Biostatistics – it’s a big word, often needs a spell checker, and something to do with p-values?
Not sure? Let’s find out!
I didn’t know much about biostatistics until I was in graduate school. Halfway through my master’s program in public health, I decided to pursue a career in biostatistics. Now, you might be wondering if I was going through an existential crisis wanting to change my career to something unknown without any prior knowledge.
Actually, no. The reason was quite simple; I wanted to make impactful changes based on evidence generated by data.
Biostatistics, at first glance, was quite intimidating. That’s why I decided to do background research and talk to professionals in the field to understand what skills I would need to become a successful biostatistician. There was a consensus among the experts of the top 4 essential skills of a biostatistician:
To be a biostatistician, you need to have contextual information to formulate a plan for your data analysis. Biostatisticians use various study designs – frameworks used to develop a procedure in data collection and analysis – to answer specific research questions. A few study design examples one should be familiar with are:
To learn more about these study designs, click here.
While this part may not be the most exciting for biostatisticians, we still need to talk about data collection and management.
The main goal of data collection is to accurately collect necessary information to answer your questions. The main goal of data management is to provide administrative oversight to manage data collection and data cleaning as you prepare for analysis.
Biostatisticians should have a working knowledge of computer programming languages to manipulate and manage data. Examples of statistical programs that many use are STATA, R, SAS, or Mplus.
All biostatisticians should feel comfortable in planning data analysis and performing standard statistical tests.
For data analysis planning, biostatisticians should have solid methodological background which includes knowledge of study design (mentioned above). They also need to know data visualization techniques such as making plots to explore patterns in the data.
Examples of a few statistical techniques biostatisticians should be familiar with are independent t-tests, paired t-tests, Chi-squared test, regression analyses, survival analysis, etc.
Biostatisticians don’t just crunch numbers and stare at excel sheets all day long. We consult and collaborate with researchers, and contribute to writing grants and manuscripts. We learn to effectively communicate our plans, designs, and results. Communication is a key factor in becoming a successful biostatistician.
But we also need to communicate to non-analytical folks. It is important to explain the meaning of p-values, graphs, and plots to people without any statistical knowledge.
One way biostatisticians can share their results to both analytical and non-analytical folks is through data storytelling.
Now that you know the 4 essential skills of becoming a biostatistician, do you think you are ready to tackle a career in this field?