The normative sample for the LDDI included 2,152 students with learning disabilities, living in 43 states and the District of Columbia. The normative sample is representative of the population of students who have learning disabilities in the United States as a whole with regard to gender, race, ethnicity, urban/rural residence, family income, educational attainment of parents, and geographic distribution. The sample characteristics were stratified by age and keyed to the demographic characteristics reported in the 1996 Statistical Abstract of the United States.
Scores for the LDDI are reported in terms of stanines (i.e. standard scores with a mean of 5 and a standard deviation of 1.96) and percentiles. The test uses stanines to: identify the likelihood of intrinsic processing disorders in the six areas assessed by the LDDI; and conduct a profile analysis to determine the extent to which a student's LDDI profile reflects that which is associated with learning disabilities.
For all scales the Internal consistency reliability coefficients exceed .90. In addition, evidence for stability and interscorer reliability is also provided and coefficients are in the .80s and .90s. Overall, the LDDI can be used with confidence to yield consistent results.
Several studies were conducted on the validity of the LDDI to ensure that the scores have content-description, criterion-prediction, and construct-identification validity. Extensive item selection and differentiation examinations were involved in these studies, and included confirmatory factor analysis; as well as studies that examined the LDDI's relationship to age, academic achievement, group differentiation, gender, and ethnicity - all of which support the validity of the LDDI scores. Factor analysis research also validated the LDDI's factor structure. These studies all provide evidence that the LDDI yields valid results that can be used with confidence to identify the presence or absence of learning disabilities in children and adolescents.
Controlling for Test Bias
The LDDI minimizes the effects of bias. Many steps were taken to detect and eliminate sources of cultural, gender, and racial bias. The first step that was taken to control the effects of bias was the inclusion of minority groups in the normative sample. Second, the examination of reliability and validity information was presented for the different racial, ethnic, and gender groups. A particularly powerful element of content-description validity is the demonstration of excellent internal consistency reliability for the different racial, ethnic, and gender groups. Finally, the use of differential item functioning analysis was used to reduce item bias during item selection. Delta score values were used to remove items that appeared to be biased against targeted groups.
Back to the top