Exploring possible sociocultural bias on the SCAN-C.

Area of Science:

• Speech-language pathology
• Child development
• Sociolinguistics

Background:

• The Structure of Communication Activity - Child (SCAN-C) is a tool used to assess communication skills in children.
• Sociocultural factors, including ethnicity and socioeconomic status (SES), may influence test performance.
• Previous research has not fully explored potential biases in the SCAN-C for diverse populations.

Purpose of the Study:

• To investigate potential sociocultural bias in the SCAN-C.
• To examine differences in SCAN-C performance between Anglo American and Latino American children.
• To assess the impact of dialectal variations on SCAN-C scores for Latino American children.

Main Methods:

• A sample of 20 Anglo American and 20 Latino American 8-year-olds from varying SES backgrounds participated.
• Univariate and repeated measures ANOVAs were used to compare group performance.
• Classification analyses were conducted, including analyses with dialectal rescoring for Latino American participants.

Main Results:

• Initial analyses revealed no significant differences between ethnic groups when clustered by ethnicity and SES.
• Classification analyses showed 10% more Latino American children scored in the borderline-to-disordered range compared to Anglo American children.
• These classification differences were most pronounced on the Filtered Words subtest (25% difference).
• Dialectal rescoring of Latino American participants' scores resulted in classification distributions more closely matching those of Anglo American children.

Conclusions:

• The SCAN-C may exhibit sociocultural bias, potentially misclassifying Latino American children.
• Caution is advised when interpreting SCAN-C results for Latino American children due to potential dialectal influences.
• Applying dialectal scoring adjustments is recommended for Latino American children scoring in the borderline-to-disordered range to improve diagnostic accuracy.