Social researchers in many different fields use both surveys and observational approaches in conducting in-depth, qualitative research. Some researchers will use one method or the other or, in some cases, a combination of both. It all depends on the nature of the research that's being conducted.
Yet there are notable differences between observation and surveys. For one thing, they each have a different overriding purpose. Surveys are concerned with identifying the needs of individuals within a certain group. So for instance, researchers conducting surveys for market research companies will be looking to identify and understand the buying habits of a designated group.
Observation, on the other hand, is concerned with recording people's behavior during their normal, everyday lives—what they say, what they do, and so on. This is a common method among anthropologists, who use observation as a way of understanding the patterns of behavior displayed by a particular group such as a tribe or clan.
As the overriding aims of surveys are generally much more modest than those of observation, the individual researcher can maintain a certain distance from those they are surveying. With observation, on the other hand, it pays to become immersed in the culture that one is observing, thus blurring the distinction between the observer and the observed. This kind of participant observation, as it's called, is a more continuous process than the conducting of surveys, which of its nature tends to require one-off communication. Those conducting such surveys do not need to immerse themselves in the daily lives of the people with whom they are dealing.