• The AYP computational formulae have several features designed to ensure that decisions are valid and reliable.
  
  – Adjusted Proficiency Targets
  Each target accounts for possible measurement error by an adjustment called a “ confidence interval ”. Schools and districts are only required to meet targets that are lowered to the bottom of range of statistical confidence. This ensures that there are no false negatives in decisions.
  
  – Safe Harbor Proficiency
  Some schools and districts may not meet proficiency requirements even when combining data across three years. If these schools are meeting requirements for attendance and/or graduation one more look is given to determine if they have at least decreased the number of students who are not proficient by 10%.
  
  – Rolling Averages
  Sometimes current year performance may be atypical. If schools or districts to not meet targets in any of the four components current year data is automatically combined with as many as two prior years to determine if an expanded data set can provide a more accurate and positive picture allowing the school or district to meet requirements.
  

For specific computational details regarding validity and reliability components refer to the following sections in the AYP Calculation: Statewide Proportion Proficiency Targets for Each Grade, Confidence Interval applied to Proficiency Targets, Safe Harbor Calculation, Adjustments to Proficiency Measurement found in the 2007 NCLB Calculations Functional Requirements Specification (July 25, 2007)