A theoretical climate change example might be:
Car A emits 10% less CO2 than car B.
Driving is 30% of your annual emissions (i.e. it's significant and one of the low hanging fruit).
That 30% of your annual emissions if multiplied by 7 billion people is likely to cause X.
If everyone drives car B instead of car A, Y will happen instead.
Y is better than X, because we end up with a more verdant Earth, cheaper food, better weather, less war (insert reason here).
Therefore, our overall quality of life is higher if we switch to car A.
Without the whole information chain, you just have isolated statistics that don't really mean anything.
Convincing argumentation is not only about mathematics, it's about appealing to the things that matter to people.
The problem is that while it's easy to calculate exactly how much less CO2 car A emits than car B, it is close to impossible to use science to predict to any useful degree of accuracy how much cheaper the food, better the weather, or less of the war we will get. It's often hard to even predict the sign, not even trying to determine the magnitude. This means that when scientists or politicians tell you that "we need to do X, so that we get Y", they're very often fooling you, because they have no good idea how much (if any) Y you get for each unit of X.
The other one that pops to mind is probably statistical calculation helping services.
Maybe there are lots of other things as well - I haven't been involved with scientific research in a long time.