ERIC Number: ED173203
Record Type: RIE
Publication Date: 1977-Dec
Reference Count: 0
Final Report for Dynamic Models for Causal Analysis of Panel Data. Methodological Overview. Part II, Chapter 1.
Hannan, Michael T.
This technical document, part of a series of chapters described in SO 011 759, describes a basic model of panel analysis used in a study of the causes of institutional and structural change in nations. Panel analysis is defined as a record of state occupancy of a sample of units at two or more points in time; for example, voters disclose voting intentions in a sequence of surveys leading up to an election. The author first defends use of the model by comparing it with a more complex alternative. Three arguments for using the basic model are presented: that the basic model approximates the true causal structure, that the model relates levels of X and Y to changes in X and Y, and that the basic model is a reduced-form of the proper model which allows for causal effects over the study period. The next section identifies special methodological issues, outlines strategies for addressing them, and cites relevant technical literatures containing more detailed treatment. Issues include autocorrelation of disturbances (variables omitted from the study such as material infrastructure), technology, cultural organization, national history, the assumption of constant error variance, measurement errors, and functional forms of relationships. (Author/KC)
Publication Type: Reports - Research
Education Level: N/A
Sponsor: National Inst. of Education (DHEW), Washington, DC.; National Science Foundation, Washington, DC.
Authoring Institution: Center for Advanced Study in the Behavioral Sciences, Stanford, CA.
Note: For related documents, see SO 011 759-772