NotesFAQContact Us
Search Tips
PDF pending restoration PDF pending restoration
ERIC Number: ED173201
Record Type: RIE
Publication Date: 1978-Dec-27
Pages: 8
Abstractor: N/A
Reference Count: 0
Final Report for Dynamic Models for Causal Analysis of Panel Data. Preface.
Hannan, Michael T.; Tuma, Nancy Brandon
This document introduces research aimed to explore methods that could be used to make inferences about causual effects of educational change over time when data are from an educational panel. This preface, the first in a series of 14 chapters described in SO 011 760-772, discusses an educational research project designed to examine affects of educational change and outlines the structure of the project's final report. First, activities of the research project are discussed. These include: integration and explications of models, methods, and results from existing relevant sociological, economic, and statistical literatures; extensions of these models, methods, and results to handle complications for which existing approaches were inadequate; investigation of the properties of different research designs and of different estimators; and performance of empirical analyses. A description of the project's final three-part report follows the discussion of project activities. Part I addresses broad issues of research design in the study of change. Two general arguments call for collecting data on the processes as well as the occurrence of change and shifting from the common discrete-time approach to continuous-time. Part II concerns analysis of data on quantitative (metric) outcomes. Analyses of common approaches and of issues involved in applying new strategies are discussed. Part III deals with dynamic analysis of qualitative outcomes such as remaining in school and kind of curriculum selected. Advantages of collecting event history data and problems in analyzing event-histories are examined. Effectiveness of Monte Carlo experiments, results of empirical analyses on the effects of schooling and ability on job changes, and research designs that minimize biases which result from ignoring interdependent systems of qualitative outcomes are discussed. (Author/KC)
Publication Type: Reports - Descriptive; Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: National Inst. of Education (DHEW), Washington, DC.
Authoring Institution: Stanford Univ., CA.
Note: For related documents, see SO 011 759-772