[Wendy Lapic Hall]

7. Cost-Effectiveness
The placement assessment system provides reasonable value to students and the college at a reasonable cost.
Costs of the assessment process are acceptable, in view of the benefits that result.

There are several factors related to cost-effectiveness of placement assessment including:
  1. Cost of assessment testing
  2. Instructional costs
  3. Opportunity cost of missing out on potential Financial Aid benefits (without testing)

Cost of assessment testing: The two predominant placement/assessment models in the Washington State Community and Technical College system are Compass (ACT) and Accuplacer (College Board). Both tests are administered on a per unit fee basis, with most colleges passing on the cost of assessment and related administration to the students. Both Compass and Accuplacer offer volume discount pricing models. A third option offered through Pearson called MyFoundationsLab offers an integrated testing, diagnostics and instructional model.

Compass offers placement assessment in math, reading and writing, with diagnostic assessment available for math and writing as well. All assessments are based on a per unit cost including the diagnostics, with the exception of the more intensive writing assessment (e-write) which is charged back at a rate of 3.5 units. Additionally, there is a 0.4 unit charge per student for collection of demographic information. If a student takes the standard placement assessment consisting of math, reading, writing and a demographic section, there is a charge for 3.4 units.

Per unit pricing for Compass is as follows:
  • 1 - 4,999 students: $1.66 per unit
  • 5,000 - 14,999: $1.50 per unit
  • 15,000 - 34,999: $1.42 per unit
  • 35,000 - 99,999: $1.34 per unit
  • 100,000 - 174,999: $1.27 per unit
  • 175,000 or more: $1.21 per unit

The maximum savings per student is $.45 per unit ($1.53 or more for the standard battery) at the system vs. institutional level. These fees are paid by students at most/all colleges so cost savings for the colleges would be minimal.

Accuplacer offerings assessment in reading comprehension, sentence skills, arithmetic, elementary algebra and college-level math. Diagnostic assessments are also available in reading comprehension, sentence skills, arithmetic, and elementary algebra. There is an essay evaluation tool as well. Each of the placement assessments is charged back at a rate of one unit per assessment. Diagnostic assessments are charged back at a rate of 2 units per assessment.

Per unit pricing for Accuplacer is as follows:
  • Standard: $2.10 per unit
  • College Board member discount: $1.95 per unit
  • State approved discount (meaning the tools have been approved for an entire state): $1.75/unit

The maximum savings per student is $.35 per unit ($1.05 or more for the standard battery) at the system vs. institutional level. These charges are paid by students at most/all colleges.

The emergent player is Pearson, which offers both a traditional assessment tool (MyReadinessTest) that is comparable to Compass and Accuplacer at a rate of $10 per 10 tests. There is a 20% volume discount offered for systems vs. individual institutions. Pearson also offers an integrated testing and instructional system called MyFoundationsLab. Rather than paying for a stand alone assessment, students pay a fee which gives them access to testing and instructional materials for a year at a time. Currently the access fee is $100, plus the cost of e-books. At one institution in Washington, the cost of the e-books is approximately 40% of the cost of traditional textbooks rendering a significant savings to students (particularly if they need to take multiple developmental courses). As with the other assessment tests, fees are paid by students.

Instructional costs: Developmental instructional costs are determined by the amount of remediation required by students. The amount of remediation can be impacted by testing--cut scores resulting in lower placements often equate to more dollars being spent on remediation--and by the instructional model. In the traditional, sequential instructional model, students place at a particular level and take courses sequentially (sometimes repeating courses) until the desired level is reached. In the newer, modularized instructional model, students take only those pieces of instruction they need (based on diagnostic testing). For example, while one student may need to take the entire level (5 credits) of pre-college algebra, another student may need only a portion (for example, 2 credits). While the cost of integrated, diagnostic testing up front may actually cost slightly more than non-diagnostic placement testing, the potential savings in terms of fewer credits of remediation is potentially great (dependent at least in part, of course, on the instructional model in use).

The potential instructional cost savings through more precise placement of students (resulting in fewer dollars spent on remediation) is substantial, particularly when you consider the large proportion of students enrolling in developmental coursework. According to a recent study published by the Community College Research Center, ("Assessing Developmental Assessment in Community Colleges"), more than half of entering community college student enroll in some kind of remediation, with a great deal more testing into but never taking a developmental course. The study suggests that the most reliable and valid assessments are those that rely on multiple indicators, such as a placement exams, high school transcripts, and assessment of noncognitive abilities. Multiple indices are often not considered due to additional costs up front; however, the study suggests that the potential for increased validity (through adoption of such measures) is just not very well known. Increased validity in placement assessment yields higher rates of student success--an important aspect of cost effectiveness for community colleges.

Math reform currently underway at a handful of community colleges in Washington State has the potential to further impact instructional costs. Utilizing a combination of technology-supported instruction and integrated assessment, the newer models of math education can reduce per student costs in several ways. A recent study by the National Center for Academic Transformation ("Cost-Effective Strategies for Improving Student Performance: Redesign + MyMathLab in Higher Education Mathematics Courses") suggests that cost savings can occur in the form of declining faculty workloads due to logistical and other routine tasks being supported by technology, increasing capacity to serve students due to computerization of some tasks such as quiz grading, and actual dollar cost savings realized through the reduction in the length of instruction and textbook costs. At one institution in Washington State, the combination of math reform (reducing the total number of pre-college math courses from four to three by eliminating redundancy in curriculum) and textbook savings (through conversion to e-books) is $1282 per student for the full sequence (academic year 2010-11 prices).

Opportunity costs related to Financial Aid: Currently students entering community colleges with no GED or high school diploma must receive an "Ability to Benefit" certification in order to receive Financial Aid. To do so, they must pass a federally approved placement test and receive a qualifying score (as determined by the U.S. Department of Education). Although this impacts a relatively small number of students, the opportunity cost (of missed opportunity to receive financial aid) would be great without the availability of standardized testing.

Conclusion: Students typically pay their own testing fees. Testing centers require staffing, regardless of which test or tests are administered there. Research suggests that increasing the number of indicators used to place students, not limiting or reducing options, is the most promising road to boosting student success. Some students need standardized testing to qualify for financial aid, so it can't be eliminated. Diagnostic testing, although it costs more, has shown more precision in terms of accurately placing students. Cut scores vary, in some cases wildly, from institution to institution, causing some confusion in the general public and even within higher education. Although correlation can occur between placement scores and student success, in most cases it's weak or even non-existent. Students can encounter difficulty in moving from one institution to another where different placement tests, or cut scores, are in place. Coordinating placement assessment across the institution has value, although much of the value appears to be in the form of public perception and, potentially, student satisfaction.

More significantly, strategies intended to reduce time spent in developmental coursework and increase student success have the potential to transform our institutional outcomes and tremendously enhance cost effectiveness. Examples of emerging and promising practices include using multiple indicators for placement, taking advantage of diagnostic aspects of assessment and/or integrated testing and instruction, reducing or eliminating redundancy in developmental curriculum, and providing contextualized learning whenever possible.