5 Tips for Analyzing Data Insights Questions on the GMAT

Especially when you see a Table or a Graph, quickly summarize for yourself the relationship between the variables in each table, chart, or graph. Do they have a direct or indirect correlation? Does the data spike or significantly decrease at certain points?

One of the most common mistakes on the Data Insights section is using the wrong information because of a slight grasp of the presented information. The data you need to solve DI questions must lie on the various screens; you just have to know where to look.

First, understand what the question is asking, then stop and consider which table, graph, chart, or part of the passage provides the relevant information you’ll need to solve for the correct answer. Harder DI questions will require you to use more than one screen or ask you to take information or figures from one screen and apply it to another.

Mentally categorize each graph, chart, and table. Do not just skip the screens entirely and go straight to the question/s on DI. While you may think this will save you time, it actually significantly decreases your accuracy.

Make sure you read every tiny piece of writing on or near the data, including titles, the labels for the x and y-axes, column names, and even footnotes, if any. Analysis and synthesis are the tested skills in DI, and you’ll need to fully comprehend the data to answer questions accurately. Unlike Reading Comprehension, skimming will never work here.

Once you understand the labels, take special care to note the units (mph, m/sec, cm², etc.). Are we dealing with seconds, minutes, or hours? Does one graph represent a year, while the other graph represents the specific months? The units may change from graph-to-graph or chart-to-table, and some DI questions might ask you to do simple conversions as you move between the screens.

Most scientific reasoning either goes from broad to specific, or from specific to broad. Pay attention to the trends in the presented information. What kinds of generalization are possible, and does the remaining data corroborate this, or contradict it?