Measuring Goodness of Story Narratives

Abstract

PURPOSE: The purpose of this article was to evaluate a new measure of story narrative performance: story completeness. It was hypothesized that by combining organizational (story grammar) and completeness measures, story "goodness" could be quantified. METHOD: Discourse samples from 46 typically developing adults were compared with those from 24 adults with acquired brain injuries. Story retellings were elicited and analyzed for episode structure (story grammar). Each story was also evaluated for the presence of 5 key components, yielding the story completeness score. Story goodness was quantified by combining the story grammar and completeness measures using a 2-coordinate grid system. A multivariate analysis of variance was performed as well as correlational analyses between the story grammar and story completeness scores. RESULTS: There were significant group differences on both story grammar and story completeness. Moderate correlations were noted between the 2 measures, suggesting that the indices were not entirely measuring the same abilities. Plotting the 2 sets of scores into quadrants discriminated the comparison group and the group with brain injury into 4 distinct categories of story "goodness." CONCLUSION: The combination of measures provided a more accurate depiction of discourse performance than either measure alone. Results suggest the measure is sensitive, is reliable, and has potential utility for investigating discourse deficits in clinical populations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 54 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.