Cost Effectiveness of Usability and Human Error

Abbas Moallem, Ph.D.

Outlines

Why Develop a User-Friendly Product?

Cost of Errors

Introduction Human Error in HCI

Classification of Human Error

Preventing Human Errors

Designing for Error Tolerances

Why Develop a User-Friendly Product?

The Palm Beach Country "Butterfly Ballot

Cost of Error

An error occurs once per week (0.2 per day).

This error delays user 2 minutes in performing a task. (0.4 minute per day)

Example Error Messages

Cost of Error

An error occur once per week (0.2 per day).

This error delays user 2 minutes in performing a task. (0.4 minute per day)

Benefits of Reducing Error

Evaluation of an interface to identify the error

20hr @ $120/hr = $2400

Programming cost to implement suggestions:

30hr @ $120/hr = $3600

Total Cost of DesignTotal cost of error

$6000 $191,666.000

Questions

What is error?

What is Error?

What is Mistake?

What is Slip?

Errors are attributable to individual.

Error reveals what a person was really thinking but did not whish to disclose.

Introduction

Error is Human.

If an error is possible, someone will make it. (Norman (1988)

Classifying Human Error

According the Webster's New World Dictionary “error" are:

 the state of believing what is untrue, and

something incorrectly done

Researcher suggest two definitions of error by defining error as:

 a divergence between the action actually performed and the action that should have been performed, or

 an action or event

Classifying Human Error

Action sequence triggered by knowledge structures (organized as memory units and called schemas). The mind comprises a hierarchy of schema that are are invoked (or triggered) if particular condition satisfied or events occur. (Norman 1981)

Apply more to skilled behavior.

Classifying Human Error

Another method of error classification is developed by Reason incorporating slips, lapses, and mistakes.

Slips and lapses are defined by attentional failures and memory failures, respectively.

Both slips and lapses are examples of where the action was unintended, whereas mistakes are associated with intended action. (Reason 1990)

takes the information processing framework to consider the implication of psychological mechanisms in error formation.

With mistakes the situation assessment and/or planning are poor, whereas the retrieval action execution is good.

With slips, the action execution is poor, whereas the situation assessment and planning are good.

With lapses, the situation assessment and action execution are good, but memory is poor. (Wickens 1992)

Classifying Human Error

knowledge of how the world works (e.g., mental models) leads to the anticipation of certain kinds of information, which in turn directs behavior to seek out certain kinds of information and provide a ready means of interpretation.

During the course of events, as the environment is sampled, the information serves to up date and modify the internal, cognitive schema of the world that will again direct further search. (Neisser 1976)

Classifying Human Error

knowledge of how the world works (e.g., mental models) leads to the anticipation of certain kinds of information, which in turn directs behavior to seek out certain kinds of information and provide a ready means of interpretation. During the course of events, as the environment is sampled, the information serves to up date and modify the internal, cognitive schema of the world that will again direct further search. (Neisser 1976)

Classification of Human Error

Error Phenotypes: Observable State Undesirable

Error Genotype: Generative Mechanisms of Observable States

Classification of Human Error

Classification of Errors by their Consequences

Classification of Error by their Underling

Slips and Mistakes

Cognitive Control and Systemic Error

Errors Relating to Learning Adaptation

Errors due Interfaces among Competing Cognitive Control Structures

Errors due to Lack of Resources

Errors due to Intrinsic Human Variability

Error Shaping factors

Classification of Errors by their Consequences

Describe Superficial Nature of error

Example:

Omissions, Substitutions

Pressing “delete”instead of “end”

Analyze “How” the Error has Happen rather than “Why”

Classification of Error by their Underling Causes

Dissimilar Errors Forms may have Similar Causes

Analyze Cognitive base of Errors

Slips and Mistakes

Slips: Failure of Execution

Mistakes: Planning Failures (the action may go as planed but the plan itself is wrong)

How to reduce errors

Making the limit of acceptable performance visible to users, while the effects are still observable and reversible

Provide feedback on the effects of actions to allow the users to cope with the delay between the execution of an intention and the observation of its effects

Make overview displays available to avoid capture errors at the skill-based level

How to reduce Errors (Conti.)

Support memory with externalization of the effective mental models

Develop consistent information transformation concepts

Support of memory of items, acts , and data which are not part of an integrated : “gestalt” can be useful

Reduce Errors

Avoiding Mode Errors

Maintaining Consistency

Facilitating Multiple Activities

Reduce in Head Memorizing

Knowledge about Users

Understanding Cultural Issues

Providing Appropriate Metaphors and Icons

Using Colors

On Line Help & Training

More Information

The Design of Everyday Things, Norman, Donald A. Currency Doubleday, 1988, USA

The Handbook of Human –Computer Interactions, Chapter 22 page 489-501

Human Error, J. Reason, Cambridge University Press, UK, 1990

For More Information

Karat, C. (1997). Cost-justifying usability engineering in the software life cycle. In Helander, M., Landauer, T., and Prabhu, P. (Eds), Handbook of Human-Computer Interaction. Elsevier Science, Amsterdam. ·

Karat, C. (1990). Cost-benefit analysis of usability engineering techniques. Proceedings of the Human Factors Society. Orlando. Fl. ·

Norman, D. (1998). The Invisible Computer: Why Good Products Can Fail. MIT Press, MA. ·

Moore, G. (1991). Crossing the Chasm. Harper Business, NY.

Pressman, R.S. (1992). Software Engineering: A Practitioner's Approach. McGraw Hill, NY.

Schlesinger, L.A., and Heskett, J.L. (1991). A service driven service company. Harvard Business Review, 69, 5, 71-81.

Wildstrom, S. (1998). A computer user's manifesto. In Technology and You, Business Week, Sept. 28.

1

COE296/AM/SCU/Session 4