A Two-Stage Approach for Generating Unbiased Estimates of Text Complexity

Many existing approaches for measuring text complexity tend to overestimate the complexity levels of informational texts while simultaneously underestimating the complexity levels of literary texts. We present a two-stage estimation technique that successfully addresses this problem. At Stage 1, each text is classified into one or another of three possible genres: informational, literary or mixed. Next, at Stage 2, a complexity score is generated for each text by applying one or another of three possible prediction models: one optimized for application to informational texts, one optimized for application to literary texts, and one optimized for application to mixed texts. Each model combines lexical, syntactic and discourse features, as appropriate, to best replicate human complexity judgments. We demonstrate that resulting text complexity predictions are both unbiased, and highly correlated with classifications provided by experienced educators.