Hone this skill to get better at data analysis
AICPA logo
AICPA logo
  • Home
man wearing headphones while working in home office

Hone this skill to get better at data analysis

1 month ago · 3 min read

One of the best ways to improve your data analysis skills isn’t by spending more time with spreadsheets — it’s by sharpening your critical thinking skills.

Critical thinking — the iterative framework we use to analyze, challenge, clarify, synthesize, test, and apply information to find solutions, as the AICPA® & CIMA® CPE course “CQ Originals: Critical Thinking” puts it — is a learned skill that’s essential for anyone in a data-centric profession.

Ray Sang, CPA/CITP, CISA, built his accounting career by honing his financial data analysis skills, but he named critical thinking as the “cornerstone” skill that’s propelled his wide-ranging career, which includes working at Google and founding his own financial robotic process automation (RPA) software company, Chipmunk Robotics.

“You can master all the computer science languages out there, but it's still not going to help you to make the data useful without critical thinking,” Sang said.

What kickstarts the critical thinking process? Getting curious

If critical thinking is a car that you steer toward solutions, curiosity is the gas pedal.

Becoming inquisitive about a data-driven problem will help you arrive at solutions faster and uncover information you might not have otherwise learned. Asking questions when interpreting data proves that you’re active and focused — key characteristics of a critical thinker.

Curiosity can also be an accelerant for a data-centric career. Sang said he found himself learning the coding language Python after searching for a better way to help his client improve data collection processes. That same insatiable curiosity led him to learn other programming languages that would expand his financial data analytics skill set, a set of technological skills that are becoming even more important as the line between financial professionals and data analysts’ blurs.

“Because I’m curious, because I want to help others, that's really how I quickly built my analytical skills and data programming skills,” Sang said. “I consider myself an accountant in my heart, but I don't really see a difference between data engineering or data science and accounting. It's just part of the nature of our job.”

3 questions to ask yourself when applying critical thinking to data

  1. How was the data collected?
    It’s easy to think of data, particularly quantitative data, as incontrovertible. The truth is there’s an unavoidable bias inherent in every data set. Learning the origin of numbers in each spreadsheet cell can reveal these biases.Suppose you’re asked to analyze the results of a nationwide poll. When you ask how pollsters collected data, you might learn that they only polled people walking through business districts in major cities between 8 a.m. and 9 a.m. on weekdays. That information might prompt you to ask if more polling is required to include the opinions of those who don’t live in major cities or don’t commute during those hours.Asking up front where data comes from can unveil potential biases in collection, for example. Sang said he always starts a project by asking about the data’s origin, including how and when it was collected, so he can understand what he’s analyzing.“Without understanding the data collection process, I think it's actually very hard to just summarize the data attributes,” he said.

  2. What’s the nature of the data?
    Data sometimes comes to you raw, meaning that it’s unprocessed and hasn’t yet been cleansed. Other times, you’ll receive a spreadsheet full of aggregated figures, like a consolidated financial statement. If the data is the result of some calculation, it could be worth investigating whether the arithmetic is reliable.

  3. Can I stand behind my interpretations?
    Data analysis should exist with the least amount of bias possible. Sang said that’s why some data projects, especially those in the pharmaceutical industry, call for what’s termed a “blind screening” or “blind summary,” in which you review data before forming a hypothesis about potential results. “You don’t want data analysts to examine data with a purpose at the beginning because they may put too much weight on the data favorable to their hypothesis,” Sang said.

The questions don’t end there. Learn to ask more revealing questions and develop critical thinking skills and receive CPE credit from the CQ Originals course “Critical Thinking.”

Ryan Lasker, CPA

Ryan Lasker is a senior technical writer at AICPA & CIMA, together as the Association of International Certified Professional Accountants.

What did you think of this?

Every bit of feedback you provide will help us improve your experience

What did you think of this?

Every bit of feedback you provide will help us improve your experience

Mentioned in this article




Manage preferences

Related content