Changes in Grammatical Aspects

Most studies on L2 learners’ written texts, as noted in Chapter 2, focus primarily or exclusively on the grammatical characteristics of those texts, particularly linguistic accuracy, lexical and syntactic complexity, and fluency (i.e., CAF). This has been the case because writing has traditionally been seen as a context for practicing, and demonstrating mastery, of L2 grammar and vocabulary. In addition, many SLA studies analyze L2 learners’ written texts in order to draw inferences about these learners’ L2 proficiency or L2 development as reflected in their writing (Wolfe-Quintero et al., 1998). A key assumption underlying these studies is that as L2 proficiency develops, L2 learners’ texts become more fluent (i.e., longer), more linguistically accurate, and more syntactically and lexically complex (Wolfe-Quintero et al., 1998). In addition, more proficient learners are presumed to produce texts that are more fluent, more linguistically accurate, and more syntactically and lexically complex than those written by less proficient learners. Finally, tasks that vary in terms of their complexity and requirements are expected to affect the fluency, linguistic accuracy, and syntactic and lexical complexity of learners’ texts differently. Given the prominence of CAF in research on L2 learners’ texts, we start by reporting findings concerning changes in fluency, accuracy, syntactic complexity, and lexical complexity.

Fluency

Table 5.1 reports descriptive statistics for fluency by task, test occasion, and ELP group for the full sample. For the full sample, there were significant effects for occasion (F(l,275) = 11.12,p = .002, ij² = .44), task (F(l,275) = 214,p = .0001, rf = .90), ELP group (F(3, 275) = 21.56, p = .0001, rf = .89), and occasion by

TABLE 5.1 Descriptive Statistics for Fluency for Full Sample

task interaction (F(l, 275) = 8.13, p = .007, ip = .16) on fluency. Similar results were obtained for the longitudinal sample. There were also significant main and interaction effects for student (within ELP group) on fluency. Overall, for the full sample, the participants wrote longer essays after instruction (AL = 222.24 words) than they did before instruction (AL = 195.47). Essays on the independent task (AL = 255.11 words) were significantly longer than essays on the integrated task (AL = 160.39). Overall, participants with higher initial ELP wrote longer essays than did participants with lower ELP. All pair wise comparisons for ELP groups were significant at p < .01, except for the difference between the medium and low groups, which was not statistically significant.

The significant interaction between task and occasion indicates that the length of the independent and integrated essays was affected differently by occasion (see Figure 5.1). Follow-up analyses indicated that the independent essays were significantly longer than the integrated essays both before and after instruction. However, while the length of the integrated essays did not differ significantly across test occasions (p = .03), the participants wrote longer essays on the independent task after instruction (A4 = 281.24 words) than they did before instruction (AL = 237.15); this difference was statistically significant t(49) = 4.14, p = .0002.

There were also significant main and interaction effects for student (nested within ELP). These significant effects indicate that students within ELP groups varied significantly in terms of their fluency and that individual students exhibited

FIGURE 5.1 Fluency Index by Occasion and Task for Full Sample

significantly different patterns in terms of fluency changes across tasks and test occasion.

In summary, fluency indices were higher for the independent task than the integrated task. In addition, participants had significantly higher fluency indices after instruction, particularly for the independent task. Participants with higher initial ELP had higher fluency than did those with lower ELP.