Bar Charts : Exploring categorical data
Bar charts effectively and clearly show distribution of categorical data (one of the most common data types), and enable comparison of trends across categories.
How to change the aggregate function
By default, Keshif shows the count of records per each category. You can change the aggregate function to average or sum, using the top part of the dashboard, or the analytics configuration panel.
The example visualizes a list of 5,000 companies.
- First, the chart shows the count of companies per each industry, where "IT services" has the lead with 733 companies.
- Next, the chart shows total revenue, where "Health" industry takes the lead with $22B.
- Next, the function is switched to average revenue. "Energy" takes the lead with $190M average revenue.
- Next, we see the average number of workers. "Human Resources" take the lead with 865 employees in average, and "Security" closely follows with 788 employees in average per company.
- Next, we see the total number of worker. While "Human Resources" still has the lead with 170k employees, "Security" no longer employs a lot of people in total, and "Business Products & Services" companies take second place with 120k employees.
- Next, we switch across different label modes, using the configuration menu, and then within the chart view. This view shows that 16% of the workers in the 5000 companies are in "Human Resources", helping us understand the global distribution in terms of whole, where the complete datasets has 1M workers - noted on the top line in the dashboard.
You can read more about the use of aggregate metrics here.
If your categories are coded in your dataset, you can simply change how each code is displayed, making it much easier to read and share results. Read more.
When you have a Likert-scale data, or ordered information, such as years of experience captured as groups, you can specify a custom order of categories with simple dragging action. Read more.
Here is our guideline if you want to analyze word frequencies in some text data with Keshif.
With an intuitive design unmatched in any other tool, Keshif supports rich and easy analysis of multiple categorical values for each record. Examples include multi-answer questions, genres of movies or tags of articles/resources. Read more.