Fluency and Placement Ratings in Second Language Writing

Sae Rhim Oh, Second Language Studies

Fluency has been traditionally investigated in L2 oral production, but the construct has attracted increasing attention in L2 writing as well (Chenoweth & Hayes, 2001; Polio, 2001). The present study investigated the relationship between fluency and placement rating scores in 67 L2 writing samples elicited on a 3-hour university-wide writing test.  The purpose was twofold: (a) to determine the extent to which fluency could predict students’ placement in two curricular levels and (b) to examine the relationship between two measures commonly used as a gauge of fluency: the number of words per essay and the mean length of T-unit (Polio, 2001).  The result of a discriminant analysis showed that both fluency measures had high predictive value for essays placed into the lower level but not for those placed into the higher level, thus supporting the contention that fluency may be less important at higher levels of proficiency (Larsen-Freeman, 1978). On the other hand, a low correlation between the two measures investigated suggested that the number of words and the mean length of the T-unit might not be measuring the same construct, thus contradicting claims by Wolfe-Quintero et al. (1998) that mean length of T-unit should be considered a fluency rather than a complexity measure. The implications of these findings will be discussed in terms of the role that fluency may play in a comprehensive theory of L2 writing competence and also in terms of recommendations for L2 writing pedagogy and assessment.