Elizabeth Franks
Issue: No Child Left Behind
Act mandates the inclusion of English Language Learners in statewide
assessments. How can these students be assessed in a way that ensures the
validity and equitability of the inferences drawn from the assessment?
Ø NCLB also mandates that large-scale assessments be attached to high stakes consequences. This fact has led to the unintended consequence of using test results of questionable validity to make important educational decisions (Messick, 1994). In a district study comparing proficiency levels and rates of passing state assessments, students at Proficiency Levels 1, 2 and 3 were not able to pass the respective state assessments in Language Arts Literacy while 52% of students at Levels 4 and 5 performed at/or above the Proficient level.
|
Proficiency
Level |
Language Arts Literacy |
|
||
NJASK4 |
200 + |
< 200 |
Total |
|
|
1 |
0 |
2 |
2 |
|
|
2 |
0 |
1 |
1 |
|
|
3 |
0 |
1 |
1 |
|
|
4 |
6 |
5 |
11 |
|
|
5 |
2 |
1 |
3 |
|
|
Subtotal |
8 |
10 |
18 |
|
GEPA |
|
|
|
|
|
1 |
0 |
0 |
0 |
|
|
2 |
0 |
7 |
7 |
|
|
3 |
0 |
11 |
11 |
|
|
4 |
1 |
4 |
5 |
|
|
5 |
1 |
0 |
1 |
|
|
Subtotal |
2 |
22 |
24 |
|
HSPA |
|
|
|
|
|
1 |
0 |
0 |
0 |
|
|
2 |
0 |
2 |
2 |
|
|
3 |
0 |
5 |
5 |
|
|
4 |
3 |
5 |
8 |
|
|
5 |
2 |
0 |
2 |
|
Subtotal |
5 |
12 |
17 |
|
|
Total |
15 |
44 |
59 |
|
When students who exited the program were added to the study, 76% of the students were at/above the Proficient level. This mini-study supports the literature that it takes 5 –7 years for students to become academically proficient in a second language.
This table also demonstrates the difference in the number of students at each level. This level varies each year and in each grade. For the NJASK4, the majority of students were levels 4 and 5 while the majority of eighth grade students fell into levels 2 and 3. Therefore, the percentage of students passing the tests varies according to the level of English language proficiency. The tests do not discriminate between the lower levels of English proficiency. Our district hovers near the benchmark number of 20 students at each grade level. Consequently, each student’s score will impact the AYP in the disaggregated subgroup data.
This information supports Olsen’s and Goldstein’s (1997) assertion that these assessments should not be used until validity and reliability issues are addressed and procedures to ensure effective learning opportunities have been established.
Ø Case Study
o Student in Grade 3 assessed in Supera scored in the 99th percentile in Language Arts Literacy. At the time, he was at Proficiency level 2 in English. The following year, he took the ESPA and scored in the Advanced Proficient range in Science and Math; and the Proficient range in Language Arts Literacy. This student’s Spanish score not only demonstrated his proficiency in Language Arts Literacy but also supports the literature that states that proficiency in first language predicts ability in the second language.
o These profiles are repeated over and over again with students who demonstrate strong literacy skills in their first language. By testing students in their native language and accepting those scores as demonstration of proficiency, these students will never be labeled as “deficient” or “remedial”.
o In addition to the native language assessment all English Language Learners will be assessed and monitored in their acquisition of English. Their strengths will be accentuated and their successes will be cumulative.
Ø Recommendations for Accommodations/Alternate Assessments
o Assessment should be related to language of instruction (Abedi, Hofstetter, Lord; p. 11; 2003)
o Accommodations will aid the high intermediate/advanced level students.
§ Glossary and extra time
§ Small group administration
§ Rephrase instructions into “simpler” English
§ SIOP
· Build background
· Develop vocabulary
o Alternate assessment
§ Native language assessment for students in bilingual Spanish classes
§ Simplified, modified English assessments for higher level ELLs and students not being instructed in Spanish bilingual classes.
§ Authentic, standardized performance-based assessment in English
and native language (if possible)
§ Follow students over time to ensure achievement in academic English.
Research has been very clear about the transferability of literacy skills from first language to second language.
“Native language assessment provides a less biased indication of what students know and can do. It also can be used to identify gaps in the native language literacy development. Finally, if students are receiving content instruction in their native language, native language assessment can also serve as a way to measure the continued development of their content knowledge.” (Stansfield, 1996)
August,
Calderon and Carlo (2001) state that “the level of L2 [second language]
competence that a bilingual child attains is partially a function of the type
of competence the child has developed in L1 [native language] at the time when
intensive exposure to L2 begins”. (based on Cummins,
1979, p. 233).
