What is an evaluation strategy?

It’s your plan for how you’ll fairly assess whether students have learned what you promised them you’d give them the opportunity to learn(!).  

Don’t think of your evaluation in terms of a “test”, think of it as an assessment system, with all the meanings we discussed earlier for a system:  formal structure, multiple parts, interactions among parts, with an eye toward incremental improvement of the system.

How does an evaluation strategy contrast with the traditional approach?

In the “old days”, the assessment system was simple:  essay midterm and final exams.  Never mind that grading was often subjective, with no rubric to specify how many points a given answer would receive; never mind that the tests only covered a few topics out of an entire course, or that writing skill was paramount, or that these tests only captured a student’s skill at producing a first draft, and never mind that the entire semester boiled down to just two opportunities to show what you’d learned.

Contrast that with a class containing 10 weekly homework assignments, 10 weekly quizzes, a term paper, four tests (each containing both objective and essay questions), and a short group project. 

How is this better than the traditional approach?

Because students are being  evaluated more fairly in terms of  both quantity and quality. 

Quantity: While quantity is no guarantee of quality, students in this class are being evaluated 26 times during the semester instead of two times (and before you howl that you’d never be able to grade so many assignments, rest assured that we’ll get to that!).  You're also getting a better understanding of students' performance across time because they're assessed incrementally throughout the course. Since test reliability tend to increase as the number of measurements increases, you're more likely to gain a true picture of a student's abilities and less likely to make an assessment error.

Quality: You're improving the quality of assessment because tests are criterion-referenced (i.e., like the actual instruction, they're written according to the instructional objectives) and because students are being  assessed on a wider variety of knowledge and skills.