11. Recommendations for Replication and Accurate Analysis

  1. Micah Altman,
  2. Jeff Gill and
  3. Michael P. McDonald

Published Online: 22 JAN 2004

DOI: 10.1002/0471475769.ch11

Numerical Issues in Statistical Computing for the Social Scientist

Numerical Issues in Statistical Computing for the Social Scientist

How to Cite

Altman, M., Gill, J. and McDonald, M. P. (2003) Recommendations for Replication and Accurate Analysis, in Numerical Issues in Statistical Computing for the Social Scientist, John Wiley & Sons, Inc., Hoboken, NJ, USA. doi: 10.1002/0471475769.ch11

Publication History

  1. Published Online: 22 JAN 2004
  2. Published Print: 12 DEC 2003

Book Series:

  1. Wiley Series in Probability and Statistics

Book Series Editors:

  1. Walter A. Shewhart and
  2. Samuel S. Wilks

ISBN Information

Print ISBN: 9780471236337

Online ISBN: 9780471475767

SEARCH

Keywords:

  • replication;
  • reproduction;
  • verification;
  • secondary analysis;
  • canonical data source recommendations

Summary

In this chapter we summarize our findings through the lens of replication. We discuss what steps are need to be taken in order to ensure reliable replication of research, with particular attention paid to issues related to statistical software. Topics include:

  • definition of terms relating to replication

  • cannonical data sources and data versioning

  • issues related to statistical software, such as data inputting, model specification, and algorithm implementation

  • recommendations to ensure reliable replication

  • recommendations to improve numerical accuracy, with specific recommendations for non-linear models, Hessian inversion, MCMC estimation, logistic regression, and spatial regression