Most observation studies begin with casual observation and then use systematic procedures for data collection.
|"People watching" is casual observation (do not confuse casual with causal.) It is what we do most of the time – observing others in a haphazard way. We may be taking mental notes, or may not be paying much attention at all. This is about as far as most people get. Some people are very good at casual observation. Novelists, journalists, and social commentators may be keenly perceptive observers and have a very accurate knowledge of what is going on with people's behavior. Unfortunately, casual observation is subject to bias that can distort information. We notice behavior that fits our stereotypes -- for example, teenage boys in huge baggy pants, and girls with bare midriffs -- without seeing the exceptions, which may in fact be the majority. Nevertheless, casual observation can be an important first step in many behavioral studies.|
Example & self-test #1
Observation versus inference
Casual observations tend to be a mixture of observation and inference. In our common language, we mix the two. People are smiling. They are happy. Right? Maybe not. Think again – a person may be putting on a good face. A guy smiles at a girl. She thinks "he likes me." Maybe, maybe not. We need to disentangle observation from the inference.
More than one observer could agree with the second description on the right. It is scientifically preferable to stick with the more direct bare-bones description. It can be rich in detail, but not rich in inference or elaboration about internal motives.
Observing children in a playground, we might make the following observation:
Timmy was mad at Ricardo. Ricardo could not understand why Timmy was mad at him.
The preceding is filled with inferences. A better description is
Timmy moved toward Ricardo, put his hand on Ricardo's shoulder, and shoved him into the sandbox. Ricardo's eyes opened wide as he fell backward into the sand. He let out a short cry, and tears rolled down his cheeks.
Separating inference from observation requires reflection on your own thought processes -- distinguishing between what you actually saw or heard, and how you interpret it.
Distinguishing between observation and inference moves us toward the process of systematic observation. Systematic observation is setting up our study so that we eliminate or reduce bias.
We set up decision rules ahead of time that reduce inferences. A decision rule is a procedure set in place before we begin data collection. We construct our observational research in such a way that if someone else did it, under the same or similar circumstances, they would come up with the same result. In order to do this, our procedure and decision rules must be clearly described.
It is essential to consider the context of behavior being studied. For example, you might be testing the hypothesis that boys on a playground are more likely to engage in risky behavior (e.g., leaping from high places vs. playing in the sandbox) than girls. It may turn out that more boys are jumping off of high places. That may be due to more boys being in the playground in the first place. To obtain an accurate picture, you need a count of the total number of boys and girls (so you know how many did NOT jump from high places).
Also, you must keep track of the number of different individuals taking the risk. You might find more risk-taking incidents among the boys, but that might be due to one or two risky individuals. In this example, you would need to record the individuals doing the behavior in addition to counting the number of incidents.
For observations to be systematic, they must be reliable. Next section: Establishing reliability