Direct observation is a flexible assessment strategy which can be used to systematically obtain both quantitative and qualitative information about ongoing behavior. The various observation methods can be uniquely applied to obtain a wide variety of information types and molded to obtain highly specific information. Consequently, direct observation has wide application in education across age levels, settings, and content areas.

Definition

Observation is defined as the process of gathering information by looking. More specifically, observation involves watching behavior and making a record of the behaviors observed. Direct observation refers to the process of first hand recording of behavior as it occurs in a specific setting. Naturalistic observation involves observations of behaviors as they occur in natural environments and clinical observation occurs in an arranged or contrived setting. Through observation, one may answer questions about what behaviors are performed under what conditions and what stimuli are related to the observed behaviors. 

Rationale

Many reasons have been identified for integrating direct observation into educational evaluation and assessment processes. By employing direct observation, for example, the teacher may (1) obtain a picture of an individual’s spontaneous behavior during daily routines in their natural environments, (2) assess difficult-to-test behaviors and skills, (3)  validate information collected from other measures, (4)  identify functional relationships between the environment and an individual’s behavior, and (5) monitoring intervention effectiveness. These are just a few examples.  Direct observation can also be used to answer a wide range of questions in a variety of settings. For example, direct observation can answer questions about the occurrence or nonoccurrence of distinctive behaviors, behavioral responses to specific conditions, and learning style. Prior to beginning the direct observation process, one should clearly specify the question to be answered by direct observation such the questions of “What behaviors does the student do independently; with support, assistance, adaptations; or not do?” “Under what conditions does a behavior occur or not occur?” and “Is progress being made?”

Techniques for Identifying Behaviors
 

Anecdotal Recording. A typical first step in the process of direct observation is anecdotal recording. An anecdotal record is a brief narrative, written after an event, objectively describing the activities surrounding a behavior of interest. The date, time, and place are recorded as well as an accurate description of the events as they occurred . For example, if a child engages in disruptive behavior during circle time, the teacher may write an anecdotal record objectively describing, in chronological order, the sequence of events that preceded the disruptive behavior, the disruptive behavior, and the events that followed the disruptive behavior. Doing so, the teacher may gain insight into why the behavior occurred and characteristics of the disruptive behavior.

Running Records.  A running record is a technique, which can be used after an anecdotal record, to obtain more precise information about a specific behavior of interest. Instead of being written after the event has occurred, running records are written while the event is occurring. Rapidly describing events in real time sequence can provide insight into antecedents (events preceding the behavior), the behavior itself, and consequences (events following the behavior). Although running records can provide valuable information about an individual’s specific behavior, running records consume time, inevitably overlook some details, and are difficult to use with groups.

Techniques for Measuring Behaviors

Step One: Defining and identifying the relevant dimensions of the behavior. The first step in the direct observation process is to develop a precise definition of the specific behavior observed. Precise definitions help observers know when the behavior is and is not occurring and determine the method to be used when measuring the behavior. 

Step Two: Selecting data collection systems and designing data sheets.  After one selects the behaviors to be the focus of direct observation, one must then select the dimensions of the behavior that are most relevant and easily recorded. Four behavior dimensions frequently addressed in direct observation are: (1) Frequency –  number of times the behavior occurs, (2) Duration – length of time the behavior lasts, (3) Latency – length of time between the end of a cue and the occurrence of a behavior, and (4) Accuracy – extent to which a behavior conforms to a model.  The behavior dimensions measured will vary from behavior to behavior. For example, to measure hitting, one might be interested in frequency of the hitting; to measure crying, the dimensions of duration might be measured; and to measure spelling, latency and accuracy might be of interest. 

In addition to the four basic types of data described above, teachers may need to measure discrete behaviors or categories of behaviors by units of time – interval recording. An interval recording is simply a measure of the occurrence or nonoccurrence of a behavior within specified time intervals. A period of time is selected for observation and divided into equal intervals based on the preferences of the teacher. Time segments can be as short as 5 seconds or as long as 30 minutes depending on the type of information the teacher desires.

Step Three: Selecting appropriate times and situations for observation. In deciding which situations to observe, the following guidelines can be helpful. Observation should occur in situations where (1) the behavior is likely to occur, (2) the student is likely to need the behavior/skill, and (3) generalization and/or application of the behavior is desired. If two or three observations yield similar information, perhaps enough data is available for decision making. However, if three observations yield substantially different information across observations, more information is needed. 

Step Four: Checking the accuracy of data collection.

Step Five: Using the results for decision making. Once gathered, data can be summarized to answer the question driving the observation process. Simply calculating the behavior’s average frequency/ duration/ latency, the percent correct (accuracy), or the rate of occurrence (average number of behaviors per unit of time) may be sufficient for decision making. Other times the data will need to be graphed to reveal subtle increases or decreases in behavior. Systematic analysis of data facilitates decision-making by answering specific questions, addressing accountability issues, and facilitating communication among members of a student’s IEP team.

Venn, J. J. (2000). Assessing students with special needs. Upper Saddle River, NJ: Merrill.