12. Assessing the Simulator's Reliability and Improving its Construct Validity

  1. Timothy C. Haas

Published Online: 12 JAN 2011

DOI: 10.1002/9780470979334.ch12

Improving Natural Resource Management: Ecological and Political Models

Improving Natural Resource Management: Ecological and Political Models

How to Cite

Haas, T. C. (2011) Assessing the Simulator's Reliability and Improving its Construct Validity, in Improving Natural Resource Management: Ecological and Political Models, John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9780470979334.ch12

Author Information

  1. Sheldon B. Lubar School of Business, University of Wisconsin-Milwaukee, USA

Publication History

  1. Published Online: 12 JAN 2011
  2. Published Print: 4 MAR 2011

ISBN Information

Print ISBN: 9780470661130

Online ISBN: 9780470979334

SEARCH

Keywords:

  • construct validity;
  • ecosystem management tool (EMT) simulator;
  • influence diagrams (IDs);
  • MC hypothesis test;
  • one-step-ahead prediction error rate;
  • sensitivity analysis;
  • simulator reliability

Summary

This chapter contains descriptions of procedures that are part of the politically realistic ecosystem management tool (EMT) that assess the reliability and construct validity of the EMT simulator. It presents a step-by-step procedure for assessing the reliability of a political—ecological system simulator. To support one of these steps, an algorithm is given for performing a sensitivity analysis. The chapter derives an algorithm for estimating the simulator’s one-step-ahead prediction error rate. It also presents instructions using the EMT to conduct an MC hypothesis test of a proposed modification to one or more of the simulator’s influence diagrams (IDs). The chapter discusses the effect on a hypothesis test that an unobserved covariate may have. Such an effect can occur when the data arises from an observational study rather than a designed experiment. A political—ecological data set in almost all cases arises from an observational study.

Controlled Vocabulary Terms

hypothesis testing; influence diagram; random error; sensitivity analysis