Computationally Assessing Expert Judgments of Freewriting Quality

Computationally Assessing Expert Judgments of Freewriting Quality

Jennifer L. Weston, Scott A. Crossley, Danielle S. McNamara
DOI: 10.4018/978-1-60960-741-8.ch021
(Individual Chapters)
No Current Special Offers


This study examines the relationship between the linguistic features of freewrites and human assessments of freewrite quality. Freewriting is a prewriting strategy that has received little experimental attention, particularly in terms of linguistic differences between high and low quality freewrites. This study builds upon the authors’ previous study, in which linguistic features of freewrites written by 9th and 11th grade students were included in a model of the freewrites’ quality (Weston, Crossley, & McNamara; 2010). The current study reexamines this model using a larger data set of freewrites. The results show that similar linguistic features reported in the Weston et al. model positively correlate with expert ratings in the new data set. Significant predictors in the current model of freewrite quality were total number of words and stem overlap. In addition, analyses suggest that 11th graders, as compared to 9th graders, wrote higher quality and longer freewrites. Overall, the results of this study support the conclusion that better freewrites are longer and more cohesive than poor freewrites.
Chapter Preview


Arguably the most important skill a student learns is how to write effectively. This notion is supported by a 2001 survey by Light wherein over 90% of professionals responded that writing was essential to their job. Nonetheless, there are many students who leave high school without the necessary proficiency in writing needed to procure a job or to be successful in higher education. One means of increasing writing proficiency is through the instruction and use of writing strategies. The use of strategies can help to activate prior knowledge and lessen the demands on working memory. In addition, the use of writing strategies helps to focus the writer on the steps needed to produce a successful written product. The present study focuses specifically on one common writing strategy: freewriting. Freewriting is a timed writing exercise during which the writer produces as many ideas as possible as quickly as possible with little regard to the rules of structure, grammar, and punctuation (Elbow, 1979). It can take different forms including focused freewriting where a person writes with a topic or prompt in mind (Hinkle & Hinkle, 1990). Freewriting is generally a prewriting task and is often part of planning (Renyolds, 1984). Planning is the first step in many writing tasks and can take many forms, including freewriting, outlining, concept maps, and lists (Loader, 1989; Brondey et al., 1999; Reese & Cumming, 1996; Vinson, 1980).

Our goal in this study is to better understand which linguistic features of a freewrite are related to freewrite quality. Identifying these features is necessary in order to build automated NLP assessments of freewrite quality. Automated freewrite assessment will allow educators and intelligent tutoring systems to provide targeted feedback to writers engaging in freewriting. Better understanding the nature and features of freewrites will also afford future investigations of the relationship between freewrite quality and essay quality. Assuming there is a link between freewrite quality and essay quality, feedback can be designed to help students produce higher quality essays. As such, this study serves as one step toward the overarching goal of providing effective tools that use artificial intelligence to help students learn how to improve their writing and help researchers and educators understand the nature of writing.

Although much has been written on the topic of freewriting, most published research has been anecdotal (Belanoff, 1991; Fontaine, 1991; Haswell, 1991; Sweedler-Brown, 1984). That is to say, the claims made in many freewriting studies are based on little to no experimental data. In addition what little research has been conducted on freewriting has been limited to qualitative research on samples of convenience. In addition, the few experimental studies conducted on freewriting were not investigating the product of freewriting (Hinkle & Hinkle, 1990; Knudson, 1989). Rather, these studies examined freewriting as a comprehension strategy to be used immediately following classroom lectures and thus focused on the effects of freewriting on comprehension scores, not the written products. Thus, these researchers never examined the freewrites that students wrote.

Complete Chapter List

Search this Book: