You can try to help making better decisions, e.g. citizen participation in policy making.

One example project is Polis (https://pol.is), which is used by the government in Taiwan. They wrote a nice paper about their approach (https://www.e-revistes.uji.es/index.php/recerca/article/view...). Maybe find ways to improve the idea? Or something completely different?

Allow me to be a Debbie Downer but the whole idea of statistics and data-driven decisions sort of falls apart if people are not able to interpret the output, distrust the input, or are just cynical.

If you’re OK with having a small, unmeasurable impact I’d recommend focussing on data literacy, or helping educate people to perform your trade in the organisations they are already in.

cool! will check it thanks

MRNA technology R&D is a data problem (to my limited understanding)

Demonstrate the effectiveness of vaccines or other medical interventions that are difficult to understand and often shrouded in mystery. If your findings support their use it could save lives if it gives people the confidence to take those interventions.