I appreciate the opportunity to be invited to express my
views related to the assessment of English Language Learners. Unquestionably,
the No Child Left Behind
Act has placed a new emphasis on assessment and accountability.
New Jersey now
faces the important task of addressing the Federal mandate to assess all
students including all English Language Learners. When approaching this task,
we must look at the mandate in light of what we have learned about second
language acquisition and ask some important questions:
First, the purpose of assessment:
- Assessment
should provide valuable information, such as:
- the
knowledge and skills the students bring with them (native language);
- what
progress are students making in English language proficiency after each
year of instruction;
- the
academic progress of students;
- the
effectiveness of curriculum and instruction;
- the effectiveness of different program models.
- Assessment
should help us in the decision-making process related to curriculum and
instruction.
What do we know
about the second language acquisition process?
- It is
a process involving stages of acquisition;
- It
begins with the development of interpersonal skills;
- It
takes as long as 5 to 7 years before most students acquire the academic
language required to achieve a level of “proficient” on state mandated tests;
- Many
student characteristics and environmental factors impact on language acquisition.
The New Jersey Core Curriculum Content Standards were
developed to define what an educated student should know and be able to do at
particular benchmarks. In the case of ELLs, content tests are often a measure
of English literacy rather than content.
How can English Language Learners demonstrate what they know
when they might have just arrived in our schools and have no knowledge of the
language used in the assessments?
They can only demonstrate the knowledge and skills acquired
over time in the language of instruction.
What does the
assessment process need to include?
We need to
establish baseline data that includes:
The knowledge students bring –this
requires assessment in the students’ native language whenever possible, and
Their level of English
proficiency at the time of entry.
What do we need to plan for in the assessment design?
We need to:
- Continue
assessing the students’ annual progress in English language proficiency;
- Align
assessment to the NJCCCS in order to determine achievement of content
knowledge and skills;
- Design
assessments aligned to levels of language proficiency.
What are the
options that would provide multiple opportunities to demonstrate progress?
- Native
language assessment;
- Alternative
assessment including portfolio assessment reflecting the same level of
standards as the state tests. These could include samples of student work
over time scored with appropriate rubrics scored by several assessors;
- Language
proficiency tests administered annually;
- Sheltered
English tests in the content areas aligned with levels of language acquisition;
- State
mandated tests with modifications (if appropriate for level of
proficiency)
Use of the
data can be extremely useful if we:
- Examine
student variables such as age, level of L1 proficiency,
- reading
level in L1, previous academic experience, time in the US,
- mobility,
and type of program when analyzing assessment data ;
- Compare
the progress of the same students from year to year.
- Individual
student growth is emphasized.
Unless these
measures are taken, we risk the following taking place:
- Levels
of language acquisition will be ignored and teaching will take place at a
level incomprehensible for the student;
- Emphasis
on teaching to the test;
- Elimination
of native language instruction;
- Tracking
students who are “not successful”;
- Advising
students not to take certain academic or elective courses;
- Higher
drop out rates.
There are no easy solutions to this very complex situation.
It requires serious and careful consideration, if we are to provide equality of
educational opportunity for all our students with their unique characteristics.
Submitted by:
Grisel López-Díaz
Assistant
Professor
New
Jersey City University
January 16, 2004