Comment: British Journal of Mathematical and Statistical Psychology, Jan 1998

Research output: Contribution to journalReview articlepeer-review

1 Citation (Scopus)

Abstract

Robust methods for description and inference about location that are based on down weighting extreme values have very little to offer empirical researchers. Problems of non-normality and heterogeneity are serious, but can be better handled by other sophisticated modern methods. These include exploratory uses of robust methods as an essential and indispensable part of any data analysis. Democracy for data: all points are equal - none are more equal than others.

Original languageEnglish
Pages (from-to)43-46
Number of pages4
JournalBritish Journal of Mathematical and Statistical Psychology
Volume51
Issue number1
DOIs
Publication statusPublished - 1 Jan 1998

Fingerprint

Dive into the research topics of 'Comment: British Journal of Mathematical and Statistical Psychology, Jan 1998'. Together they form a unique fingerprint.

Cite this