Academic journal article The Journal of Research in Business Education

Measuring Speech Recognition Proficiency: A Psychometric Analysis of Speed and Accuracy

Academic journal article The Journal of Research in Business Education

Measuring Speech Recognition Proficiency: A Psychometric Analysis of Speed and Accuracy

Article excerpt


This study examined the validity of various measures of speed and accuracy for assessing proficiency in speech recognition. The study specifically compared two different word-count indices for speed and accuracy (the 5-stroke word and the 1.4-syllable standard word) on a timing administered to 114 speech recognition students measured at 1-, 2-, and 3-minute intervals. Speed scores measured by 5-stroke words were consistently 9% gross words a minute higher on 1-, 2-, and 3-minute timings than scores measured by 1.4-syllable standard words on the same timings; this difference was statistically significant. Speed increased with the length of timing; the effect was small but statistically significant. Accuracy scores were similar across assessment methods and length of timing. The findings indicate that the most valid and reliable measure of speech recognition speed was gross words a minute on timings with 1.4 syllabic intensity measured by the standard 1.4 syllable word count. In contrast to keyboarding timings, 1-minute, 2-minute, and 3-minute timings in speech recognition are reliable measures of speed and accuracy.


SPEECH RECOGNITION TECHNOLOGY has evolved into a viable product for all users as the technology has improved and the hardware and software requirements have become less expensive. Because speech recognition improves input speed and productivity, it is currently being taught at all levels of the curriculum, from elementary to college level (Erthal & Rader, 2004; Lear & Zimmerman, 2005). In elementary and secondary schools, speech recognition is usually taught as a unit in another course such as keyboarding or business technology. In community colleges and four-year colleges, it is taught either as a separate course or as a unit in other courses such as computer applications and medical transcription (Erthal & Rader, 2004).

The National Standards for Business Education state that students should "Develop proper input techniques (e.g., keyboarding, 10-key touch pad, scanning, speech recognition . . .)" (NBEA, 2007, p. 90). However, the National Standards do not specify the proficiency level or assessment methods that should be used to measure if and to what degree the standard has been met.

Classroom assessment is normally used to determine if students are meeting performance objectives (Duncan-Evans, 2002). The terminal objective of speech recognition should be a usable skill that the individual can transfer from the classroom to the workplace. Speech recognition should be evaluated according to performance assessment: the requirement that individuals achieve learning outcomes that can be measured (Madaus & O'Dwyer, 1999). Because speech recognition produces measurable outcomes, the product can be evaluated according to traditional keyboarding standards such as speed, accuracy, and production. These criteria are traditionally utilized in both formative and summative assessments. Summative assessment is primarily product oriented (Lambrecht, 2000), particularly in skill courses such as keyboarding. In contrast, formative assessment often includes process-oriented criteria such as the evaluation of keyboarding technique.

Need for the Study

Keyboarding instructors historically have used technique, speed and accuracy based on straight-copy timings, and production work to assess keyboarding proficiency. Technique is generally assessed by observation and checklists in the early stages of keyboarding instruction. After students have learned the keyboard, speed and accuracy are emphasized and measured by straight-copy timings throughout the course, and production work is assessed in the later stages of learning (McLean, 1994; Robinson, 1979; West, 1974, 1983). Keyboarding skill development is based on psychomotor coordination; in contrast, skill development in speech recognition is based on dictation (verbal) fluency (Rader & Kurth, 2007). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.