Helwig, R., Stieber, S., Tindal, G., Hollenbeck, K., Heath, B., & Almond, P. (1999). A comparison of factor analyses of handwritten and word-processed writing of middle school students . University of Oregon Research Consultation, and Teaching Program.

Not available online

This study compared the assessment scores in handwritten-response format and transposed typed format, as well as student-composed word-processed format, in order to determine construct validity. Based on technology available, about 20% of the students had access to a spell-checking feature of the software used.

A total of 117 grade 8 students participated, from seven language arts classrooms in three middle schools involving two school districts in western Oregon (U.S.). Ten students (about 8%) were receiving special education instruction, and five of these received assistance in language arts. Additional demographic data were reported for students, and amount of teaching experience of their teachers was also documented.

The Oregon state writing assessment comprised the handwritten-response format condition, and similar items were then administered and required use of a computer keyboard as the response format. Writing scores were based on six traits: conventions, ideas and content, organization, sentence fluency, voice, and word choice. For further comparison purposes, teachers were asked to rate students on their writing proficiency on a 5-point scale. Students reported their amount of computer use.

Students' word-processed composition scores were consistently higher than handwritten composition scores across all six tested traits. Paired t-test results indicated that 3 of the 6 traits reached a level of experimental significance: ideas and content, organization, and voice. Correlations among the six test traits within each test condition were stronger than between test conditions, that is, between each trait in handwritten and word-processed modes. Factor analysis yielded two factors, which were equivalent to the two separate response modes. Separate factor analyses completed on subsets of the participant population (but not students with disabilities vs. students without disabilities) replicated the same results. Limitations of the study, as well as alternative explanations of the findings, were reported.