UWC Executive Director Valerie Balester is beginning the process of assessing the effectiveness of the Wcourse program, an undertaking made more difficult by the organic nature of how students learn to write.

As we move into the second phase of implementing writing-intensive courses at Texas A&M, we naturally want to know how well we are doing. Fortunately, research data about writing instruction can help us design an assessment. We know, for instance, that requiring more writing will not alone improve writing ability. Writers need guidance, and, as David E. Harris and Robert Schaible point out in their review of research, writers must be evaluated not only on content, but also on organization, style, grammar, spelling, and presentation.

W courses have been designed with such research in mind. Now we need to evaluate if our efforts are paying off at Texas A&M. Are our students better writers as a result of taking W courses?

It seems a simple enough question with a simple enough means of assessment: test student writing at the beginning of a W course, re-test at the end, and measure improvement. Unfortunately, many factors complicate the process.

Let’s start with the obvious: the inherent difficulties of measuring writing ability. With careful definition of a writing task and training of raters, a reliable measure of a particular document can be achieved. However, even if students can write one type of document well in one instance, they may not be equally competent writing other types of documents. Measuring a single performance on a single assignment provides only a partial view of writing ability. One way to address this limitation is to examine portfolios of student work, written and revised over time.

Another, perhaps less obvious, complication of assessing writing is that writing skill develops over time and at different rates. Some students show only minimal improvement over the course of a semester, with more substantial results becoming apparent only later, assuming that the students continue to write. Further, other variables can affect performance, from time-on-task and prior knowledge of a subject to how seriously students view the writing requirement.

Our expectations, therefore, must be realistic, and our measurements should not be limited to a single semester or too narrowly focused on individuals.

By the same token, we should not seek results that define better writing merely by better written products. We could achieve better written products easily enough by editing our students’ work for them. But that wouldn’t make them better writ- ers. To help our students achieve that, we must:

1. 1. Increase their repertoire of rhetorical strategies, for example, by enhancing their ability to analyze and address various audiences and understand specific genres.
2. 2. Influence their writing habits and processes, for example, by encouraging them to plan, revise, edit, and proofread, and to seek out and value feedback.
3. 3. Improve their attitudes about writing, for example, by decreasing writing anxiety and raising awareness about the importance of writing to their future.

To measure changes in strategies, habits, and attitudes, our assessment should also include indirect measures of writing ability, such as reflective writing or surveys of students and of faculty who observe them.

Over the course of the next year, I will devise an assessment strategy for the W course program. It will be a realistic plan that will include student writing and surveys of students and faculty, as well as data from sources such as the National Survey of Student Engagement conducted on this campus. What we learn will help us decide how to invest wisely in a program we cannot afford to do without. In the end, I know we’ll discover new questions we’ll want to ask. After all, assessment, if it is to be of use, must be ongoing.