Skip to main content

See no evil in the voice-to-voice customer service context

Cite this dataset

Peasley, Michael et al. (2022). See no evil in the voice-to-voice customer service context [Dataset]. Dryad.


A sample of more than 28,000 front-line employee (FLE) - customer interactions, extrapolating from foundational framing, we pit conventional service approaches against one another to propose a dual-process model, situating customer frustration/satisfaction as mediators of the indirect relationships between resolution/relational tactics and call duration – a key customer service efficiency outcome.


In order to provide an internally valid empirical foundation to test our model, the sampling frame we adopt was based on a random sample of inbound calls received over a one-year time frame, resulting in 28,103 dyadic interactions, handled by a total of two-hundred and seventeen separate FLEs, employed at four call centers located across the US. The data on which our analyses are based was provided by a Fortune 500, nationally – branded, market-leading insurance company headquartered in the US.

To account for the nested structure of the complex data that underlie our analyses, we estimated one omnibus moderated-mediation, random-effects model in Mplus v. 8 (Muthén & Muthén, 2017). This approach allows us to take into account complex sampling features related to our panel data, including generating robust standard errors (Thompson, 2011), teasing out variations potentially caused between FLEs, thereby controlling for omitted, unobserved effects related to the ways in which the 217 FLEs handled each call (Germann et al., 2015).

Usage notes

We are including our analysis code because the data is not available due to its proprietary nature and a non-disclosure agreement. An example of the data has been provided.