Histogram

A Histogram, or Frequency Diagram, is a bar chart that shows the frequency of how often a cause or issue occurs. The chart displays the variation of continuous data such as time, weight, size or temperature. These values are distributed along the X-axis of the chart.
Resource Type: 
  • Guide
Topic(s): 

    REFERENCES

     

    Bassard, M., Ritter, D.(2010). The Memory Jogger 2. Tools for Continuous Improvement and Effective Planning.Second Edition.Salem, NH: GOAL/QPC

    Math is Fun.  (2014).  Frequency Distribution.  https://www.mathsisfun.com/data/frequency-distribution.html  Website. Retrieved 25 May 2017

    Sollecito, W. A., Johnson, J. K. (2013). Continuous Quality Improvement in Health Care Fourth Edition.Burlington, MA: Jones & Bartlett Learning.

    Spath, P. L. (2013). Introduction to Healthcare Quality Management, Second Edition.Chicago, IL: Health Administration Press.

    REFERENCES

    A Histogram is Used to:

    • Summarize large data sets graphically.
    • Show the frequencies of occurrence of the data values.
    • Compare measurements to specifications.
    • Assist in decision making for future performance.
    • Provide information for predicting future process performance.
    • Help to answer whether the process is capable of meeting customer requirements.

    Materials You'll Need:

    • Data points
    • Excel spreadsheet or other electronic tools that can compute and display data
    • Calculator
    • Graph paper (if not using an electronic spreadsheet)
    • Pencils/pens

    Some Helpful Terms and Definitions:

    Frequency:
    How often something occurs.

    Frequency Distribution Table:
    A simple table that includes the pattern of the frequencies of all data points.

    Class Interval:
    The grouping of numerical data into equal groups (bars on the chart).

    How to Use the Histogram:

    Process Measure:

    Decide on the process measure. The data should be variable and measured on a continuous scale (temperature, time, weight, speed, size, etc.).

    Project Team:

    Select a team to work on the project.

    Gather Data

    • Collect enough data points (we recommend between 30 and 100). This helps to identify patterns and calculate results.
    • Collect data for a specific period of time: hour, shift, day, week, etc.

    Place Your Data

    Place the data in a table. Example: Number of minutes patients had to wait to get their call answered on day x.
    A 5x8 grid of numeric values ranging from 4 to 28, each cell representing a measured value such as a wait time in minutes.

    Move into One Column

    If using Excel, enter each number in a different cell of a single column. Example:

    Reorganize

    Sort the data points in numerical order. Example:
    This table displays a 40-value dataset arranged in a 5x8 matrix. The values range from 4 to 28 minutes, likely representing observed wait times or cycle durations. This raw data serves as the foundation for subsequent frequency distribution and histogram visualizations.
    An Excel column labeled “Data” shows the same 40 numeric values sorted in ascending order from 4 to 28.

    Determine the Number of Class Intervals (Bars) Needed

    Note: The number of intervals can influence the pattern. Too few will produce a tight pattern, too many will produce a spread out and flat pattern. Our Class Intervals (bars) will be as follows:
    A frequency table with seven rows displaying class intervals (e.g., 1–4 minutes, 5–8 minutes) and corresponding frequencies ranging from 1 to 12.

    Draw a Histogram from the Frequency Table

    • On the vertical line (Y axis) draw the frequency scale to cover the class interval with the highest frequency count. In our example, that number is 14.
    • On the horizontal line (X axis), draw the scale related to what you are measuring.
    • For each class interval, draw a bar with the height equal to the frequency total of that class. See example below.

    Interpret the Histogram

    Review the histogram to determine the spread of the data to determine if your process is capable of meeting your customer requirements.

    Ask the following questions:
    • Where do most of our data points lie?
    • What does this tell us about our process?
    • Does it meet the needs of our patients and employees?
    • Where can we make changes in our process to improve our outcome?

    Example:  Histogram from the data we collected

    A bar chart labeled “Wait time” on the x-axis and “Frequency” on the y-axis. It displays seven vertical bars corresponding to class intervals with varying heights based on frequency, peaking at 13–16 minutes.

    Conclusion: We can observe a normal or symmetrical distribution. Most data points lie between 9 and 20 minutes (with an average wait time of 15.6 minutes). Our goal for the call wait time is 8 minutes or less. Using histogram has helped us to determine, that we need to make changes in our process to improve wait times.

    Improving the Process

    1.  Study the data to find a pattern and to determine the causes. Ask the following questions:
     – Is the recent data within the usual range?
     – Is there a daily (weekly, monthly, yearly) pattern?  

    NOTE: The above example of Patient Call Wait Time was collected over one day. To better understand wait times, it may be necessary to collect data for a longer period of time to be able to look for patterns or similarities. 
    Variables that could be affecting your data:
    – Number of patients during the day
    – Number of staff present
    – Technology errors

    2.  Determine the areas that need correction.

    3.  Utilize a Process Improvement Methodology such as Plan-Do-Study-Act (PDSA) to make improvements. 

    How Can We Help Your Team Work?