The data science field is maturing into its own. The hype that heralded the discipline’s arrival on the public stage is giving way to a more measured view of how its tools can enhance decision-making. But amid growing expectations and the rapid pace of innovation, those in the field experience frequent challenges to their current skill and their attempts to grow. I believe that mentorship should play a stronger role in data scientists’ professional development to help overcome those challenges.
In a survey we at Anaconda conducted to get an idea of a day in the life of a data scientist, more than half of the roughly 2,000 respondents reported getting “stuck” every week, which we defined as hitting an obstacle that requires them to do more research, consult a colleague or friend, or acquire additional skills to complete the task. Some of these uncertainties trace back to the rapidly evolving nature of the field; with innovations coming out so frequently, it can be hard to keep up with the latest best practices. In their search for answers on how to get “unstuck,” nearly 80 percent of respondents said they resort to external sources like Q&A sites.
There are also other challenges aside from the more technical ones. For instance, working with business leaders can present its own difficulties. Organizations bring data practitioners into business decisions, but there can be a disconnect between delivering insights and how leadership interprets them. In the 2021 State of Data Science survey, only 36 percent of respondents said their organization’s decision-makers are very data literate and understand the stories told by visualizations and models. On the other hand, nearly a quarter of enterprise respondents listed “business knowledge” as lacking from data practitioners at their organization. With these gaps in understanding, it’s unsurprising that data scientists sometimes feel they struggle to impact business decisions.
Potential Advisees and Mentors Need to Be Connected
Amid these types of challenges, it makes sense that data practitioners are looking for advice. Nearly 70 percent of respondents to our “Day in the Life” survey were interested in finding a mentor. These data scientists could be seeking mentorship for various reasons, whether to solve technical challenges, better understand how to interface with business leaders, or navigate their career paths. It’s a common misconception that data scientists are always behind a computer just crunching numbers. They often spend time explaining the models, presenting, and closely working with others across teams – skills that might require mentorship to develop, especially if they are new to the field. With Generation Z entering the workforce, the need for mentors is increasing year over year; compared to 2020, this year’s State of Data Science survey saw a 15 percent increase in respondents from Gen Z.
While the desire for mentorship is clear, there appears to be a disconnect between demand and access, with more than half of respondents to our “Day in the Life” survey saying that they weren’t sure where they could find a mentor, even though nearly 40 percent of those surveyed said they’d be willing to serve as one. With the relative newness of the data science industry, there aren’t as many established venues to build these relationships, and with the variety of titles that encompass data science, practitioners aren’t always sure who to look to for advice. This is a key area where our field needs to improve.
The Disconnect Between Mentors and Advisees in Data Science
- Nearly 70 percent of respondents surveyed were interested in finding a mentor.
- More than half of respondents weren't sure where to find a mentor.
- Almost 40 percent of those surveyed were willing to serve as a mentor.
Mentorships Offer Growth Opportunities for Advisors, Too
Professional development gaps aren’t surprising for an emerging industry but will impede its future development if left unaddressed. For those working alone or in small enterprises, it’s unlikely they can find support internally. How can data science professionals and organizations work together to build the networks and relationships to propel the discipline forward?
Data scientists can improve their resumes by becoming mentors themselves, even if they don’t feel like titans in the profession. A growing field like data science offers an excellent opportunity to build leadership skills. Similarly, the best mentors are not necessarily the most well-known or organizationally senior, but rather provide an informed yet objective perspective. Teaching and doing are very different skills. When looking for a mentor, seeking advice from someone who has experience, either academically or professionally, will help you gain more knowledge of the field and grow your network.
Build Community to Advance the Field as a Whole
Practitioners should also look outside programming forums and consider professional associations, networking groups, meet-ups (virtual or in-person), or reaching out to those who lead courses on Coursera or Udacity. Along with broader groups and cohorts that you can find on Reddit and Github, there are many more specialized, sector-specific, and regional groups out there or are waiting to be started – such as a governing body for data science. A professional cohort for data scientists can better equip them to tackle complex but essential issues. Similar to how other professions have a governing body, like the American Medical Association, creating a cohort for the data science community would not only serve as a place for practitioners to elect a governing board to guide industry-wide development, but also allow individuals to connect and share best practices, and put an emphasis on the value of community in the field.
Finally, it’s in companies’ best interest to support the data science community. That means sending their data science teams to conferences, covering the cost of additional education, sponsoring working groups, and providing opportunities to build mentor/advisee relationships. An investment in data science tools is, first and foremost, an investment in people.
Data scientists are shaping the future. While there is no set path for entering the profession, data science organizations need to shift their focus from proving what’s possible to nurturing the craft and those leading it. The field still has vast untapped potential in front of it, but to realize this vision, every community member should support the philosophy of going further together.
