Science education should be method-oriented
I would argue that the most important reason why we teach science in school is to transmit the scientific method. Many people ascribe to the view that science education is utilitaristic, meaning that what matters is the content, which leads to discussions such as replacing analysis with statistics. However, they miss the point of what is robust over time. We live in a world where access to information is getting simpler and simpler. The number of things I am supposed to remember is smaller, because I can retrieve them very quickly (see: retrieval and storage strength).
However, what is missing is the method to validate whether the sources we find are valid or not, for example. And we are also lacking the idea that everything can be learned. I am confident I can learn about biology, or philosophy, or politics. But many people belonging to other fields think that natural sciences are un-approachable. This is mostly due to the lack of initial method and to the lack of resources that follow a different pattern, which would emerge from Curiosity-driven education.
Therefore, what must be strengthened is not the ability to calculate an integral of a trigonometric function, but the ability to pursue whatever question we pose ourselves. Especially at a younger age, when the cognitive capacities are expanding, schooling should not focus into limiting the access to questions in favor of ready-made solutions.
These are the other notes that link to this one.