An Experimental Test of the Satisfaction Mirror between Frontline Public Servants and Citizens

The aim of this randomized controlled trial is to test for the effect of job satisfaction on citizen satisfaction during face-to-face encounters between public service employees and citizens. The experiment addresses the question: Is high job satisfaction mirrored by high citizen satisfaction? This is a between-subjects experiment with a 3*2 factorial design (i.e. there are six arms). Data collection will be conducted in two phases. Phase 1 draws from members of the general population who reside in two public housing estates near City University of Hong Kong. Phase 2 draws from students and employees at City University of Hong Kong (employee participation pending City University Finance Office approval). Phase 1 will be conducted in a "lab-in-the-field" setting in a community center. Phase 2 will be conducted in computer labs at City University of Hong Kong. Both settings are chosen to achieve a near distraction-free environment. The experiment involves a short pre-treatment survey, the viewing of a randomly chosen version of a video (produced by the research team), and a post-treatment survey including attention checks, manipulation checks, outcome measurements, and the collection of covariates for correlational analysis and balance testing. There are six different versions of the video, all depicting an interaction between a public library assistant and a citizen. Both are actors. All filming was done in the same location on the same day (December 28, 2023). The videos are similar except for two experimental manipulations. The first experimental manipulation tests the direct pathway from job satisfaction to citizen satisfaction. It consists of the public library assistant's expressed job satisfaction. She expresses high job satisfaction in two versions of the video, low job satisfaction in two more, and acts neutrally in the remaining two. The second experimental manipulation is whether or not the public library assistant makes a mistake that is made obvious in the video.