Are the Steps on Likert Scales Equidistant? Responses on Visual Analog Scales Allow Estimating Their Distances

Area of Science:

• Psychometrics
• Quantitative Psychology
• Statistical Modeling

Background:

• Likert items are widely used but their discrete steps may not represent equal intervals on an underlying continuum.
• This assumption of equal intervals is crucial for accurate data interpretation and analysis.
• Traditional methods lack direct ways to assess the equality of intervals in Likert response formats.

Purpose of the Study:

• To develop and validate a method for estimating the distances between consecutive steps of Likert items.
• To investigate the assumption of equal intervals in Likert scales using item response theory.
• To explore the implications for scoring Likert-based measurements.

Main Methods:

• Utilized item response theory, specifically the Continuous Response Model (CRM) and Graded Response Model (GRM).
• Administered items in both Likert and Visual Analog Scale (VAS) formats to capture continuous response data.
• Established a formal relationship between CRM and GRM parameters to estimate interval boundaries.

Main Results:

• Demonstrated a valid method for estimating interval distances through simulation studies.
• Empirical analysis of personality scale data confirmed the correspondence between VAS and Likert formats.
• Revealed significant and variable unequal distances between successive Likert scale steps across different items.

Conclusions:

• The assumption of equal intervals for Likert scale steps is often violated.
• The proposed method provides a quantitative approach to assess and understand interval properties of Likert items.
• Findings have important implications for the accurate scoring and interpretation of Likert scale data in research and practice.