Computer Science Instruction and Assessment
What Domain 7 tests
Computer Science Instruction and Assessment tests pedagogical content knowledge: how to sequence CS concepts, surface misconceptions, design inclusive instruction, and gather valid evidence of learning. Expect items on standards, formative and summative assessment, rubrics, differentiation, and equitable participation.
Core rules
- Standards frameworks such as CSTA K–12 Computer Science Standards and the K–12 CS Framework organize concepts (computing systems, networks, data, algorithms, impacts) and practices (fostering an inclusive culture, collaborating, abstracting, testing).
- Constructivism treats learners as builders of mental models; worked examples and Parsons problems reduce cognitive load before open-ended coding.
- PRIMM sequences instruction as Predict, Run, Investigate, Modify, Make, so students read and explain code before writing from scratch.
- Formative assessment is low-stakes and feeds instruction; summative assessment judges achievement at the end of a unit.
- A rubric should describe observable criteria at distinct performance levels. Validity means the assessment measures the intended construct; reliability means results are consistent.
- UDL and differentiation provide multiple means of representation, action, and engagement without lowering the target standard.
Worked example
A teacher wants to know whether students understand loop bounds, not whether they can copy syntax. She writes a selected-response item asking for the number of times a loop body runs when the counter starts at 1 and the condition is counter < 5. Distractor A is 4 (correct), B is 5 (counts the stop value), C is 3 (off by one), D is 0 (assumes a pretest failure). Because B and C map to documented misconceptions, the item yields diagnostic information, not just a score.
Traps
Do not call a quiz “formative” if it is graded heavily and only records a grade; formative use depends on how results feed back into teaching. A rubric with vague adjectives such as “good” is unreliable. Group work without individual accountability can hide whether each student can code. Avoid assessing tool-specific menus when the standard targets a concept.