Toward the development of optimal calling protocols for telephone surveys: a conditional probabilities approach
This study confirms the results of previous research indicating that the chances of finding a respondent at home and conducting a telephone interview on the first call are much better on weekday evenings and on weekends than during weekday daytime hours. By examining call outcomes across a series of three unanswered calls, it was also demonstrated that this general principle applies to second and third “cold” calls as well, while also providing some evidence that repeated calls during the same “optimal” time period may yield diminishing returns. We also see how one might best “recover” from a non-optimal initial call through more efficient scheduling of second and third calls. However, when optimal and nonoptimal time slots were mixed, better results were obtained when the former were called earlier in the sequence, thereby emphasizing the critical importance of optimal timing for the first call in particular.
Kulka, R., & Weeks, M. (1988). Toward the development of optimal calling protocols for telephone surveys: a conditional probabilities approach. Journal of Official Statistics, 4(4), 319-332.