ERIC Number: ED243974
Record Type: RIE
Publication Date: 1984
Reference Count: 0
Conceptions of Statistical Relations within the Context of Thinking about Causal Relations. 1984:03.
The results concerning conceptions of statistical relations, presented in this report are based on interviews with fourteen nurses and fourteen technicians. The interviews were about a medical and a technical case of a causal relation. The starting point was a statement about the existence of the relation and the subjects were asked what evidence there might and should be behind such a statement. In an unstructured in-depth interview they outlined their conceptions of the problem of evidence in each case by suggesting the kinds of investigations that would be relevant and the statistical relations that might be a part of the evidence. The focus in this report is on the meaning of the statistical relations suggested in each case and its dependence on the context of reasoning about evidence for causal interpretations. Similarities and differences between the conceptions of statistical relations are described in the form of some main categories of conceptions. The relation between the conceptions of statistical relations and the conceptions of evidence for causal interpretations is described. The forms of the statistical relations suggested are compared to the general form of the coefficient of correlation. There are clear differences in the conceptions of statistical relations both within and between the medical and technical cases and amongst and between the nurses and the technicians for each case. (Author)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: Swedish Council for Research in the Humanities and Social Sciences, Stockholm.
Authoring Institution: Goteborg Univ., Molndal (Sweden). Dept. of Education.
Identifiers: Causal Analysis; Data Interpretation; Evidence
Note: Also supported by the Tercentenary Fund of the Bank of Sweden.