A lot many times the manager has to step in the shoes of a Business Analyst and understand the data around him.
You can think of a Project Manager looking to baseline metrics that will be achieved from the project or a manager trying to report the process performance achieved over the month or a quarter.
One of the tools that really stands out from the competition is a Box Plot.
Box plots are an excellent tool for conveying variation information in data sets, particularly for detecting and illustrating variation changes between different groups of data.
The # 1 Surprise Trick Out Of An Analyst’s Bag – Yes It’s The Box Plot!!!
Imagine 3 agents making credit card sales each day. Here is credit card applications files by the three agents
You are supposed to analyze this data and find out how these are fairing against each other?
I had be scratching my head!!!
The Surprise Trick Out Of An Analyst’s Bag – Yes It’s The Box Plot!!!
Let us see a box plot of multiple data sets & how we interpret it!
- Depicts the Means of the agents.
- Agent’s ‘C’s performance is better than other associates.
- The * is a outlier (which is away from the normal data set and is an exception)
For running this tool in minitab you can follow the following instructions:
- Click on the ‘Graph’ menu and select ‘Box Plot’ a window with Box plot options will open
- Choose Multiple Y’s Simple(as data belongs to three agents). Double click on the column numbers where the stacked data is stored Click on ‘OK’ button and the box plot is generated
When you hover your mouse over Agent 3 the descriptive Statistics show up: Q1: 45, Q3: 85. Interquartile range or Q3 – Q1: 40. This is important and tells you that 50% of the observations are between 45 and 85. This observation may also prove beneficial for Project Managers trying to baseline targets for their Projects:
- Interquartile Range is an important measure and represents the range or midspread -middle 50% of data
- Usually If the data is spread normally, the mean could be a better representation of baseline targets to be followed
- If the data is skewed the analyst could look at other statistical tools to understand the reasons for skewness
- In some cases where the data is skewed the median of the data set could be a good representation of baseline targets to be followed
Caveat: Although the box plot is a good tool to understand variation and spread of data, it is often used in conjunction with other tools to make an informed decision