Limits of observational methods

B.Participant observation

A.Structured observations

Observational methods

Lecture 7 (part 2) Research methods

Самостоятельно изучить государственное право античного периода.

Государство и право периода кризиса и разложения рабовладения. Кризис республики. Возникновение территориальных государств и причины перехода к монархической форме правления в конце позднего рабовладельческого периода. Эллинистические правы. Особенности общества периода кризиса и разложения рабовладения. Экзимированная собственность.

Like interviews, observation can be structured or unstructured.

· are most commonly associated with experimental or evaluative research designs.

· can also take place in naturalistic settings. For ex., Michelle Stanworth (1983) - systematically recorded the amount of direct contact time teachers give to male and to female students

However, the vast majority of observational research studies in sociology are unstructured, and most of them use a method called participant observation where the researcher participates directly in the life of the people being studied. The research work involves detailing observations, listening to what is being said and asking questions.

Participant observation is the method most commonly used in ethnographic research designs, however ethnography and participant observation are not the same.

There is a richness of detail in participant observation research that tends to be lacking in other methods.

In participant observation sociologists are able to see for themselves how people behave in their natural contexts. This authentic knowledge means that data from participant observation usually fulfils the key criterion of validity far better than data obtained from other methods.

It also offers flexibility and can provide the basis for inductively generating new theoretical explanations.

‘Chicago School’ of sociology used to claim that participant observation ‘tells it like it is’.

Do you think participant observation always ‘tells it like it is’ or do you think there may be some problems with this view?

① The observer effect

In essence, this means that those being observed may change their behaviour simply because they are being studied. If this happens, and the researcher is not seeing the subjects of the research as they really are but as they want to be seen, then the ecological validity of the research is compromised.

Sometimes researchers try to get round this problem by using covert observational methods and concealing their true identity from the group being studied.

For example, Goffman (1987), Holdaway (1983),Humphries (1970).

This ‘undercover’ research raises ethical issues, as those being studied have not given their consent to the research, and it has the limitation that the researcher is unable to ‘stop the action’ and ask questions freely and openly.

Participant observation methods also tend to be unreliable, data collection is not standardised and, like the unstructured interview, selection of data is very much dependent on the researcher’s subjective views of what should (and should not) be included.

It is also time consuming and, because it is often based on a single case study or a small and non-representative sample, it is hard to generalise from the results.

Furthermore, there are many areas of social life – domestic violence, sexuality, suicide and childhood experiences for example – that cannot usually be studied in this way.