The data science landscape has changed drastically, making it tough for analytics professionals to know where to focus their growth. As you know, we at IADSS are exploring various data sources to better understand the knowledge and skills that professionals in analytics and data science field are expected to have. In this brief article, we want to share some initial insights from two data sets we are working on:
The skills listed by the universe of Data Scientist, Data Engineer and Data Analyst profiles on LinkedIn.
200+ job postings from Fortune 2000 companies which were posted in 2018.
We also assumed that some professionals might prefer not to mention what skills they have in their LinkedIn profiles. Thus, skill percentages were normalized in order to provide a better comparison.
Comparing these two sources will also provide us an understanding of what employers expect in certain roles vs. what the people with those titles typically list as their skills.
One important point we would like to highlight is when companies or individuals think of data science and analytics related skills or knowledge, the majority of the items are programming languages or specific technologies. While we recognize these are important to perform on the job, it is critical to know, as many of you who have been in this field for some time can attest to, they are also usually transient and can change over time or from position to position. So at IADSS, we are looking into a much broader and deeper set of knowledge and skills inventory, and recommend job seekers to focus more on fundamental skills on which they can build tool skills. You can join the research as a contributor or follower via a super-short form at https://www.iadss.org/contact.
So with that in mind, here are some findings with the first 10 core skills for Data Scientists, Data Engineers and Data Analysts, in relation with relevant job postings featured in 2018.
You can see Python listed in four out of every five Data Scientist job postings. It is the most common skill demanded by companies and the most common skill that all data scientists have on their LinkedIn profiles. The second most popular programming language in the job postings is R compared to C++ for individuals. Also on job postings, you can find Java and Scala but they are not as common on Data Scientist LinkedIn profiles.
Companies prefer to list tools more specifically, e.g. frameworks like "Hadoop" and "Spark" in a job post, whereas individuals put them in bundles like "data analysis" or "machine learning".
54% of whole data scientists list MatLab, and 23% of them list Tableau among their skills. But those are rarely mentioned data scientist job postings which further point out to the confusion in defining what data science is made up of.
See the graphs below:
Python comes at the top for Data Engineers job postings, whereas SQL is clearly dominant as a data engineering skill. Java, Python, and Hadoop are other tools somewhat popular and common among Data Engineers.
For Data Engineer job postings, you see the heavy investment by companies into various big data technologies and tools. Check the graphs below:
You can see a high concentration of querying, basic analysis and visualization tools listed by both Data Analysts and employers seeking for Data Analysts. Top skills are Microsoft Excel and SQL for employees, yet some companies do not specifically indicate Microsoft Office knowledge as a requirement, whereas SQL seems to be the most outstanding skill for hiring. One fourth of data analysts also mentioned they are skilled in Tableau platform, where one third of companies look for Tableau skill as a requirement for their Data Analyst roles.
Tableau, Hadoop, Python are all sought by companies, as commonly mentioned in job postings, however none of them comes to the forefront as a runner-up in data analyst job market.
"Research", as a general term is commonly mentioned by more than half of Data Analysts in their skill-sets. You may see the graphs below:
Although we can see the gap between the knowledge/skills desired by the employers vs. what data science and analytics professionals list at a high level look at these data sets, the real insights are not in the averages but in the details of how organizations define data science and analytics roles, how individuals position themselves in the data science and analytics job market, and probably more importantly how these sets of analytics knowledge and skills can be assessed and measured.
Help Set the Standards for Analytics Professionals!
At IADSS, we will be diving deeper into data and expanding our research. You can join the Analytics and Data Science Standards research study as a contributor or follower via a super-short form at https://www.iadss.org/contact. Later, you can choose to fill in a 5-10 minute survey, and provide us with your valuable insight into the analytics professional landscape.