Assessing writing for placement into college composition courses has traditionally been limited to one or more of a handful of methods: multiple-choice exams (indirect assessment), timed-essay tests, portfolio assessment (direct measures), and directed self-placement (DSP).¹ In the past, standardized tests such as the ACT, SAT, and TOEFL included only multiple-choice questions, but now the SAT and TOEFL include a timed-essay test as well. Indirect assessment measures correlate poorly with writing ability, whereas direct measures increase these correlations, with assessments of multiple samples of student writing providing a better measure than a single sample, though the training and expertise of the raters can affect outcomes, too.

Huot’s (1994) nationwide survey of writing placement practices of 1,037 public and private institutions indicates that a writing sample is the most widely used placement method (51%), followed by standardized test scores (42%), and a combination of a writing sample and standardized test scores (23%). Crusan (2002) reviewed ESL writing placement practices at 10 large public universities, with a particular focus on ESL writers, and found that 3 (Penn State, Purdue, and Wisconsin) used only indirect measures, 2 (Northwestern and Ohio State) used only direct measures, and the rest (Michigan State and the Universities of Illinois, Iowa, Michigan, and Minnesota) used a combination of direct and indirect measures (pp. 24–25).

The CCCC Position Statement on Writing Assessment (2009) notes the following:

Decision-makers should carefully weigh the educational costs and benefits of timed tests, portfolios, directed self-placement, etc. In the minds of those assessed, each of these methods implicitly establishes its value over that of others, so the first impact is likely to be on what students come to believe about writing.” (p. 4)

While “educational costs” refers to more than financial costs, it is worth noting that multiple choice exams, timed essay tests, and portfolio assessment each cost successively more, according to White (1995) and Peckham (2009), who both estimated the expense ratio of essay to portfolio scoring is about 1 to 5. (Peckham estimated the cost of scoring a single essay sample at $5 while a writing portfolio costs $25).

Other educational costs noted in the CCCC Position Statement on Writing Assessment relate to using indirect measures include the loss of professional development opportunities for the teachers who score student writing samples; the distorted message to students about writing and literacy that indirect measures suggest by their traditional emphasis on form over content; and, in the case of machine scoring of essays used by many testing companies today, a reduction in reader-writer interactions.

The Conference on College Composition and Communication (CCCC; 2009) Committee on Second Language Writing, in its Statement on Second-Language Writing and Writers states:

Decisions regarding the placement of second language writers into first-year writing courses should be based on students’ writing proficiency rather than their race, native-language background, nationality, or immigration status. Nor should the decisions be based solely on the scores from standardized tests of general language proficiency or of spoken language proficiency. Instead, scores from the direct assessment of students’ writing proficiency should be used, and multiple writing samples should be consulted whenever possible. Writing programs should work toward making a wide variety of placement options available—including mainstreaming, basic writing, and second language writing as well as courses that systematically integrate native and nonnative speakers of English, such as cross-cultural composition courses. (p. 3)

Both CCCC Statements on Writing Assessment (2006) and Second-Language Writing and Writers (2009) recommend DSP because, in the assessment group’s opinion, “reflection by the writer on his or her own writing processes and performances holds particular promise as a way of generating knowledge about writing and increasing the ability to write successfully” (p. 2), and in the ESL group’s view, “writing programs should inform students of the advantages and disadvantages of each placement option so that students can make informed decisions, and should make this opportunity available to both international and residential second language students” (p. 3).

Nevertheless, some reservations about DSP are evident in the professional literature. Crusan (2006) describes the arguments for and against DSP that were voiced on her campus. Some questioned whether the cost of tuition might influence students (and their parents) to make unrealistic appraisals of their writing ability. Others argue that because the training international students receive in writing may differ from that which U.S. students receive, the students and their U.S. university teachers might have different understandings of the strategies and qualities that writing well entails. Crusan also cites Zamel (1995) in noting that, in general, international students tend to have a higher regard for authority and their teachers, and thus may not feel it is culturally appropriate to question placement decisions. Indeed, in a 1995 survey of ESL students’ satisfaction with their placement in writing courses, Crusan found that 2% felt they had been placed into course below their ability, but none challenged the placement decision (pp. 212–213). For these reasons, critics of DSP have argued that it may lead all students, but especially international students, to bypass the courses most needed for their academic success (Crusan, 2006, p. 211).

