The calibration of a sit-ups task using the Rasch Poisson Counts model.

W. Zhu, M. J. Safrit

Research output: Contribution to journalArticlepeer-review


The purpose of this study was to calibrate a nationally used sit-ups test using the Rasch Poisson Counts model and evaluate the model-data fit. The total number of subjects was 8,723, consisting of 4,486 girls and 4,237 boys, ages 10 to 18. The estimated difficulty of the sit-ups task was -2.80, which was appropriate for a majority of examinees whose ability levels ranged from .09 to 1.39. After the calibration, boys and girls as well as different age groups were compared under the same metric. Graphs of the model-data fit demonstrated that the model-data fit at a low ability level was not as good as the fit at a high ability, which could be caused by violation of assumptions of the model that examinees have the same performance speed throughout the test and that the speed at a given time is independent of the number of sit-ups completed so far.

Original languageEnglish (US)
Pages (from-to)207-219
Number of pages13
JournalCanadian journal of applied physiology = Revue canadienne de physiologie appliquée
Issue number2
StatePublished - Jun 1993
Externally publishedYes

ASJC Scopus subject areas

  • Physiology
  • Orthopedics and Sports Medicine


Dive into the research topics of 'The calibration of a sit-ups task using the Rasch Poisson Counts model.'. Together they form a unique fingerprint.

Cite this