Frequency Distribution: Histogram Diagram
Histograms are a great way to visualise data and track key performance indicators because they are so clear and simple to read. They are the preferred method of presenting large amounts of data in a simple straightforward manner. What is a histogram and how does it help you analyse your data?
What is a histogram diagram?
The histogram is a graph that is often used in mathematics and statistics. Histograms are used to measure how frequently values or value ranges appear in a set of data. The horizontal axis typically displays the measured value – either a continuous numerical variable, such as height, distance or time or a discrete, countable value, such as number of items. The vertical axis shows the frequency that this value or value range appears.
Histogram divisions can either be discrete number blocks (1, 2, 3) or, in the case of a range, class intervals or bins (0-10, 10-20, 20-30). The most important thing to remember is that there should be no gaps between the numbers or number ranges – every section of the value range is displayed along the horizontal axis.
For a continuous measured variable, class intervals can be a judgement call or worked out through trial and error. They should be chosen so that the shape of the graph resembles a distribution curve similar to the histograms shown above.
Bar charts and histograms: categorical and quantitative
Histogram diagrams have characteristics in common with traditional bar charts – they both measure frequency and use a similar layout. However, there is a key difference:
- Bar charts measure categorical data: data that can be split into different categories or types
- Histograms measure continuous, quantitative data: data that can be counted
Bar charts are certainly a useful tool to visualise the size of each category, but histograms are a better way to display frequency distribution over a range. Histograms also allow us to better analyse the data set and find its mean, median and mode.
How to read histogram diagrams
Averages: mean, median and mode
Averages can be calculated in three ways. The different methods can give the same or different values depending on the data set involved. Consider this simple dataset:
1, 2, 2, 3, 3, 4, 5, 5, 5, 8
- The mean is the sum of all of the values in the data set divided by the total number of values. For this dataset, the mean is 3.8. When an average is referred to without specifying if it is a mean, median or mode value, it is almost always the mean.
- The median refers to the middle value of the dataset. If there is an even number of values, the midpoint between the two closest values is taken. For this dataset, the median value is 3.5.
- The mode is simply the value that appears most often. For this dataset, the mode is 5.
Three methods of calculation, three different averages. The goal of an average is to work out the central tendency of your data – the value your data clusters around. Looking at the shape of your frequency distribution shows you which average best reflects this central tendency.
WE UNCOVER THE EFFICIENCY OF YOUR WORKFLOW
Optimise your performance with Kanban analyticsSee a dashboard with your data
Frequency Distribution Shapes
The most common frequency distribution type is the normal distribution (also known as a Gaussian distribution or bell curve). This symmetrical shape shows values clustering around the central peak with fewer instances further away. In a normal distribution, the mode, median and mean are the same value.
Datasets can also be skewed to the left (negative) or right (positive). Instead of clustering symmetrically around a central value, much higher or lower values skew the shape of the graph. In these cases, the mode, median and mean are different. For skewed data, the best reflection of the central tendency is the median.
Some histograms will show two peaks. This is known as a bimodal distribution. This distribution indicates that there are two overlapping groups in your dataset. We recommend trying to separate the groups to get a clearer picture of the data.
One of the most important metrics in the Kanban method is the productivity of your team. It is measured by the number of work items delivered over a time period (day, week, month). This metric is known as throughput. The most efficient manner to visualise how throughput varies over time is using Throughput histogram. Tracking your team productivity over time will enable you to measure and improve your capacity to deliver.
Do you use histograms to monitor KPIs? What patterns do you notice in your data? Tell us about your experience in the comments!
Meet the Author
Sonya Siderova is an independent consultant who helps organisations deliver successful projects as a Product Manager and Agile Coach. She is a proud mother of a daughter and a son, and enjoys good food and heavyweight boxing championships. Sonya is a regular blogger and founder at Nave.
Many product managers appreciate the benefits of Kanban, but don’t know how best to prioritize tasks. How do we han… https://t.co/Zi9tBiItL4Follow
High flow efficiency is vital to sustaining a scalable, productive business. Learn how you can calculate, analyze a… https://t.co/OysMenyLeRFollow
Two Kanban roles, Service Request Manager and Service Delivery Manager have emerged to equip teams with important d… https://t.co/X2zVYhq1oeFollow
To make your processes more efficient, first you need to identify where you are falling behind. Learn how to recogn… https://t.co/UbIDPd3tfwFollow
Are long cycle times holding your business back? By following these tips and integrating data-driven analytics into… https://t.co/Xgxmq6CU6PFollow
The main flow metrics in Kanban let you monitor your productivity and give you the information to make data-driven… https://t.co/ndLIyWqNqDFollow
Large teams and complex projects face a common issue – the Kanban board becomes cluttered and hard to read. Learn h… https://t.co/auLny9yNnEFollow
How did supermarket restocking techniques revolutionize project management? Learn the history of Kanban and how it… https://t.co/WZUw8z8UqwFollow
Optimizing workflow efficiency is often at the forefront of project managers’ minds, but eliminating idle time isn’… https://t.co/ZmEK020Yu0Follow
To improve workflows, first you must understand how to identify problem areas. Learn how to recognise the most comm… https://t.co/1cpIY9wVOJFollow
Wasteful activities that add no value to your business, also called Muda, can have a detrimental effect on your pro… https://t.co/RG5ZTG3raqFollow
Though it's easy to focus on less complex tasks first, neglecting work in progress can put your team at immense ris… https://t.co/wi4QFY03AjFollow
The Kanban Method focuses on making iterative, incremental changes to workflows to increase efficiency and producti… https://t.co/He4AtSVJXzFollow
Two Kanban roles, Service Request Manager and Service Delivery Manager have emerged to equip teams with important d… https://t.co/N6uQzpDQLhFollow
What makes the difference between an effective daily standup and one that just wastes your team’s time? Our short g… https://t.co/IJU8u5Let3Follow
Are you always short on time and on budget, even though your team is motivated and skilled? Learn how our team at N… https://t.co/f5PKT4gdq1Follow
In this infographic, we’ll be explaining how to identify bottlenecks and stabilize your workflow with the cumulativ… https://t.co/IyNAPj0C2UFollow
When making forecasts, it is impossible to escape uncertainty. Monte Carlo simulation uses permutation of numbers t… https://t.co/1yB4d3nIaXFollow
Lasting change happens in increments, over time. Learn how applying Kaizen principles brings continuous improvement… https://t.co/D6hRs2LfWmFollow