-
ELL Students Transition: Once ELL students become proficient in English, they are moved out of the ELL category. This causes the ELL category to perpetually remain low-performing as new arrivals with low proficiency are constantly added.
- Impact: The performance metrics for ELL students cannot improve significantly under this system.
-
Blame on Students, Teachers, and Parents: Often, ELL students, their teachers, and parents are unfairly blamed for a lack of progress.
- Reason: These stakeholders are not provided with enough appropriate instruction and accommodations.
-
Assessment Issues for ELL: Tests constructed and field-tested for native English speakers are often not relevant for ELL students.
- Reliability Coefficients: Tests should have high reliability (0.8 or above).
- ELL students: Tests specifically constructed for them have reliability above 0.8.
- Non-ELL Tests: Often have reliabilities as low as 0.5 for ELL students, indicating 50% measurement error.
-
Consequences of Low Reliability Tests:
- AYP Reporting: Tests with low reliability are still used for accountability purposes, classification, placement, and graduation decisions.
-
Psychometric Studies:
- Factor Analysis: Non-ELL tests showed all items loading on the first factor, indicating they measure the same construct.
- ELL Tests: Only 50% of variance measured by test items are due to content knowledge.
- Other factors include language complexity and cultural issues.
-
Language Complexity: ELL students' low performance may be due to difficulties understanding the linguistic structure of the questions rather than a lack of content knowledge.
- Example: A paper titled "Impact of Language on Assessment" was retitled to "Would you explain the question to me" reflecting the difficulties ELL students face.
-
Construct Relevance: Ideally, tests should measure the intended construct (e.g., Math) without being influenced by irrelevant factors like language complexity.