Gere, Aull, Green, and Porter (2010) analyzed a decade of data since DSP was introduced at the University of Michigan in 1999, focusing on different conceptions of test validity to determine if DSP was leading to a better match between students’ writing ability and the writing courses in which they enrolled. They note the inherent difficulty of aligning survey questions and demographic data on the student with course content: “The limited correlation of standardized test scores and GPAs with DSP questions, along with the relatively complex profiles of students with regards to these measures, suggests the need for more nuanced ways of describing student achievement” (p. 171). Additionally, a majority of students who participated in surveys and interviews about their experiences with DSP and first-year composition indicated that they followed the advice of their advisors in deciding which courses to enroll in more than the advice received from DSP. Nevertheless, for those students who enrolled in the lower level writing course as recommended by DSP, more than 71% felt they had made the right choice because the course increased their confidence in producing college-level writing successfully (Gere et al., 2010, p. 169).

Both Wright State University, where Crusan works, and the University of Michigan, where Gere works, eventually shifted toward an online directed self-placement system (ODSP). One advantage to ODSP is the global access students have to the system as well as the advanced notice students and administrators have of placement decisions. Peckham (2009) notes that the savings associated with “not having to adjust class sizes up and down during the first two weeks of classes” in addition to “having the students settled in a class by the time they arrived” on campus were significant (p. 522).

The University of Michigan (Gere et al., 2010) and Louisiana State University (LSU; Peckham, 2009) both use variants of ODSP that ask students to write essays in response to assigned reading(s). As in traditional DSP, students select their writing class based on the course information provided. At the University of Michigan, student essays are submitted online and reviewed by their instructors prior to the beginning of classes to identify strengths and weaknesses. At LSU, students are initially placed by standardized test scores. Then they receive course information and are given the option of challenging the writing placement online by logging into a secure system, reading the essay prompt and several assigned texts, and submitting their written response within a set period of time. The more motivated students, those with self-agency that isn’t captured in a standardized test score, pursue this option. Roughly 10% of the students at LSU challenge their initial placement, and of these, about half are moved up or down in the course sequence after their writing sample is scored by writing instructors in the program (Peckham, 2009).

This review of the professional literature to date on assessing writing for placement into college composition indicates that there is no one “best method” for placing students into first-year writing courses, for each method or even combination of methods has its own advantages and disadvantages for the university, the writing program, and the students they serve.

Note

1. Unlike other assessment procedures, directed self-placement does not ask students to produce writing or answer grammaticality and usage questions. Rather, students are presented with detailed information about available courses, guided in an evaluation of their own background and abilities, and allowed to enroll in the course they feel best meets their needs.

References

Conference on College Composition and Communication. (2006). Writing assessment: A position statement (rev. ed.). Urbana, IL: National Council of Teachers of English. Retrieved from http://www.ncte.org/cccc/resources/positions/writingassessment.

Conference on College Composition and Communication. (2009). Statement on second-language writing and writers (rev. ed.). Urbana, IL: National Council of Teachers of English. Retrieved from http://www.ncte.org/cccc/resources/positions/secondlangwriting.

Crusan, D. (2002). An assessment of ESL writing placement assessment. Assessing Writing, 8, 17–30.

Crusan, D. (2006). The politics of implementing online directed self-placement for second language writers. In P. K. Matsuda, C. Ortmeier-Hooper, & X. You (Eds.), The politics of second language writing: In search of the promised land (pp. 205–217). West Lafayette, IN: Parlor Press.

Gere, A. R., Aull, L., Green, T., & Porter, A. (2010). Assessing the validity of directed self-placement at a large university.Assessing Writing, 15, 154–176. doi:10.1016/j.asw.2010.08.003

Huot, B. (1994). A survey of college and university writing placement practices. Writing Program Administration, 17(3), 49–65.

Peckham, I. (2009). Online placement in first-year writing. College Composition and Communication, 6(3), 517–540.

White, E. M. (1995). An apologia for the timed impromptu essay test. College Composition and Communication, 46(1), 30–45.

Zamel, V. (1995). Strangers in academia: The experiences of faculty and ESL students across the curriculum. Reprinted in V. Zamel & R. Spack (Eds.), Negotiating academic literacies: Teaching and learning across languages and cultures (pp. 249-264). Mahwah, N.J.: Lawrence Erlbaum.