FL writing development and general proficiency development

1
FL writing development and general FL proficiency
development Pieter de Haan, Kees van
Esch Department of English, Department of
Spanish, Radboud University Nijmegen, The
Netherlands URL http//www.let.ru.nl/p.dehaan/ Em
ail P.deHaan_at_let.ru.nl
Background Attempts have been made in the past
to relate FL writing development to students
proficiency level (cf. de Haan van Esch, 2005).
Proficiency level classes were established on the
basis of teachers holistic assessment of FL
essays in the first year of the university
course. Writing development was established on
the basis of a number of linguistic features in
essays written by a single cohort of students in
three consecutive years. So far it has proved
impossible to establish any clear relationship
between writing development and proficiency
level. For this presentation we established
students proficiency levels on the basis of
general proficiency exam results, and relate
these to the linguistic features found in their
essays in the three consecutive years.
Table 1 Comparing teachers holistic assessment
of essays with general proficiency

Method From a single cohort of Dutch-speaking
students of English, we selected fifteen for whom
we had access to proficiency exam results in the
first three years of the university course, and
who had written three essays each in three
consecutive years on the same topic (stating
their preferred news source, and giving reasons
for their preference), under equal
conditions. Proficiency levels were established
per year on the basis of proficiency exam
results. These included practical grammar, oral
communication skills, writing, and translation
Dutch-English (Figs.1 2 stud.01 least
proficient stud.15 most proficient in first
year). For each year we looked at three
linguistic features which are potentially
indicative of development 1. essay length
(general fluency) 2. word length (lexical
sophistication) and 3. type/token ratio (lexical
variation). These were examined in individual
students per year, and for the entire cohort per
year.
Little correspondence between general proficiency
scores in first year and holistic assessment of
first year essay
Great variation in individual development of FL
proficiency
Weaker students in year 1 appear to become
relatively more proficient in FL than better
students in year 1
Result Proficiency exam results yield a
different classification of proficiency levels
than teachers holistic assessment of essays
(table 1). General proficiency appears to
increase more in relatively weaker students
(Figs. 1 2). However, development of essay
length, mean word length, and type/token ratio
occurs according to different patterns in
individual students (Figs. 3, 4, 5). A
significant difference in essay length is found
only between years 1 and 2 and between years 1
and 3 (Fig. 6 ANOVA F 3.25 P 0.05 df 2,
42). No significant differences are found for
mean word length or type/token ratio. After
division of students into most proficient (top
5), average, and least proficient (bottom 5), a
significant difference in word length is found
only between the least proficient and the most
proficient group (at a .10 Fig. 7). No
further significant differences are found.
Majority are more fluent in year 3 than in year
1 Majority are less fluent in year 3 than in
year 2
Majority use longer words in year 3 than in year
1 Majority use shorter words in year 3 than in
year 2
Majority have higher type/token ratio in year 3
than in year 1 Majority have higher type/token
ratio in year 3 than in year 2
Fig 6 Essay length by year
Fig 7 Word length by proficiency class
Conclusion The data in this study show only
partial relationship between general proficiency
development and the three linguistic features
examined in the student essays. We recommend
studying individual lexical features (lexical
density, use of discourse markers and hedges) and
syntactic items (length and complexity of
T-units) in order to assess writing development.
Reference de Haan, P. K. van Esch (2005) The
development of writing in English and Spanish as
foreign languages. Assessing Writing, 10,
100116.