Skip Navigation
small NCES header image

Table of Contents  |  Search Technical Documentation  |  References

NAEP Assessment Weighting Procedures → Weighting Procedures for the 2005 Assessment → Quality Control for the 2005 Assessment Weighting → Main Quality Control Findings of Interest: Nonresponse Bias Analyses for the 2005 Assessment

Main Quality Control Findings of Interest: Nonresponse Bias Analyses for the 2005 Assessment

       

Private Schools at Grades 4 and 8 (67kb)

Private Schools at Grade 12 (90kb)

Students at Grades 8 and 12 (76kb)

School-level nonresponse bias analyses were conducted on private schools at all three grades (4, 8, and 12) since the school-level participation rate of each of these student groups fell below the 85 percent participation threshold. The analyses involved three components. The first analysis looked for potential bias introduced through the school nonresponse. The second analysis examined the remaining potential for nonresponse bias after accounting for mitigating effects of school substitution. The third analysis examined the remaining potential for nonresponse bias after accounting for mitigating effects of both school substitution and school-level nonresponse weight adjustments.

The analyses suggest that although school nonresponse for private schools was quite high at all three grades, there is little evidence of bias.

Student-level nonresponse bias analyses were conducted on grade 8 students in the Cleveland and New York City TUDA districts and students in grade 12 public schools since the student-level participation rates of these student groups fell below the 85 percent participation threshold. The analyses involved two components. The first analysis looked for potential bias introduced through student nonresponse. The second analysis examined the remaining potential for nonresponse bias after accounting for mitigating effects of the student-level nonresponse weighting adjustment.

Although a number of differences were found to be statistically significant, the large sample sizes resulted in large statistical power, and hence small differences were detected as being statistically significant. Especially after the student nonresponse adjustments were applied, the differences between the respondents and the full samples were very small, even for grade 12 public schools where the student response rate was substantially below 85 percent.


Last updated 07 July 2009 (GF)

Printer-friendly Version


Would you like to help us improve our products and website by taking a short survey?

YES, I would like to take the survey

or

No Thanks

The survey consists of a few short questions and takes less than one minute to complete.
National Center for Education Statistics - http://nces.ed.gov
U.S. Department of Education