How to Gauge Scientific Aptitude?

Today on twitter, I read a complaint with hundreds of likes and dozens of comments about the review committee commenting on F31/F32 applicants undergrad/grad grades. The tweet mentioned that the applicants should be reviewed on their scientific aptitude only, not their grades. Dozens of commenters concurred with this tweet and shared their personal stories about being dinged because of a past low grade.

I have to say that I absolutely agree that grades are not always a reliable predictor in any professional environment. I myself have seen high graders with a horrible working attitude and vice versa. I also rarely make decisions based on grades, or place grades as an important factor for consideration at all. However, I can do this often because, at the current stage, I can still interview most of the applicants, and learn about them in person, which can be quite helpful in further determining whether he or she fits a position or not, putting grade aside.

In the case of a grant or position with a large number of applicants, making interviewing every single one of them impossible, I want to discuss: what factors then should we consider when selecting an applicant for award/grant/fellowship/position? Or more specifically, how to objectively decide the aptitude of a person for a position? In the same example of a scientific grant, should we just consider the track record and the research plan? Yes to research plan, because a good applicant should have a clear and well-founded idea of what to pursue. And yes to track record, but only in a perfect world where all applicants have the same or similar level of access to publication. In my experience, a good-looking track record is a blessing given to the very lucky and hardworking applicants. Therefore, the heavily luck-dependent factor (of course hardworking on the top of it), in my opinion, should also be balanced by heavily hardwork-dependent factors. One example is grade. Of course, grade itself can give more information than just being a balancing factor, but I do believe its importance in the case of “blind-selection”.

Additionally, when multiple applicants share similar “scientific aptitude” as determined by the track record and their research plan, using grade to differentiate them would be at least less subjective than “I think this research topic is more important”.

All in all, grade does not define scientific aptitude for sure, but it can be a quite useful and objective criterion when making blind selections, which makes it irreplaceable. I do hope there are better metrics, but will there be? Humans are not that easy to measure, right?