The rubric contained four components addressing different aspects of summarization abilities: Main Idea Coverage, Integration, Language Use, and Source Use.
Source text had significant and relatively larger effects than the summarizers' language abilities on summarization performance.
The relationships between various language abilities and summarization performance were examined to see to what extent English reading and writing affected students' English summarization performance.
Given this discrepancy, the present study examined the two important language abilities in summarization to improve our understanding of the construct the task assesses.
Table 8 Summarization performance and language abilities in reading and writing DV IV R2 F sig. Standardized coefficients Beta t sig.
Findings concerning the individual as well as collective contribution of these abilities to summarization performance also, to some extent, contradicted with some previous studies in this line of research (e.g., Asención Delaney 2008; Yu 2008).
Table 7 Correlation between language abilities and summarization performance CET4 reading CET4 writing Summary writing CET4 reading 1.000 CET4 writing.389 1.000 Summary writing.411.605 1.000 n = 60; **Correlation is significant at the 0.01 level (2-tailed).
This verification is meaningful if related to, among other considerations, the caution issued against a danger of "muddied measurement" (Urquhart & Weir 1998, p. 121; Weir 2005, p. 101) in terms of the confounding effects of reading and writing abilities on summarization performance (Alderson et al. 1995).
As the number of documents increases in multi-document summarization, the summarizer gets more difficulties in performing the summarization.
Studies have shown that instructing students in summarization not only enhances their writing ability (Taylor & Beach 1984), but can also facilitate learning (Brown et al. 1981).
