Handwriting and neatness of presentation has long been seen as a contaminating factor in the assessment of writing ability. In particular it has been invoked as a possible reason why girls tend to perform better in relation to boys on free-response writing tests than they do in forced-choice formats in tests of first and second language proficiency.
Over the years there have been a number of studies in the area of first language writing assessment which have investigated the impact of handwriting on overall judgements of writing quality. Some of these have involved correlations of teacher-assigned ratings of writing quality with independent judgements of handwriting (eg. Stewart and Grobe, 1979; Chou, Kirkland and Smith, 1982). Others have involved experimental designs where the same essays are presented to raters in different presentation formats involving good handwriting, poor handwriting and, in some case, typed scripts (Chase, 1968; Marshall and Powers, 1969; Briggs, 1970, Sloan and McGinnis, 1978. Bull and Stevens, 1979, McGuire 1995). The findings indicate in general that the quality of handwriting does have an impact on the scores awarded to essays, and that increased legibility results in higher ratings; in all the studies except that by McGuire (1995), the essays with better handwriting or the typed scripts received higher scores.
Given the great interest over the years in handwriting and its impact on assessments of writing proficiency within the field of first language literacy, it is surprising that there are hardly any studies of the effect of handwriting in the assessment of second language writing. The only study that could be traced involving essays written by non-native speakers, Robinson (1985), produced similar findings to the majority of the first language writing studies; essays written by students whose Ll did not use the Roman alphabet tended to receive lower scores than essays written by 'expert' writers.
The lack of research into the impact of handwriting on assessments of L2 writing proficiency is all the more surprising in a field where reliability and validity issues are generally well understood, and where much attention is paid in the research literature to identifying and examining the impact of construct-irrelevant variance on test scores. One could argue that it is particularly important in formal L2 writing test contexts to examine and evaluate the impact of extraneous variables such as handwriting and presentation, because it is often on the basis of such tests that decisions regarding candidates' future life or study opportunities are made. Moreover, it is particularly in writing contexts such as these, where the writers typically have to write under considerable time pressure, that it may be most difficult for them to control the quality of handwriting and general neatness of layout. It is rare, for example, in formal tests that writers have time to transcribe a draft of the essay into a more legible and well-presented script. Also, as Charney (1984) points out, in a test context the constraints imposed on the rater may result in handwriting playing a larger part in the assessment than it should. Charney argues that the assessment constraints, that is limited time and multiple assessment foci, mean that raters have to read essays rapidly and this may force them to "depend on those characteristics [such as handwriting] in the essays which are easy to pick out but which are irrelevant to 'true writing ability"'.