Estimating the true extent of gender differences in scholastic achievement: A neural network approach

Philipp Manuel Loesche

Intelligence, 2019, vol. 77, issue C

Abstract: In this study neural networks are employed to analyze individual item scores in a large-scale achievement test. They are able to correctly identify the participant's gender in 65.1% of cases, performing much better than a competing model based on differences in subject domain performances. It follows that substantial gender-related information is contained in the items, and comparisons based on performance can only provide a limited view of gender differences in scholastic achievement. An exploratory view of what the networks learn is presented and perspectives for further research are discussed.

Keywords: Gender differences; Sex differences; Neural networks; Deep learning; Scholastic achievement; Large-scale assessment (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0160289619301801
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:intell:v:77:y:2019:i:c:s0160289619301801

DOI: 10.1016/j.intell.2019.101398

Access Statistics for this article

Intelligence is currently edited by R.J. Haier

More articles in Intelligence from Elsevier
Bibliographic data for series maintained by Catherine Liu ().