Understanding sources of user variability in computer-based data entry performance

Sara J. Czaja, Joseph Sharit, Sankaran Nair, Mark Rubert

Research output: Contribution to journalArticlepeer-review

28 Scopus citations


The pervasive use of computers in work settings implies that an increased number of workers, with varying levels of skills and abilities, will be performing computer-based tasks. This study investigated the impact of age, cognitive abilities, and computer experience on the performance of a real world data entry task. One hundred and ten subjects, ranging in age from 20-75 years, performed the task for nine hours following task training. The results indicated that abilities such as visuo-spatial skills, motor skills and processing speed had a significant impact on performance as did age and prior computer experience. With respect to age, the older participants completed less work than the younger and middle-aged subjects. Age differences in psychomotor skills and processing speed appeared to be important factors underlying age effects. In fact, the data indicated that after controlling for differences in these abilities age was no longer a significant predictor of work output. Further, after controlling for differences in work output the older people made fewer errors than the younger people. Overall the data suggest that older people will be at a disadvantage in the performance of computer-based data entry work to the extent to which speed of responding is emphasized. However, if speed of responding is not a critical element of performance they will be able to achieve comparable levels of performance to that of younger people.

Original languageEnglish (US)
Pages (from-to)282-293
Number of pages12
JournalBehaviour and Information Technology
Issue number5
StatePublished - Jan 1 1998

ASJC Scopus subject areas

  • Developmental and Educational Psychology
  • Arts and Humanities (miscellaneous)
  • Social Sciences(all)
  • Human-Computer Interaction


Dive into the research topics of 'Understanding sources of user variability in computer-based data entry performance'. Together they form a unique fingerprint.

Cite this