(Nicole Strangman and Tracey Hall National Center o

1
Examining Background Variables of Students with
Disabilities that Affect Reading

• Jamal Abedi, CRESST/University of California,
  Davis
• Seth Leon and Jenny Kao, CRESST/University of
  California, Los Angeles

2
Examining Background Variables of Students with
Disabilities that Affect Reading

• The reauthorization of the Individuals with
  Disabilities Education Act (IDEA) demands equity
  and accountability in education for the
  approximately 6.5 million children and youth with
  disabilities in the United States (U.S.
  Department of Education, 2004)
• In the 2003-2004 school year, almost half of all
  students with disabilities were in regular
  classrooms for 80 percent or more of the school
  day (U.S. Department of Education, 2005)
• Including these students into assessments
  requires first having a valid and reliable
  measure of their knowledge and skills

3
Focus on Reading

• Among the content areas, reading is of greater
  importance.
• Literature shows lower performance on academic
  achievement tests for students with disabilities
  (SD).
• Students with disabilities usually fail at higher
  rates than other students, even on tests of basic
  skills for graduation (Heubert, 2000)
• These students perform significantly lower than
  students with no apparent disabilities (Abedi,
  Leon, Mirocha, 2003 Thurlow, et. al., 2000
  Tindal, et. al, 2000 Ysseldyke et al., 1998)
• In 1998, data from fourteen states revealed that
  students with disabilities consistently failed
  state graduation tests at rates 35 to 40
  percentage points higher than those for non-SD
  (Ysseldyke et al., 1998)

4
Factors contributing to performance-gap between
SDs and non-SDs

• There are many factors that contribute to
  performance-gap between SDs and non-SDs
• This paper focuses on the background factors that
  affect reading
• This paper uses the term background in a
  broader sense to include both students
  background variables and background variables
  related to the assessment
• We focused on students background knowledge and
  Test format

5
Differential pattern of performance across
SD/non-SD categories

• We examined the possibility of differential
  impact of these variables on the reading
  performance of students with disabilities
• The results of this study along with the findings
  of other studies showed differential impact
  across SD/non-SD categories
• There are many different forms of detecting item
  bias (Matlock-Hetzel, 1997 ONeal, 1991)

6
Data Source

• The data were obtained for the 1997-1998 academic
  year from a state with a large number of ELLs and
  students with disabilities
• Included item-level information on grades 3 and 9
  students responses to Stanford 9 Reading
  Comprehension (RC) and Word Analysis (WA)
• Multiple-choice items in two reading subscales
  (RC WA) were selected for this study. Analyses
  were conducted for grade 3 and grade 9 students
• Students in higher grades may have more
  difficulty with reading

7
Statistical Design

• Logistic regression approach was used to examine
  the possibility of differential distractor
  functioning (DDF) across the SD and non-SD
  categories
• Responses were grouped into two categories (1)
  students who selected the most common distractor,
  and (2) students who selected one of the two less
  common distractors
• This indicator of distractor selection was used
  as the criterion variable
• SD status, ELL status and the total standardized
  Z-score on the subscale being analyzed were used
  as predictors
• Odds Ratios (OR) of selecting the less common
  distractors as compared with the most common
  distractor were computed

8
ResultsReading Comprehension, Grade 3

• The large majority of the items exhibit DDF based
  on reading comprehension ability
• Of the 54 items, 46 showed significant DDF on the
  ability measure
• A significant odds ratio greater than one
  indicates that SDs were less likely to choose the
  most commonly chosen distractor

9
Results Reading Comprehension, Grade 9

• The large majority of the items exhibit DDF based
  on reading comprehension ability
• Of the 54 items, 46 showed significant DDF
• Significant DDF for SDs suggests that SDs are
  generally less likely to choose the most commonly
  chosen distractor when compared to non-SDs
• Results show that a substantial number of items
  exhibit DDF for SDs even when controlling for
  reading comprehension ability
• It also appears both for SDs that there are more
  items in grade 9 that exhibit DDF as compared to
  grade 3 on the reading comprehension subscale

10
Word Analysis, Grade 3.

• Just over one half (16 of 30) of the items
  exhibit DDF based on the word analyses ability.
  This was a considerably smaller proportion as
  compared the RC subscale in which 46 of 54 items
  showed significant ability DDF
• A significant DDF for SDs suggest that they were
  generally less likely to choose the most commonly
  chosen distractor when compared to non-SDs

11
Word Analysis, Grade 9

• Of the 30 items, 22 exhibited DDF based on
  reading comprehension ability
• The odd rations for SDs were most often greater
  then one
• The grade 9 results show that a substantial
  number of items exhibit DDF for SDs even while
  controlling for reading comprehension ability
• Similar to the RC results, SDs had more items in
  grade 9 that exhibited DDF as compared to grade 3
  on the WA subscale

12
Conclusion

• Findings of this study provide evidence that in
  addition to test content other factors may
  contribute to the performance-gap between SDs and
  non-SDs
• Controlling for these factors that are not
  related to content being assessed may help test
  developers provide more accessible and more valid
  assessments for students with disabilities

13
Limitations

• This study did not differentiate between
  different categories of disabilities
• Student performance across different categories
  of disabilities may be quite different and these
  factors may affect their performance quite
  differently
• We could not include other test characteristics
  (such as type of items, fatigue and frustration
  factors, graphic layout, and size and type of
  font) that may be considered as extraneous or
  nuisance variables and may impact performance
  students with disabilities

14
Background knowledge(Nicole Strangman and Tracey
Hall National Center on Accessing the General
Curriculum)

• Students who lack sufficient background knowledge
  may struggle to access, participate, and progress
  throughout the general curriculum
• The terms background knowledge and prior
  knowledge are generally used interchangeably
• Prior knowledge and background knowledge include
  dimensions such as conceptual knowledge,
  metacognitive knowledge, subject matter
  knowledge, strategy knowledge, personal
  knowledge, and self-knowledge

15
Background knowledge (continues)

• There is a well established correlation between
  prior knowledge and reading comprehension
  (Langer, 1984 Long, Winograd, Bridget, 1989
  Stevens, 1980)
• Irrespective of students reading ability, high
  prior knowledge of a subject area or key
  vocabulary for a text often means higher scores
  on reading comprehension measures (Langer, 1984
  Long et al., 1989 Stevens, 1980)
• After controlling for reading ability in the
  sample, a significant effect of prior knowledge
  building on reading comprehension was observed
  (Stevens, 1982 Dole et al., 1991 Graves et
  al., 1983)

16
Student characteristics

• Weisberg (1988) claims that students with
  disabilities, as a group, demonstrate a
  considerable over reliance on prior knowledge
  when text material is inconsistent with their
  preconceptions
• This raises another issue, which is whether a
  students educational group or disability status
  influences the effectiveness of prior knowledge
  activation strategies
• A few of these studies with students with
  learning disabilities revealed differences in
  responsiveness to prior knowledge activation
  across educational groups (Carr et al., 1996
  Langer, 1984 Pflaum et al., 1982 Carr et al.,
  1996 Croll, Idol-Maestas, Heal, Pearson, 1986
  Pflaum, Pascarella, Auer, Augustyn, Boswick,
  1982 Walraven et al., 1993)

17
Overall Summary/Conclusion

• Many factors contribute to the performance-gap
  between SDs and non-SDs
• Among these factors, background variables may
  have substantial impact on SDs performance
• To have a more valid and reliable test and at the
  same time more accessible assessment for SDs,
  these factors and their level of contribution
  must be identified and controlled

18
For more information contact

• Jamal Abedi
• jabedi_at_ucdavis.edu
• [phone]