Comparing 'challenge-based' and 'code-based' internet voting verification implementations

Internet-enabled voting introduces an element of invisibility and unfamiliarity into the voting process, which makes it very different from traditional voting. Voters might be concerned about their vote being recorded correctly and included in the final tally. To mitigate mistrust, many Internet-enabled voting systems build verifiability into their systems. This allows voters to verify that their votes have been cast as intended, stored as cast and tallied as stored at the conclusion of the voting period. Verification implementations have not been universally successful, mostly due to voter difficulties using them. Here, we evaluate two cast as intended verification approaches in a lab study: (1) "Challenge-Based" and (2) "Code-Based". We assessed cast-as-intended vote verification efficacy, and identified usability issues related to verifying and/or vote casting. We also explored acceptance issues post-verification, to see whether our participants were willing to engage with Internet voting in a real election. Our study revealed the superiority of the code-based approach, in terms of ability to verify effectively. In terms of real-life Internet voting acceptance, convenience encourages acceptance, while security concerns and complexity might lead to rejection.