There are many reasons for wanting to go to graduate school: You want to become a professor, you love pipetting colorless liquids, you really liked your biology classes in college, or you are burning with an insatiable desire to understand biological mechanisms. Whatever your reason, you should know what you're getting into: Not only does your future happiness depend on it, but this knowledge may also determine your success. Let me explain.
The difficulty of predicting future scientific success is often cited as a reason for graduate programs to admit many students—the reasoning goes that it's impossible to tell which students will succeed. In a recent MBoC Perspective, University of California San Francisco (UCSF) Associate Professor Orion Weiner (who was on my thesis committee) provides some data that suggests that this might not be so true (though his data have been debated here and here). The piece is titled "How should we be selecting our graduate students?" and is targeted at those involved in graduate school admissions.
By identifying two group of students, "highest-" and "lowest-ranked" (as rated by professors), the author was able to determine which characteristics that were visible during the admissions process, were the best predictors of success. Interestingly, GPA and general GRE scores were not good indicators (though there are probably cutoffs for these metrics in place early in the admissions process at UCSF that would skew the results). The subject GRE was somewhat correlated, but the best predictor was previous research experience.
What does this mean for prospective students? Weiner notes that the "strength of the letters (particularly from the primary research advisor) and the quality of the applicant's previous research and essays are massively important" and continue to be the primary factors UCSF considers in admissions. But if other graduate schools take this new information into account for their own admissions processes, it might also be wise for prospective graduate students to seek research opportunities early so that they can accumulate two to three years of experience before applying.
I'd argue, however, that spending a lot of time working in a lab is extremely valuable for reasons other than wooing the admissions committee. Not only will you have a head start on the technical as well as more esoteric aspects of doing science (much like a massive Kindergartener held back a year), you'll also be able to say, with greater confidence, whether benchwork is really right for you. As Weiner writes, "it is likely that students who have done <2 years of research do not know what they are getting into."
I suggest all students considering graduate school learn what it really is they are getting into, which requires reading beyond the descriptions on graduate program websites. In addition to reflecting on the ways you respond to success and failure in a lab environment, talk to current graduate students at all stages of their careers. Usually the senior ones are not as available during recruitment visits, but their opinions are also valuable. Keep in mind that over a third of students in biomedical PhD programs in the United States drop out, though those statistics will vary widely on an institutional basis. This number, as well as placement of recent graduates, is important data that prospective students must demand to see when considering a graduate school. It might also be useful to consider the big picture of the history (and potential future) of the biomedical research enterprise.
If, after reflection, reading, and most importantly, actual lab work, the chance to do research in graduate school is still irresistibly tantalizing to you, apply. By facing the cautionary tales head on, you will be better prepared to enjoy and benefit from this amazing opportunity to immerse yourself in research.