Chapter 4: Cause Analysis

Introduction to Cause Analysis

Cause analysis is the second step in the HPT Model (Figure 4-1) and is based on Thomas F. Gilbert's Behavior Engineering Model ^[1] (see Table 4-1). Although thorough performance analysis yields valuable information about the organization, the environment, and performance gaps, a cause analysis will determine why the performance gap exists. This section of the chapter will provide the reader with an overview of cause analysis. The remaining sections will focus on six performance gap causes that are rooted in either the environment or the individual worker.

 

Figure 4-1: HPT MODEL ” CAUSE ANALYSIS PHASE

Definition and Scope

"Cause analysis is the process of determining the root cause(s) of past, present, or future performance gaps. It follows , but is integrally related to, performance analysis." ^[2] Rosenberg sees cause analysis as the bridge between performance analysis and the appropriate intervention(s) that will eliminate the performance gap. ^[3] Using the roots-of-a-tree analogy, he tells us "to look under the performance gap to discover its roots ... select intervention(s) ... to both feed the high performance roots and eliminate the roots that caused the performance gap." ^[4]

Robinson and Robinson, in linking training to business needs, say that cause analysis "provides us with the data to respond to either of these two questions: What is causing an end-result and/or on-the-job performance deficiency? What might prevent newly learned skills from being transferred to the job?" ^[5]

Historically, theorists like Robert Mager, Peter Pipe, Thomas Gilbert, Geary Rummler, Alan Brache, and Joe Harless suggest that performance problems are essentially deficiencies of knowledge, skills, training, or a host of management deficiencies. ^[6] Allison Rossett identified four kinds of drivers (e.g., causes, barriers, or obstacles): lack of skill, knowledge, and information; flawed incentives, flawed environment; tools and processes; and lack of motivation. ^[7] She defined drivers as "everything that it takes to enable performance to 'grow.'" ^[8] Robinson and Robinson identified three major causes of performance deficiencies: those due to the learner, those due to the manager (boss) of the learner, and those due to the organization. ^[9] Perhaps the most influential listing of performance gap causes comes from Gilbert's work on human competence, especially his Behavior Engineering Model, above. ^[10]

Rothwell and Dean have adapted the Behavior Engineering Model to guide the PT practitioner in the search for causes (see Table 4-2). ^[11] , ^[12] They focus on performance drivers, asking key performance questions and suggesting examples of performance deficiencies.

Table 4-2: BEHAVIOR ENGINEERING MODEL ADAPTATION

Performance Drivers (or Causes)	Performance Questions	Performance Deficiency Examples
Data, information, feedback	How well are people given data, information, feedback to perform when they are needed?	Information not given on a timely basis Lack of feedback mechanisms Little documentation Performance standards are nonexistent Is data tied to performance?
Environment support, resources, tools	How well are people supported with resources, tools, equipment?	Ergonomic deficiencies Inadequate working conditions Tools unavailable or not optimally arranged Insufficient time to get things done
Consequences, incentives, rewards	How well do performers see the results or consequences of what they do? How well are they rewarded or provided with incentives?	Work unrelated to organization's mission and needs Rewards not performance-based Competing incentives Poor performance rewarded
Skills and knowledge	How well do performers' knowledge and skills match performance requirements?	Lack of knowledge, skills, training, education Unable to maneuver in the system
Individual capacity	How well can people perform?	Lack of aptitude , ability, physical or manual dexterity Inadequate job analysis
Motivation and expectation	How well are people motivated to perform? Are expectations realistic?	Boring and punishing performance system Unrealistic payoffs

Conducting a Cause Analysis

Too often PT practitioners are overly confident in their diagnostic abilities , and they jump from performance analysis to intervention selection before thoroughly understanding the cause(s) of the performance gap. Symptoms are often disguised as causes. Without a solid understanding of cause analysis and how it affects work, workers, and workplace, the choice of interventions is flawed. Cause analysis determines why the performance gap exists and how it affects the unique mission, needs, wants, and desires of the organization. Once the cause of a performance problem is known, the PT practitioner can choose the appropriate intervention(s).

Techniques

The following steps can be used to conduct a cause analysis of a performance gap:

1. Identify the driver or cause of the performance gap.

2. Classify the driver or cause by determining where it originates within Rothwell's classification of the environment of human performance: ^[13]

   • Organizational environment

   • Work environment

   • Work

   • Worker

3. Prioritize the driver or cause according to high or low impact on the performance environment.

4. During cause analysis it is important to generate as many cause examples as possible. This gives the opportunity to look at many sides of the problem. If either skills or knowledge is identified as a cause, the example list may include these: employee has forgotten how to use the skill, doesn't have adequate information, lacks training in specific processes and procedures, doesn't know the components of the system, etc. If consequences, incentives, and rewards are identified, the example list may read: poor performance is rewarded, no management planning exists, competing incentives send mixed messages, etc.

5. The PT practitioner should also verify causes and corresponding examples with clients. Ask who, what, when, where, and why questions, and involve them ” clients may shed light on the topic and client support and buyin are necessary for the next phases of the HPT Model.

Tools

There are a variety of tools employed for conducting cause analysis. Table 4-3 illustrates cause analysis tools suggested by Rossett ^[14] and Rothwell. ^[15]

Table 4-3: CAUSE ANALYSIS TOOLS

Name of Tool	Brief Description
Interview	Structured or unstructured Excellent for rapport building and follow-up questions Questions can be open -ended or probing
Observation	Captures current skills and knowledge as well as context Inferences about work are generated Takes time to use effectively
Surveys or Questionnaires	Anonymous Large numbers of people can be surveyed Require clarity, effective directions, user -friendly questions, and skill in constructing
Focus Groups	Structure opportunities for soliciting information Participants must be briefed and debriefed Roles of facilitator and scribe are crucial
Root Cause Analysis	"Traces the causes and effects of accidents or other problems ... tends to be past oriented and focused only on pinpointing the causes of negative performance gaps." [a] Chronology of events reported and recorded on paper Relationship of one event to another becomes clear Result is a wall- sized flowchart Participants questioned to pinpoint root cause of problem [b]
Fishbone Diagram	Cause-and-effect diagram, used in total quality management Past-oriented with focus on identifying negative performance gaps All causes traced to people, policies/procedures, equipment, climate Troubleshooting tool [c]
Portfolio Analysis	Financial management tool with focus on positive performance gaps Develop grid Make decisions based on likelihood of payoff Great for stakeholder involvement [d]
[a] Rothwell, 1996b, p. 170 [b] Rothwell, 1996b, pp. 170 “171 [c] Rothwell, 1996b, pp. 171 “173 [d] Rothwell, 1996b, p. 173

In addition to these tools, flowcharts, histograms, Pareto charts, and run charts are useful for identifying causes of performance gaps. ^[16]

^[1] Gilbert, T. E, 1978, pp. 88 “92

^[2] Rothwell, 1996a, p. 79

^[3] Rosenberg, 1996a

^[4] Rosenberg, 1996a, p. 380

^[5] Robinson and Robinson, 1989, p. 109

^[6] Rothwell, 1996b, pp. 153 “154

^[7] Rossett, 1999, pp. 32 “37

^[8] Rossett, 1999, p. 38

^[9] Robinson and Robinson, 1989

^[10] Gilbert, 1978, p. 88

^[11] Rothwell, 1996b, pp. 13 “14

^[12] Dean, 1997, pp. 45 “51

^[13] Rothwell, 1996b, p. 33

^[14] Rossett, 1992

^[15] Rothwell, 1996b

^[16] Rothwell, 1996b