Percentiles

Percentiles can be used to divide an ordered data set into percentage sections. So, a 10th percentile will show the point at which 10% of the data points are below. Thus, the 1% percentile will show the lowest 1% of data points, the 99% percentile will show the majority of data points, with only the top 1% excluded.
This can be useful for a business when measuring aspects of performance. It can show what level of performance is achieved 99% (or 95%, 50%, etc.) of the time, and can provide a comparison to measure a value against (e.g. an applicantâ€™s psychometric test scores).

Correlation Coefficient

Correlation coefficient is calculated to show the strength of a correlation observed between two variables. Thus it shows the strength of the perceived relationship between the variables. The score can range from -1 to 1. The closer to 0, the less strong the correlation is. With 0 implying no correlation whatsoever. 1 implies a perfect positive correlation, and -1 a perfect negative correlation.

Correlation coefficient is a very important part of building predictive models, as it can show whether variables are likely to influence each other. This makes it very important for forecasting in a business, such as sales forecasting.