A new review of past studies looked at how well language sample analysis (LSA) works to identify developmental language disorder (DLD) in bilingual children. The researchers combined the results from many different studies to get a clearer picture of LSA's effectiveness.
The main finding was that LSA correctly identified children with DLD about 70% of the time. It was also good at ruling out DLD in children who did not have it, with an accuracy of about 83%. However, the overall diagnostic accuracy of LSA varied from poor to good, depending on the specific measures used.
The review also found that the language used to get a child to talk (elicitation language) did not change how well LSA worked. The most accurate results came from using a combination of different LSA measures together, rather than relying on just one.
This research suggests that LSA is a useful tool but should not be used alone. It is best as part of a larger assessment that includes other methods to get a full understanding of a child's language abilities.