Examination of an Automated Procedure for Calculating Morphological Complexity
- Resource Type
- Reports - Research
- Authors
- Wood, Carla (ORCID 0000-0002-0551-0638); Garcia-Salas, Miguel (ORCID 0000-0003-2880-0512); Schatschneider, Christopher (ORCID 0000-0002-1700-7685)
- Source
- Grantee Submission. 2023.
- Subject
- Automation
Computer Assisted Testing
Scoring
Computation
Morphology (Languages)
Written Language
Transcripts (Written Records)
Elementary School Students
Grade 5
Coding
Computational Linguistics
Writing Evaluation
Predictive Validity
Efficiency
- Language
- English
Purpose: The aim of this study was to advance the analysis of written language transcripts by validating an automated scoring procedure using an automated open-access tool for calculating morphological complexity (MC) from written transcripts. Method: The MC of words in 146 written responses of students in fifth grade was assessed using two procedures: (1) hand-coding of words containing derivational morphemes by trained scorers and (2) an automated analysis of MC using Morpholex, a newly developed web-based tool. Correlational analysis between the different MC calculations was examined to consider the relation between hand-coded derivational morpheme counts and the automated measures. Additionally, all MC measures were compared to a previously gathered rating of writing quality to consider predictive validity between the automated Morpholex score and teachers' ratings of writing quality. Results: Automated measures of MC had a strong relation (r = 0.63) with hand-coding of the number of words with derivational morphemes. Additionally, the number of derivational and inflectional and derivational morphemes accounted for a significant amount of the variation in teachers' overall ratings of writing quality. Conclusion: Automated scoring of MC has potential utility as a valid alternative to hand-coding language samples, which may be valuable for progress monitoring of growth in complexity across repeated samples and measuring components that influence perceived quality of academic writing. [This is the online version of an article published in "American Journal of Speech Language Pathology."]