You are a senior MEAL specialist with expertise in education measurement and learning assessment design. Your task is to design a Learning Assessment Tool for "education program name" (from the material, if given).
Assessment Context:
- Program name: "program name" (from the material, if given)
- Assessment purpose: "purpose" (from the material, if given)
- Target learners: "learners" (from the material, if given)
- Language of instruction: "language" (from the material, if given)
- Grade or competency levels: "levels" (from the material, if given)
- Geographic scope: "location" (from the material, if given)
- Donor: "donor" (from the material, if given)
- Comparable assessments: "comparisons" (from the material, if given)
Deliverables:
1. Assessment Framework
- Constructs to measure (reading, mathematics, or both)
- Alignment to national curriculum learning objectives for the target grade levels
- Relationship to international frameworks: EGRA (RTI International) for literacy, EGMA (RTI International) for numeracy
- Assessment type: individual oral assessment, group written assessment, or hybrid
- Criterion-referenced benchmarks: what constitutes "meeting grade-level expectations"
2. Literacy Assessment (EGRA-Aligned) Design subtasks with specifications:
| Subtask | Construct Measured | Number of Items | Time Limit | Scoring Method | Auto-Stop Rule |
|---|
Include at minimum: Letter/Sound Recognition (100 items, 60s, CLPM); Familiar Word Reading (50 items, 60s, CWPM); Non-Word Reading (50 items, 60s); Oral Reading Fluency (60-80 word passage, 60s, CWPM plus comprehension); Reading Comprehension (5 literal/inferential questions, untimed); Listening Comprehension (read-aloud passage with 5 questions, for pre-readers, untimed).
For each subtask, specify the auto-stop rule, assessor script for practice items, and scoring conventions (self-corrections = correct, skips = errors).
3. Numeracy Assessment (EGMA-Aligned) Design subtasks with specifications:
| Subtask | Construct Measured | Number of Items | Time Limit | Scoring Method | Auto-Stop Rule |
|---|
Include at minimum: Number Identification (20 items, 60s), Quantity Discrimination (10 pairs, untimed), Missing Number Patterns (10 sequences, untimed), Addition Level 1 (20 single-digit, 60s), Subtraction Level 1 (20 single-digit, 60s), Addition Level 2 (5 double-digit, untimed), Subtraction Level 2 (5 double-digit, untimed), and Word Problems (6 items, increasing complexity, untimed).
4. Benchmarks and Scoring Rubrics Define performance levels for each subtask:
| Subtask | Non-Reader/Non-Numerate | Emergent | Proficient | Advanced |
|---|
Example: Oral Reading Fluency: 0 CWPM = Non-Reader, 1-20 CWPM = Emergent, 21-45 CWPM = Proficient, 46+ CWPM = Advanced.
Provide a composite score methodology:
- How to combine subtask scores into an overall literacy and numeracy proficiency level
- Weighting of subtasks if applicable
- Cut-scores for "meeting minimum proficiency" aligned to SDG 4.1.1 reporting requirements
5. Administration Protocol Cover: assessor qualifications and selection; training plan (minimum 5 days, including practice assessments and reliability testing targeting >0.85 agreement via dual-scoring of 10% of assessments); consent scripts for parents and children; assessment environment requirements; estimated 15-20 minutes per child; accommodations for learners with disabilities; and protocol for distressed children.
6. Sampling and Logistics If sample-based: sample size for detecting a meaningful effect (e.g., 0.25 SD gain). Cover school/learner selection, daily assessment targets per assessor, materials list (stimulus sheets, scoring booklets, timers, consent forms), and digital data entry setup (e.g., Tangerine, KoboToolbox, or SurveyCTO with timed modules).
7. Analysis Plan Specify: descriptive statistics per subtask (mean, SD, distribution); percentage meeting each benchmark disaggregated by sex, age, disability, and location; gain scores from baseline to endline (effect size or paired t-test); zero-score analysis; subtask correlation for coherence validation; and equity analysis of learning gaps by gender, disability, and socioeconomic proxy.
Where the material is silent, write "Not in the material" rather than guessing.
Source material: Attached, or pasted below. Base everything above on it and do not invent details that are not in it.