Durkheim used suicide to demonstrate that a scientific sociology was possible. Using official statistics, it would study social facts that shape behaviour. Two social facts, integration and regulation, determine the type and level of suicide.
Douglas criticises Durkheim for using statistics, since these are merely the product of coroners' labels, and for ignoring actors' meanings. He recommends qualitative case studies to discover these and the real rate of suicide.
Atkinson disagrees that it is possible to discover the real rate. Instead we should seek to uncover the common-sense theories and assumptions that coroners use in reaching a verdict, such as ideas about the mode of death.
Taylor's realist approach aims to reveal the underlying structures of meaning that cause suicide. He classifies suicides according to degree of certainty and whether they are self or other-directed.