Academic journal article College Composition and Communication

The Legal and the Local: Using Disparate Impact Analysis to Understand the Consequences of Writing Assessment

Academic journal article College Composition and Communication

The Legal and the Local: Using Disparate Impact Analysis to Understand the Consequences of Writing Assessment

Article excerpt

Demographically, the United States is changing. While people who identify as white, non-Hispanic on the US Census still outnumber other groups (63.4% to 36.6%), the birthrate for people who identify as African American, Native American, Hispanic, Asian, and mixed-race descent now totals more than 50% of all births in the United States (US, Census). This remarkable demographic change in the United States is having a profound impact on education. In some states, such as Texas and California, students of color now outnumber white students in elementary and high schools (National Center). Similarly, at universities across the country, including Rutgers University in Newark, Stanford University, and many of the colleges in the CUNY system, students of color outnumber white students (Rutgers; Stanford; City University). And at institutions like California State University, Long Beach, Hispanic students are now the largest group on campus, comprising 30.8% of the undergraduate population (California State).

Such demographic changes mean that composition researchers must continue to understand how writing assessment practices impact diverse student populations (Haswell and Haswell; Hamp-Lyons; Inoue and Poe, Race and WritingAssessment). For example, we know that students of color are more likely to ex-perience the negative effects of assessment because of rigid institutional requirements (Sternglass; Soliday). We also know that different writing assessment practices may yield quite different results that, in turn, yield different consequences (Kelly-Riley; Inoue). And when it comes to placement testing, we know that enrollment in noncredit, basic writing courses may either support or impede student writers. On one hand, basic writing may be an important, supportive environment for first-year students (Horner), as a number of advances have been made in the last two decades to help struggling writers in ways that situ-ate basic writing students positively within institutional structures (Glau). On the other hand, remedial identity remains defined, in large part, by a model of writing assessment mired in a narrow vision of writing (Condon), including highly constrained lexico-grammatical interpretations of use (Shapiro)-that is, decontextualized drills in grammar and usage. Moreover, students assigned to remedial courses may resist being required to take additional courses, may not enroll in those courses, may complete them at lower rates, and may graduate at much lower rates than their peers (Complete College; Scott-Clayton). Thus, validation tools that help us understand the local effects of writing assessment can be critically important: use of these tools may mean the difference between college success and failure for many students.

As articulated in the assessment literature, the definition of validity has shifted over time (Kane, "Validation"). Currently, the measurement community has forwarded a definition of validity based on interpretation and use of test scores (Kane, "Validating"). In their acknowledgment that use of assessment information is at the center of validity, measurement researchers have advanced four sources of validity evidence in the call for validation studies: scoring (estab-lishment of test taker's performance through a scoring framework); generaliza-tion (evaluation of the different conditions that impact student performance), extrapolation (inference linking the test to a range of performances associated with the concept under investigation), and consequence (anticipation of intended outcomes, adverse impact, and systemic effects). Likewise, the composition community has begun to employ validation studies in a variety of contexts, including studies of how raters' own language backgrounds and academic experiences influence decision making (Wiseman), design of integrated writing tasks for English language learners (Knoch and Sitajalabhorn), relationships between directed self-placement methods and curricular success (Gere et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.