The main strength of observation is that it provides direct access to the social phenomena under consideration. Instead of relying on some kind of self-report, such as asking people what they would do in a certain situation, you actually observe and record their behavior in that situation. This, in principle at least, avoids the wide range of problems associated with self-report. In an interview situation or in response to a questionnaire item, for example, a person may not always provide accurate or complete information, or they might answer in ways that correspond to what is socially desirable. There is a recognized source of bias in self-report techniques referred to as a 'social desirability set', which means that in many spheres of social life there are socially desirable ways of behaving and, consciously or unconsciously, individuals will tend to respond in that way, although in the 'real world' they might behave differently. Asking children whether they would tend to be helpful to other children in certain situations, for example, would be susceptible to such a source of bias.