PURPOSE
Medical students, particularly those pursuing procedural-based specialties,
are expected to develop strong fine motor skills, manual dexterity, and
hand–eye coordination. The O’Connor Tweezer Dexterity Assessment (OCTDA),
originally developed for occupational selection, has been used in various
applied fields but has limited evaluation within medical education. This
study examined the measurement consistency to validate the reliability of
OCTDA as an indicator of medical students’ psychomotor proficiency.
METHODS
Forty-two medical students completed the OCTDA in a pre-clerkship and
post-clerkship design. Two timing metrics of both hands were recorded using a
digital stopwatch: single-lap recorded time and accumulated time across five
laps. To ensure accuracy, students were excluded if recorded time differed
from accumulated time by more than five seconds, resulting in 38 students in
the final analysis. Linear mixed-effects models were used to evaluate sources
of score variability and estimate reliability.
RESULTS
OCTDA showed moderate to good reliability (Intraclass Correlation
Coefficient (ICC) ? .72 for right hand; ICC ? .61 for left hand), indicating
that most performance differences reflected true individual variation. The
left-hand model demonstrated a significant moment effect (b = –33.09), with
faster post-test times suggesting a practice effect. No moment effect emerged
for right-hand performance, reflecting stable dominant-hand proficiency.
Proctor effects were negligible in the left-hand model and small in the
right-hand model, indicating minimal evaluator influence. A significant dominance
effect was observed in the right-hand model (b = 86.22), with
left-hand–dominant students performing worse using their right hand.
CONCLUSIONS
OCTDA demonstrated reliable measurement of student psychomotor performance.
The improvement in left-hand scores suggests sensitivity to practice effects
while right-hand performance remained stable. Minimal evaluator impact
supports the fairness and standardization of the assessment. Residual
variability could be further reduced through additional practice trials.