A Likert Scale Versus a Visual Analogue Scale and the Participant Response: A cross Sectional Study

This study aims to assess whether the VAS responses are as reliable as the Likert responses from the perspective of dental students. The preference of each scale and the influence of gender and level of education on the preferred format were also assessed. This causal comparative quantitative study was conducted at the College of Dentistry (COD) at King Saud bin Abdulaziz University for Health Sciences (KSAU-HS) using a non-probability convenient sampling technique with all second, third and fourth year male and female dental students. The survey instrument used was the well-established course evaluation at KSAU-HS. The questionnaire was distributed at two different times to the same respondents with the different measurement scales. Participants who completed both formats of the survey were included. NCSS software (Version 20) ® was used for data entry and analysis Of the 244 participants invited to participate, 127 responded to both surveys yielding a response rate of ~52%. Comparing the Likert and the scroll bar responses, the equivalence threshold criterion (a minimum of 0.7 lower band of the 95% confidence interval for ICC) was met for all the scale/item scores, except for four of the 16 items. The estimated ICCs for the rest of the items ranged between 0.67 and 0.84. The measurement equivalence criteria was preserved for the overall ICC (0.854, 95% CI 0.793 – 0.897). The males were 2.9 times more likely to prefer Likert type responses than the females. There was no conclusive evidence that either the VAS or the Likert scale was superior.