Observation

In one line

Observation is watching behaviour where it happens and recording it systematically, and it is the only tool that shows what people do rather than what they say they do.

In the wording a student can write in an exam: observation as a tool of data collection is the systematic and purposive watching and recording of behaviour, events and interaction as they occur in their natural setting, with a view to obtaining first-hand data. P. V. Young describes it as a deliberate study through the eye, and distinguishes it from casual seeing by its being planned, recorded, related to a research purpose and subject to checks on validity and reliability.

What makes it scientific rather than casual

Everyone observes; a researcher observes differently, and naming the differences is the first part of the answer.

1. It serves a formulated purpose, decided in advance.
2. It is planned: what is to be observed, where, when, and for how long.
3. It is recorded systematically, immediately, and by a stated procedure, rather than remembered.
4. It is checked for reliability, ideally by a second observer, and for validity against other sources.
5. It is related to a wider body of knowledge, so that what is seen is interpreted rather than merely listed.

The types

Participant observation

The observer joins the group and takes part in its life while observing it.

Its degrees: complete participant, where the observer's role is concealed; participant as observer, where the group knows the researcher's purpose and the researcher takes part; observer as participant, where the researcher is known and participates minimally; and complete observer, where there is no participation.

Its merits: it reaches meanings, motives and the group's own point of view, which no other tool does; it observes behaviour in its natural setting; and it reaches groups that would not answer a questionnaire at all.

Its limitations: the observer's presence changes the group; going native, in which the observer identifies with the group and loses the capacity to analyse it; the difficulty of recording without disrupting; the impossibility of observing more than one small group; findings that cannot be generalised; and, where the role is concealed, serious ethical objections, since the observed have not consented.

Non-participant observation

The observer watches without taking part.

Its merits: greater detachment; easier and more accurate recording; and no risk of going native.

Its limitations: it reaches behaviour and not meaning; the observer may misinterpret what they see for want of the insider's knowledge; and a visible outsider still alters behaviour.

Controlled and uncontrolled observation

Controlled: conducted under prearranged conditions, with a schedule listing what is to be recorded, defined categories and often a time sample, so that observations are comparable and can be counted. Its merit is precision and comparability; its limitation is that a schedule can only record what was anticipated.