Previous
research on the transfer of skills from Spanish to English indicates that there
is transfer in phonological awareness (Durgunoglu,
Nagy, and Hancin-Bhatt, in press), word reading
(Lambert and Tucker, 1972; Kendall, Lajeunesse, Chmilar, Shapson, and Shapson, 1987), vocabulary knowledge (see Fitzgerald 1995
for a review), and comprehension (Escamilla, 1987; Jimenez, Garcia and Pearson,
1995,1996). Our investigation builds on Cummins'
theoretical framework and previous research on transfer.
Results
from preliminary analyses indicate that Spanish phonemic awareness and Spanish
word reading and fluency are reliable predictors of English performance on
parallel tasks in English at the end of third grade. The effect of Spanish
phonemic awareness on English phonemic awareness emerged for all students.
However, the effect of Spanish word reading on English word reading emerged
only for students who received formal instruction in Spanish reading. These
preliminary findings support the practice of providing literacy instruction in
Spanish to Spanish-speaking English-language learners as a means of helping
them acquire literacy skills in English. Strengthening these students' Spanish literacy
also enables them to use their native language well, enhancing their bilingual
capacity.
Native
language assessment is an option that provides a means of incorporating more
students into assessment programs. It is particularly feasible in schools where
large numbers of students speak the same non-English language. It is also
possible to develop native language versions of a test in several languages.
For instance, the New York State Department of Education offers its high school
graduation test, the Regent's Competency Exam, in 20 languages. Additionally,
the Rhode Island Department of Education currently offers native language
versions of its tests for grades 4, 8, and 10 in four languages which include
Spanish, Portuguese, Laotian, and Cambodian. Students who respond in Spanish
are scored in Spanish. Responses in Portuguese, Laotian, and Cambodian are
translated to English and then scored. These very interesting innovations ensure that more comprehensive data
are collected on student achievement and that schools strive to help all
students attain high content standards”(August, Calder, Carlo; 2001).
Roberts
(1994) found that “how literacy is defined has important implications for
decision making. While it may be possible to limit literacy to a narrow definition
of encoding/decoding skills, this view fails to acknowledge that literacy
confers a powerful tool for participating in and changing society. Further,
limiting oneself to a single definition has the effect of denying the
complexity of the concept and of failing to recognize the conflicts that result
from literacy as it functions at individual, group, and societal levels. Above
all, it is imperative to remember that educational outcomes and the allotment
of educational resources are greatly influenced by outside forces (including
availability of funds, competing needs, community and societal attitudes, even
world events, …).
Even
with the complexity inherent in defining literacy, research evidence is clear
that basic literacy skills developed in L1 transfer to L2. Unfortunately, this
evidence is not always acted upon. Such evidence can and should be used to
promote L1 literacy programs
where possible, and where not possible, practitioners can be reassured that
there is evidence that L2 literacy can be developed through educational
practices that recognize and take advantage of a literate environment. In
addition, practitioners can be of great help to learners who sense the
conflicts implicit in English literacy acquisition. If learners can recognize and
explore the political and sociocultural implications
of literacy with respect to their own lives and communities, they will be
better equipped to make decisions that will meet their own needs and goals.
Literacy can be the process and product of inquiry.
Finally,
while bilingual practitioners may be familiar with this issue, it is often
clouded for others by the intricate interplay of social and political factors.
Practitioners can become advocates and cross-trainers by basing their own
choices and activities on the research that is known and by sharing what is
known with others. Teachers are often not only the most effective advocates for
ESL/BE students, but sometimes they are the only advocates. It is important to be
aware of the complexity of issues surrounding the question of the transfer of
literacy skills and to use that knowledge to promote and facilitate
educational equity for students.” (Roberts, 1994).
August, D., Calderon, M., Carlo, M. (2001). The Transfer of skills
from Spanish to English: A study of young
learners. NABE News (March/April 2001) summarized from a study
conducted jointly in 1998-2001 by the Center for Applied Linguistics,
The Journal of Educational Issues of Language Minority
Students, v13 p. 209-221, Spring 1994.
Stansfield, Charles W. (1996). Content assessment in the native language. Practical Assessment, Research & Evaluation, 5(9).