Robots such as those serving as educational tutors, healthcare supporters, and collaborative partners must develop “rapport,” a construct that encompasses mutual understanding and interpersonal connection with people, to ensure their long-term success. In earlier work 60, we constructed, evaluated, and validated an 18-item Connection-Coordination Rapport (CCR) scale to measure human-robot rapport (Studies 1-3). Even though the full-length 18-item CCR scale measures rapport thoroughly, it may not always be practical for researchers to adopt given its relatively long length. Therefore, in this work, we developed a reduced-length version of the CCR scale that still effectively measures rapport using just 8 items. Following recommended practices for short-form development and validation, we leveraged the input of HRI experts (Study 4, \(N=30\) ) to shorten the CCR scale from 18 items to 8 items (4 items per factor). Then, we evaluated this reduced-length CCR scale on a new sample (Study 5, \(N=186\) ) where online participants watched a human-robot interaction video and evaluated it using both the full-length and reduced-length CCR scales. We validated the reduced-length CCR scale by showing that it has high internal reliability, high overlap with the full-length CCR scale, a consistent factor structure, high construct validity, and significant time savings.
Lin et al. (Mon,) studied this question.