A Common Test That Does Not Test Proficiency
In a district with seven world language teachers and five sections per level, the midyear common assessment is built from textbook test-bank items: 40 multiple-choice vocabulary questions, 30 fill-in grammar items, and one five-sentence writing prompt. Results show 41% of students below 'proficient.' The pacing guide lists nine units for the year, and teachers are on different units. One teacher uses integrated performance assessments; others rely on the textbook. The administration wants a board presentation comparing schools, and the department must submit a improvement plan in three weeks. Several teachers say the test is the only common measure and resist changing it midyear. The district has also adopted a proficiency-based grading policy, but no one has defined what proficiency looks like in each course. The department chair wants a plan that is feasible, defensible, and does not publicly blame teachers.
1. Analyze whether the assessment supports the claim that 41% of students are below proficient. Identify the validity issues and what evidence you would need before reporting to the board.
Read the worked answer
The claim '41% below proficient' is not interpretable without construct validity evidence. The test is mostly recognition and discrete-point grammar plus a five-sentence prompt; that samples vocabulary recall and form, not proficiency in the three modes. A student can score low because they have not memorized the textbook's word list, not because they cannot communicate. The fill-in items may have more than one correct answer, and the prompt is too short to sample discourse. There is also a mismatch between the pacing guide's nine units and the assessment's blueprint. If teachers taught different units, the common test is not common. Before reporting to the board, the department should audit the test against the standards and the taught curriculum, sample student responses, and compare with a performance assessment in interpretive, interpersonal, and presentational modes. The 41% figure may be a curriculum-alignment problem, an assessment-design problem, or both; it is not yet evidence of student failure.
2. Propose a six-week plan to align curriculum, instruction, and assessment. What will you keep, drop, or compress from the pacing guide, and how will you make the plan feasible for seven teachers?
Read the worked answer
Week 1: audit and agreement. The seven teachers map the nine units to proficiency targets and identify three or four priority standards per level. Drop or compress low-value units, such as a second vocabulary list on the same theme, rather than rush all nine. Week 2: build one common performance assessment, integrated-performance style: interpretive reading or listening, an interpersonal task, and a presentational prompt, with a shared rubric. Weeks 3 and 4: teach with backward design; each teacher pilots one task and brings student work to a 45-minute professional learning session. Use a protocol: what does the work show, and what is the next instructional move. Week 5: common assessment. Week 6: calibrate scoring with anchor papers, analyze by skill, and adjust. Keep the textbook as a resource, not the syllabus. Protect time by replacing weekly quizzes with one unit assessment. Ask the teacher who already uses performance assessments to lead calibration, not to be the lone exception. The plan should be feasible: one common assessment, not five.
3. How will you communicate with the administration, the board, and resistant colleagues? What is your advocacy message, and what evidence will you bring?
Read the worked answer
To the administration, present a clear message: the current test measures textbook recall, not proficiency; the department is moving to a balanced assessment system and will report growth on interpretive, interpersonal, and presentational tasks. Bring evidence: sample test items, student work, and the percentage of items that require production. Do not blame teachers publicly. Ask for two things: time for collaborative scoring and permission to adjust pacing. To colleagues, acknowledge the real fear that changing grading will make students look worse; propose a transition year in which both measures are reported. Use the resistant teacher's concerns as design constraints: if the test is too time-consuming, start with one mode. To the board, avoid jargon; show a 90-second example of what a proficient student can do and what the old test cannot capture. Advocacy is not a speech; it is a sustained case built on student work and a feasible plan.