Open School_Facilitator Template_Putting Patients at the Center of Care

educational, not-for-profit uses, provided that the contents are not altered in any way and that proper attribution is given to IHI as the source of

the content. This resource may not be reproduced for commercial, for-profit use in any form or by any means, or republished under any

circumstances, without the written permission of IHI.

Open School

IHI Open School Course Summary Sheet

QI 104: Interpreting Data: Run Charts, Control Charts, and

other Measurement Tools

Lesson 1: How to Display Data on a Run Chart

 A run chart is an essential improvement tool because it displays change over time.

 Steps for drawing a basic run chart include:

o Plot time along the X axis.

o Plot the key measure you’re tracking along the Y axis.

o Label both the X and Y axes, and give the graph a useful title.

o Calculate and place a median of the data on the run chart.

o Add other information as needed, such as a goal line and annotations.

 It’s easy and often sufficient to build a run chart by hand.

 There are many computer programs, such as Microsoft Excel, Libre Office, or Google

Docs that can help you draw a run chart.

 IHI has a run chart template for Microsoft Excel freely available at:

http://app.ihi.org/LMS/Content/77a180e3-18be-4969-a23b-

d0e96e57e39f/Upload/QI104_RunChartTemplate.xls

Lesson 2: How to Learn from Run Charts and Control Charts

 If you want a stable, predictable system, you need to separate common causes of

variation from special causes of variation and remove the special causes.

o Common (random) causes of variation are inherent to the system.

o Special (non-random) causes of variation are due to irregular or

unnatural influences on the system.

 Being able to identify and count runs is the first step for analyzing a run chart.

o A run consists of one or more consecutive data points on the same side of the

median, excluding data points that fall on the median.

IHI Open School Course Summary Sheet | QI 104: Interpreting Data: Run Charts, Control Charts, and Other Measurement Tools | 2

 Applying four simple rules will allow you to identify four types of non-random

patterns in the data displayed on a run chart:

o Rule 1: A shift is six or more consecutive points above or below the median.

o Rule 2: A trend is five or more consecutive points all increasing or

decreasing.

o Rule 3: Too many or too few runs is a non-random number of runs

based on a mathematical formula.

o Rule 4: An astronomical data point is a data point that appears far away

from the others.

 A Shewhart Chart (or control chart) looks like a run chart but has the added

feature of control limits. Data outside the limits indicates special cause variation.

Lesson 3: Histograms, Pareto Charts, and Scatter Plots

 A histogram is a special type of bar chart, used to display the variation in

continuous data — such as time, weight, size, or temperature.

 The Pareto chart (or ordered bar chart) is a type of bar chart on which the

various factors that contribute to an overall effect are arranged in order according to

the magnitude of their effect.

o The Pareto principle refers to the idea that, in many situations, 20 percent

of contributing factors account for 80 percent of the results.

 Ordering the factors by magnitude allows teams to distinguish between the “vital

few” (factors in the 20 percent category) and the “trivial many” (factors in the 80

percent category).

o Focusing improvement efforts on the vital few will have the biggest payoff.

 A scatter plot is a graphic representation of the relationship between two variables.

Scatter diagrams help teams identify and understand cause and effect relationships.