Separate lead-form age effects from voice-based age effects

Determine whether callers’ greater persistence with older-presenting targets is caused by the date of birth or age printed on the seeded lead-generation form, by the target’s apparent age conveyed through voice, or by both mechanisms.

Background

The experiment randomized complete identities rather than apparent age in isolation. Each identity differed in several attributes, including voice, name, biography, address, and date of birth, and the analysis found that apparent age was the dimension along which the identities ordered most strongly.

Because the seeded form included a date of birth while the live call conveyed a voice, the design cannot determine whether callers responded to information visible before dialing, to vocal cues during the conversation, or to both. A future experiment would need to vary these channels independently.

References

Nor can we separate two channels by which age could act, since the seeded form carries a date of birth: a caller may be scripting from the age printed on the lead, or reacting to the voice on the line.

— Anatomy of a Scam Call: What 10,000 real scam and spam calls reveal about how phone scammers operate  (2608.24127 - Traister et al., 25 Aug 2026) in Section 4.5, subsection “What the experiment cannot separate”