The admittedly provocative title of this post is definitely wrong. Grants matter because they can pay people that work on the grant, and grants provide funds to conduct research. Grants matter for a lot of institutions because they often include “overhead” or “F and A” or “Indirect” funds that essentially pay for the maintenance costs of conducting research at universities. But where I think grants don’t matter, or at least don’t matter as much as they are currently weighted, is in evaluating research quality and productivity.
The only thing that makes what we do science, rather than an expensive hobby, is that the results get shared via scientific publications. Of course, results can also be shared in grey literature like reports, or via database acquisitions, but these are not universally available and so are not the focus of my argument. Hence, the only research that matters for the broader scientific enterprise is research that gets shared with the broader community in scientific publications
My basic argument is that grants are only mediators of research productivity and quality, and are not themselves a type of research product. Grants are frequently important for evaluation of scientists, and at research-intensive institutions, are the most important factors for tenure and promotion. However, the grants only matter for the progression of science if they result in publications. It does not matter how much money you bring in if that money does not result in published research.
Because obtaining grants is so competitive, they can be a signal of research prominence or quality. But grants alone are an imperfect signal for two reasons. First, not all grant funds are equal. Some research funding is more like a contract, rather than the funding of a research program. For example, scientists can be funded for monitoring work that may well be important for informing agency decisions, but is not hypothesis driven and is less likely to result in publications. In my experience, this work is less likely to push science forward, although it can contribute to other important societal outcomes. But for evaluating research quality and productivity, receiving the grant is not a good signal, although the resulting publications would be a good signal. Second, some grants fund specific research programs (e.g., regular NIH and NSF grants), and this work is more likely to be hypothesis driven and advance the scientific enterprise. However, the work has to be published to actually matter, and so we should use the resulting scientific publications to evaluate productivity, rather than simply receiving the grant.
There are other reasons to conduct research than advancing knowledge in your field. As mentioned before, monitoring research can be important for a number of reasons. Research can be a crucial part of teaching in labs and field courses, even if the goal is not publications. But then I would argue that this research (and the funding that supports it) is best evaluated as broader impacts or part of teaching, not as a way to evaluate research productivity.
I understand why institutions value the acquisition of external funding for evaluations like tenure and promotion. Requiring all faculty to generate some income for the university in the form of F and A is a great way for the university to recoup salary and startup of faculty. But as an individual faculty member, whether some gets funding or not does not really matter to me for evaluating research productivity. Of course, getting funding might be a good measure of how they support the graduate program, or training of undergraduates, but that is a different type of evaluation. Indeed, research productivity is only one of several evaluation categories to consider for hiring or promotion. However, for evaluating research productivity, I pay attention to the quantity and quality of scientific publications, which is the only real direct evidence of how the research of an individual faculty has influenced their field and contributed to the corpus of human knowledge.
The only thing that makes what we do science, rather than an expensive hobby, is that the results get shared via scientific publications. Of course, results can also be shared in grey literature like reports, or via database acquisitions, but these are not universally available and so are not the focus of my argument. Hence, the only research that matters for the broader scientific enterprise is research that gets shared with the broader community in scientific publications
My basic argument is that grants are only mediators of research productivity and quality, and are not themselves a type of research product. Grants are frequently important for evaluation of scientists, and at research-intensive institutions, are the most important factors for tenure and promotion. However, the grants only matter for the progression of science if they result in publications. It does not matter how much money you bring in if that money does not result in published research.
Because obtaining grants is so competitive, they can be a signal of research prominence or quality. But grants alone are an imperfect signal for two reasons. First, not all grant funds are equal. Some research funding is more like a contract, rather than the funding of a research program. For example, scientists can be funded for monitoring work that may well be important for informing agency decisions, but is not hypothesis driven and is less likely to result in publications. In my experience, this work is less likely to push science forward, although it can contribute to other important societal outcomes. But for evaluating research quality and productivity, receiving the grant is not a good signal, although the resulting publications would be a good signal. Second, some grants fund specific research programs (e.g., regular NIH and NSF grants), and this work is more likely to be hypothesis driven and advance the scientific enterprise. However, the work has to be published to actually matter, and so we should use the resulting scientific publications to evaluate productivity, rather than simply receiving the grant.
There are other reasons to conduct research than advancing knowledge in your field. As mentioned before, monitoring research can be important for a number of reasons. Research can be a crucial part of teaching in labs and field courses, even if the goal is not publications. But then I would argue that this research (and the funding that supports it) is best evaluated as broader impacts or part of teaching, not as a way to evaluate research productivity.
I understand why institutions value the acquisition of external funding for evaluations like tenure and promotion. Requiring all faculty to generate some income for the university in the form of F and A is a great way for the university to recoup salary and startup of faculty. But as an individual faculty member, whether some gets funding or not does not really matter to me for evaluating research productivity. Of course, getting funding might be a good measure of how they support the graduate program, or training of undergraduates, but that is a different type of evaluation. Indeed, research productivity is only one of several evaluation categories to consider for hiring or promotion. However, for evaluating research productivity, I pay attention to the quantity and quality of scientific publications, which is the only real direct evidence of how the research of an individual faculty has influenced their field and contributed to the corpus of human knowledge.
RSS Feed