How Will NCLB Affect Schools and Students?

1
How Will NCLB Affect Schools and Students?

• Rebecca Zwick
• University of California,
• Santa Barbara
• AAAS Meeting, Seattle, February 15, 2004

2
NCLB assessment provisions

• Grades 3-8 tested annually in math and reading by
  2005-6
• All students 100 proficient by 2013-14
• Schools must show adequate yearly progress for
  each subject area and group (Results are
  percentage declared proficient)
• At least 95 of each group to be tested

3
Results to be reported separately for groups
defined by

• Gender
• Racial/ethnic category
• English proficiency status
• Disability status
• Migrant status
• Socioeconomic status
• (US DOE, 2003)

4
NCLB Sanctions for failure to meetAYP objectives
(Title I-eligible schools)

• 2 years in a row in need of improvement
  Students must be offered the chance to transfer
  to another public school.
• 3 years in a rowstudents must also be offered
  supplemental services, e.g., tutoring.
• Continued failure School is subject to
  corrective measures and possibly takeover.

5
Benefits of NCLB

• Reflects national desire to improve education and
  reduce achievement gap
• Forces schools to focus on needs of groups that
  are sometimes neglected

6
Unintended effects

• Focus on assessment can narrow instruction
• Unrealistic goals and fear of sanctions can
  encourage reduction in state standards
• Morale may decay because of perceived inequities
  in the sanctioning process
• Sanctions can lead to double jeopardy
• Legislative action opposing NCLB in Virginia,
  Ohio, Utah, and North Dakota

7
Focus on assessment can narrow instruction

• Education Weeks National Survey of Public School
  Teachers (see Olson, Jan. 11, 2001)
• Conducted in 2000, pre-NCLB (n 1,019)
• 66 State testing leads to focus on tested
  material to the detriment of other areas
• 67 say there is far too much or somewhat too
  much focus on tests

8

• Unrealistic goals and fear of sanctions can
  encourage reduction in state standards

9
Percentage at or above NAEP Achievement Levels
Grade 4 Reading
1992 1994 1998 2000 2002 2003
Basic 62 60 60 59 64 63
Profi-cient 29 30 29 29 31 31
10
Notes to table

• 6 to 8 of each cohort reached Advanced
• SEs for tabled s range from .3 to 1.4
• All results are for accommodations allowed
  conditions except 1992 and 1994.

11
Fear of sanctions provides incentives for lower
standards of proficiency

• How is NCLB proficient defined?
• Louisiana set equal to states basic
• Colorado set equal to states partially
  proficient
• Connecticut Set lower than states own goals
• (Source, Hoff, D., Education Week, 10/9/02)

12

• Morale may decay because of perceived inequities
  in the sanctioning process.

13
Criteria for sanctions vary across states

• Tests selected or developed
• Definition of proficient
• Minimum sample size required
• Adequate yearly progress (AYP) definition and
  starting points
• Enforcement (waiver policies)

14
Examples of variability across states

• According to 8/6/03 Education Week
• Minimum sample size for accountability ranges
  from 5 to 50 (more later)
• AYP 21 accelerating 25 approx. linear
• For 2003, some states provided waivers for
  schools with participation rates under 95

15
Criteria for sanctions may be unreliable

• Harsh sanctions can be triggered by results on a
  single fallible indicator
• Imprecision due to small samplesas small as 5
  (29 states include CI or SE criteria)
• Inaccuracy due to student mobility
• No credit for changes that dont affect the
  percentage proficient (Linn et al., 2002)

16
Sanctions can lead to double jeopardy

• Low socioeconomic-status schools are more likely
  to be judged in need of improvement.
• Sanctions could then further drain school
  resources.
• Great care must be taken to avoid unjustified
  imposition of sanctions.

17
Recommendations What can statistics and
measurement professionals offer?

• Encourage inclusion of information about sampling
  and measurement error
• Study the probability of misclassifying schools
  under realistic assumptions.
• Encourage methods that reduce the likelihood of
  misclassification.
• Educate school personnel and policymakers about
  educational measurement statistics

18
Bibliography

• Goertz, M. E. (2001, Sept.) The federal role in
  defining adequate yearly progress. Consortium
  for Policy Research in Education Report. cpre.
  org.
• Hoff, D. J. (2002, Oct. 9). States revised
  meaning of proficient. Education Week, pp. 1,
  24-25.
• Linn, R. L., Baker, E. L., Betebenner, D. W.
  (2002). Accountability systems Implications of
  requirements of the No Child Left Behind Act of
  2001. Educational Researcher, 31, 3-16.
• Olson, L. (2001, Jan.11). Overboard on testing?
  Education Week Quality Counts 2001, pp. 23-30.

19
References (continued)

• Olson, L. (2003, Aug. 6). Approved is relative
  term for Ed. Dept. Education Week, pp. 1, 34-36.
• Olson, L. (2004, Jan. 7). Data doubts plague
  states, federal law. Education Week, pp. 1, 26.
• Rebora, A. (updated 2003, Aug. 6). No Child
  Left Behind. Summary Education Week document.
  edweek.com.
• Rogosa, D. (1999). Accuracy of individual scores
  expressed in percentile ranks classical test
  theory calculations. (CSE Technical Report 509.)
  cse.ucla.edu.

20
References (continued)

• Thum, Y. M. (2003). No Child Left Behind
  Methodological challenges recommendations for
  measuring adequate yearly progress. (CSE
  Technical Report 590.) cse.ucla.edu.
• US Department of Education (2003). No Child Left
  Behind A Parents Guide. nclb.gov/next.
• U.S. Department of Education, National Center for
  Education Statistics, NAEP 2003, 2002, 2000,
  1998, 1994 and 1992 Reading Assessments.