ONESC: a database of orthographic neighbors for Spanish read by children

Behav Res Methods. 2008 Feb;40(1):191-7. doi: 10.3758/brm.40.1.191.

Abstract

In this article, we present a database of orthographic neighbors for words that Spanish children read during elementary education. The reference dictionary for lexical entries and frequencies (which had its origin in Martínez & García, 2004) comprises approximately 100,000 words and is the result of accumulating the words read by a sample of children from first to sixth grades. Using the criterion for orthographic neighbors described by Coltheart, Davelaar, Jonasson, and Besner (1977), we present basic statistics related to neighborhood size as a function of the positions of divergent letters, the cumulative frequency of the neighbors, and the numbers of neighbors of higher, lower, and equal frequency. We also attempt to illustrate and unravel the nature of the relationships among the variables neighborhood size, length, and frequency in the distribution of neighbors. The database described in this article is available at www.psychonomic.org/archive.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Child
  • Data Interpretation, Statistical
  • Databases, Factual
  • Humans
  • Language*
  • Linguistics / statistics & numerical data*
  • Reading
  • Spain