DOI 10.1007/s10590-005-2467-1
Machine Translation (2005) 19: 37–58 © Springer 2006
Methodologies for Measuring the Correlations
between Post-Editing Effort and Machine
Translatability
SHARON O’BRIEN
Centre for Translation and Textual Studies, School of Applied Language and Intercultural
Studies, Dublin City University, Dublin 9, Ireland
E-mail: sharon.obrien@dcu.ie
Abstract. Against the background of a wider research project that aims to investigate the
correlation, if any, between post-editing effort and the presence of negative translatability
indicators in source texts submitted to Machine Translation (MT), this paper sets out to
assess the potential of two methods for measuring the effort involved in post-editing MT
output. The first is based on the use of the keyboard-monitoring program Translog; the
second on Choice Network Analysis (CNA). The paper reviews relevant research in both
machine translatability and MT post-editing, and appraises, in particular, the suitability of
think-aloud protocols in assessing post-editing effort. The combined use of Translog and
CNA is proposed as a way of overcoming some of the difficulties presented by the use
of think-aloud protocols in the current context. Initial results from a study conducted at
Dublin City University confirm that triangulating data from Translog and CNA can cast
light on the temporal, cognitive and technical aspects of post-editing effort.
Key words: post-editing, machine translatability, think-aloud protocol, Translog, choice net-
work analysis, controlled language
1. Introduction
The aim of this paper is to assess the potential of two methodologies, i.e.
keyboard-monitoring using Translog and Choice Network Analysis (CNA),
to portray accurately the temporal, technical and cognitive effort involved
in post-editing machine translation (MT) output. Research to date on
assessing a source text’s suitability for MT and on post-editing is first
summarised. The suitability of an alternative method for measuring post-
editing effort, i.e. think-aloud protocol (TAP), is considered before the
alternative methods of keyboard monitoring and CNA are explained. Some
data are then presented which illustrate how these alternative methodolo-
gies might be used to meet the research objective.