• Editors' Summary
• Credits

1. Overview
2. Getting Started
3. Define Desired Performance
4. Describe Actual Performance
5. Measure/Describe Performance Gaps
6. Find the Root Causes
• Root Cause Analysis Techniques
• Common Root Causes
7. Select Interventions
8. Implement Interventions
9. Monitor and Evaluate Performance
10. Managing Change

• Figures
• Tables
• Side-bars
• Bibliography

• PI approach offers many benefits
• Pilot project holds lessons
• Helping people do their best
• Burkina Faso: Assessing needs
• India: PI helps private providers
• Using the “Why Tree” in Ghana

• POPLINE
• Media/Materials Clearinghouse
• Other issues
• To order copies
• CCP Home Page

Published by the Population Information Program, Center for Communication Programs, The Johns Hopkins University Bloomberg School of Public Health, 111 Market Place, Suite 310, Baltimore, Maryland 21202, USA

Find the Root Causes

Root cause analysis is the main diagnostic step in the PI process. It is the transition between the description of the problem and the development of solutions. Performance problems need to be attacked at their root, or they will persist. For example, a root cause of the gap in counseling among private providers in India was loss of income. Clients did not pay providers for counseling but only for products (90). The PI facilitators concluded that despite training, clear expectations, and supplies, the gap would not close as long as this root cause, related to the incentive performance factor, remained (see side-bar, Performance Improvement in the Private Sector: India).

Faced with several root causes, stakeholders need to identify the ones that have the greatest effect on performance. The root causes are constraints or bottlenecks in the system or work process. Working on weaker constraints will not help if the most serious bottlenecks in the process remain (27).

The process also coaxes stakeholders to see beyond explanations that they feel they can do nothing about. Staff members often blame lack of funding, bad management, or corruption for problems when there are other causes that they can influence, such as unclear expectations or infrequent performance appraisals (63). An apparent lack of funding could instead be caused by misallocation, poor planning, or poor coordination, which could be corrected. PI facilitators need to encourage positive thinking about causes that can be addressed.

Stakeholders find root causes by discussing the information collected from records, site visits, interviews, and meetings, and by using analysis techniques. Two techniques that have proved useful for reproductive health programs are the Why Tree technique and the cause-and-effect diagram. The two techniques encourage careful inquiry into causes and discourage jumping to conclusions.

The Why Tree. Stakeholders identify chains of causes of a performance gap with the Why Tree technique, also known as the Why-Why-Why technique. When stakeholders can think of no more causes in one chain—no more answers to the question “Why?”—the PI facilitator asks if there are any other causes of the gap and begins another chain. Recorded on paper, the performance gap appears at the top of the page with a root system of causes (see Figure 3).

A project in Ghana to strengthen regional resource teams used the Why Tree technique to explore the lack of supervisory visits to a large proportion of providers. The stakeholders identified two main causes: the resource teams did not know how many supervision visits to make, and they had no transportation. The first cause, lack of knowledge, had three roots: no job description, no support system, and no information during training about frequency of supervision. The second cause, lack of transportation, had one root: no training in proposal writing to get funds for transportation.

Each root of the Why Tree describes a cause of the performance gap, and the lowest item in the root indicates how to address the cause, in this case by drafting a job description, establishing a support system for the Ghana teams, and training (87). The Why Tree technique helped stakeholders uncover an unexpected root cause—lack of training in proposal writing. Such training could help solve the transportation problem and other problems caused by lack of funding. If stakeholders end a chain of causes with health-sector or societal problems that they cannot control, then they address the next higher cause under their control (63).

Cause-and-effect diagrams. Sorting the root causes according to the performance factors suggests the type of solutions that would address the root causes. To help with the sorting, stakeholders can use a cause-and-effect diagram, also known as a fishbone diagram, or an Ishikawa diagram after its inventor, Kaoru Ishikawa (70, 108) (see Figure 4). The spine of the fishbone diagram extends from the performance gap in a box on the right. The long bones extending from the spine stand for the performance factors. Causes are diagrammed on lines extending from each performance factor, and further explanations extend from each cause (98).

Since performance factors overlap, some causes may fit under more than one factor. In the Ghana example in Figure 4, “No supervisor” could be placed under expectations, feedback, or organizational support. Also, the explanation for a cause under one factor can connect that cause to another factor. Thus the lack of transportation, a cause classified under workspace/equipment/supplies, turned out to be related to knowledge/skills and expectations.