Toolpack Consulting: research-based change including surveys

Direct Observations

Observation is said to have been the basis for Piaget's theories of development, and has been useful in consumer behavior and interface design studies (to name a few). Sometimes, there is no effective alternative to direct observation.


Simply observing people bypasses all the problems of self-report measures: there can be no covering up, no false reports. Direct observation allows people to discuss real, indisputable actions as they occur. For some techniques, such as process consulting, direct observation is necessary and a part of the process.

Direct observation can be used to check the validity of other data collection methods (to reduce bias, observation should be done by someone who does not have an investment in a particular point of view).


Observation takes a great deal of time, preparation, and, therefore, money (except in experimental work where people volunteer their time).

Unless the actions observed are defined very tightly and are very simple, interpretation and coding are needed. This takes time and adds the possibility of bias.

There are also more sampling issues: in addition to which people to observe, there are also issues of when to observe them, where, and during what events. There is also the question of whether the observer should be visible or hidden. One of the major questions with observation is whether people will "play to their audience." The presence of the observer may change what is observed.

Cautionary note

The skills of the observer affects the quality and usefulness of the observations!


Planning should include timing, locations of observations, what to observe, and how to record it. All plans should be specific.

Always remember that the observer may have a strong effect on the observed behavior, and try to build in ways to minimize this source of bias.


Structure could range from the extreme (note the number of clicks) to the general (watch for defensive behavior).

There can be little or no directions on what should be observed and recorded. However, to make observations most effective, there should be a model or general directions to help focus the observer's attention.

The most structure occurs when the observer has very specific instructions with regard to what is being observed and how it will be recorded.

In the middle is using less guidance with regard to observation, but keeping specific rules for recording information.

Copyright © 2001-2024, Toolpack Consulting, LLC. All rights reserved.