Prove the Quality of Your Assessment
Third Party Assurance
ASI Certification helps ensure that the assessments you use meet recognized standards for reliability, validity, fairness, and quality.
In addition to certification, ASI helps organizations identify the strengths and weaknesses of their assessments and provides specific, actionable recommendations for improvement.
Products that meet ASI standards earn the designation of being “ASI Certified” and are authorized to display the applicable ASI Certification Mark, demonstrating their commitment to achieving national standards
Assessment Certification Services
Theoretical Basis
The information provided by the assessment author should enable the prospective user to judge whether the test or assessment is suitable for the required purpose. There must thus be a clear description of the construct the assessment intends to measure as well as a detailed justification of the choice of the assessment content and the way the constructs are to be measured.
Application Design
In general, meaningful interpretation of an assessment score demands that the assessment must be administered and scored in such a way that no unintended factors can exert influence on the calculation of the score. The administration and the instruction must be standardized in such a way that the influence of varying instructions or different administrators (for paper tests) or differing test situations or decision rules (for an adaptive test) is eliminated, or in any case kept as limited as possible. The score must also be as objective as possible.
Data Norms
Sample sizes and types that are used to provide population norms are examined and compared to ASI standards. These recommendations are either based on the calculation of standard errors in parameters such as the mean and the median, or on experiential data with respect to the stability of scale values. The requirement with respect to sample size naturally applies to each norm group for which norming takes place.
Data Reliability
- Stability - We measure your tests or assessment's ability to yield the same measurement over a period of time.
- Chronbach's Alpha - Quantifies the internal consistency of the response data for assessment questions.
Validity
- Construct Validity - is the degree to which a test measures what it claims to measure. Constructs are used to conceptualize latent variables. Our data measures correlations to the latent variable and to each other.
- Factor Analysis - is a method used to describe the variability observed. The method searches for joint variances with the goal of reducing redundant or ineffective variables.
- Criterion Validity - is the mathematical relationship between measurement critenon and a concurrent or predictive outcome.
- Structural Equation Modeling - is the use of mathematical models to establish the relationship between observable and latent or dependent variables.
Disparate Impact
- The 80% Rule: A measurement established to by the Department of Labor which states that the selection rate of the protected group should be at least 80% of the selection rate of the non-protected group.
- T-Test: is used to determine whether there is a significant difference in the scores of a protected group as compared to the dominant group.
- ANOVA: A statistical method used to determine if a significant difference exists between the mean scores of the protected class and the non-protected class groups results.