Data Insights (DI) Fundamentals Flashcards
Multi-source Reasoning Questions
What are the best strategies for answering Multi-Source Reasoning questions?
DI Question type
What you get:
* Data on 2-3 tabbed pages (text, tables, or graphics).
- Requires analysis of different sources of information to answer the question.
- Some questions are multiple choice, others are true/false or yes/no.
Strategy:
* Don’t expect to know the material; focus on analyzing given data.
- Examine each source carefully, as questions require detailed understanding.
- Familiarize yourself with any graphics (tables, graphs) to understand the information presented.
- Select the answer with the most support based on the data provided, regardless of prior knowledge.
Data Sufficiency Questions
What are the best strategies for answering these questions?
DI Question type
What you get:
* A question followed by two statements (1) and (2).
- Assess whether either statement alone, both together, or neither provides enough data to answer the question.
Strategy:
* Don’t solve the problem outright, only determine if you have enough info.
- First, evaluate each statement individually, then consider them together.
- Pay attention to whether the problem requires a specific value or a range.
Table Analysis Questions
What are the best strategies for answering these questions?
DI Question type
What you get:
* A sortable table of data (like a spreadsheet) along with some brief text explaining the table.
- Questions about whether statements are consistent with the data presented.
Strategy:
* Examine the table first and note the type of information it presents.
- Sort the table when necessary to help clarify relationships or identify specific data.
- Read each statement carefully to see if it matches the table’s data, ensuring accuracy.
Graphics Interpretation Questions
What are the best strategies for answering these questions?
DI Question type
What you get:
* A graph or other graphic (scatter plot, bar chart, etc.) with statements to complete using a dropdown menu.
Strategy:
* Carefully interpret the graph, noting values, labels, and scales.
- Read any accompanying text for additional context.
- Complete the statements using the option that makes the statement most accurate or logical, focusing on precision.
Two-Part Analysis Questions
What are the best strategies for answering these questions?
DI Question type
What you get:
* A brief scenario with two interrelated tasks.
* Answer choices are presented in a table format with one column for each task.
Strategy:
* Read the problem carefully and understand both tasks.
- Make choices for each task and review all options before selecting your final answers.
- Determine if the tasks are independent or dependent on each other—if dependent, both must be correct to have a complete answer.
Data Concepts
Data set
- A data set is a structured collection of related data about a specific topic. It is often displayed in tables, charts, or both.
- Almost every GMAT DI question is based on a data set — e.g., company profits by year, employee details, or product sales.
- Understanding that all visuals come from a larger data set helps you mentally group and track variables and relationships.
- It allows you to analyze multiple variables across multiple cases efficiently and spot trends, relationships, and anomalies.
Data Concepts
Case
A case is an individual unit or thing being studied — like an employee, a product, a region, or a year.
Each case gives context to the data—it lets you understand who or what the variables are describing. Usually, each row in a table represents one case.
Data Concepts
Variable
A variable is a specific kind of data collected for each case — like salary, position, or year hired.
Variables help categorize and analyze specific aspects of each case. Each column in a data table usually represents one variable. Many DI questions ask you to:
- Compare variables across cases
- Calculate differences between variables (e.g., profit = revenue − expense)
- Interpret trends in variables across time
- If you don’t understand what the variable means, you might misinterpret what’s being asked — e.g., confusing “total cost” with “cost per unit.”
Data Concepts
Data Point
A data point is the value of one variable for one case — usually found in a single cell of a table.
Data points form the foundation of all analysis—patterns, trends, and conclusions come from interpreting individual data points across cases and variables.
If a question asks:
“What was the net profit of the electronics division in Q4?”
You’re looking for a specific data point — one variable, one case.
Precision matters here. If you look in the wrong row or column, you lose points.
Data Concepts
Record
A record is a full set of data points for one case — usually a row in a table. It provides a complete profile of a case, allowing for case-by-case comparison or identification.
DI questions often ask you to:
- Evaluate all the details about a single record (e.g., one employee)
- Identify patterns across multiple records
- Sort or filter records based on certain criteria
Example:
“Which two employees hired after 2020 earn over $60K and are full-time?”
You need to check the record (row) of each employee.
Data Concepts
Dependent Variables
A dependent variable is derived from others (e.g., profit = revenue − expenses).
It’s value depends on one or more other variables (usually independent ones).
Dependent variables are typically the outcome or result being studied. They’re central in identifying relationships and making predictions (e.g., profit depends on revenue and expense).
Data Concepts
Independent Variables
An independent variable stands on its own (e.g., revenue, expenses).
It’s value is not affected by other variables in the dataset.
Why it matters:
Independent variables often explain or influence changes in other variables (i.e., they help you understand why something changes).
Data Concepts
Qualitative Data
Qualitative Data
Qualitative data is any type of data that isn’t a number used for calculations. It includes words, names, symbols, checkmarks, colors, file links — even numbers that don’t represent quantities (like phone numbers or ZIP codes).
- Essential for interpreting text-based tables and dropdowns in Table Analysis, Multi-Source Reasoning, and Graphics Interpretation.
- Recognizing qualitative data prevents inappropriate math operations like averaging labels.
- DI visuals often include non-numeric data: names, categories, weekdays, status labels (“Yes/No,” “Red/Green”), etc.
- Knowing that not all data can be averaged or subtracted prevents calculation errors.
Data Concepts
Nominal Data
Qualitative Data
- Nominal data has no meaningful order. Think of first names, colors, departments.
- You can’t compute mean, median, or range with nominal data — but you can find the mode (most frequent value).
- Example DI Question: “Which product category had the highest number of sales?” You’d look for the mode (most frequent category) — not do math on the names.
Data Concepts
Ordinal Data
Qualitative Data
Ordinal data has a meaningful order, but values still aren’t numeric. Think: ratings (“Poor, Fair, Good”), days of the week, priority levels.
- You can’t do arithmetic, but you can find the median and the mode.
- You can rank or sequence values, but not compute with them.
- Important in Table Analysis and Graphics Interpretation when sorting or comparing ranks, grades, or levels.
You must know:
Mean/range = ❌ not valid here
Median/mode = ✅ useful
Data Concepts
Binary Data
Qualitative Data
- Binary data only has two possible values — like Yes/No, True/False, or checkmark/blank.
- Understanding binary structure helps you scan and count only one of the two categories quickly.
- Useful for filtering, tagging, and calculating proportions.
- Can be nominal (if order doesn’t matter) or ordinal (if order does).
- Frequently seen in Table Analysis (e.g., Full-Time column = checkmark or blank), and in Multi-Source Reasoning for yes/no judgments.
- Tip: If values are shown with symbols or blanks, this is still binary data — you just need to identify which value each symbol represents.
Data Concepts
Partly Ordered Data
Qualitative Data
Partly ordered data: is when some values can be ranked, but not all—like a family tree without birthdates. You may know A > B and B > C, but not A vs. D.
You can only draw limited conclusions—e.g., Haruto > Minato in age, but not Akari vs. Minato.
Median does not apply unless you’re isolating a fully ordered subset.
Useful for identifying incomplete hierarchies or relationships in Multi-Source Reasoning and Two-Part Analysis.
You may be asked:
“Who is definitely older?”
“Which task must come before another?”
Knowing what can and can’t be inferred helps avoid trap answers based on assumptions not supported by data.
Statistical Use:
Median often can’t be determined for partly ordered data, unless a subset is fully ordered.
Data Concepts
Quantitative Data
- Quantitative data refers to numerical values that can be added, multiplied, averaged, and more.
- Examples: Salaries, temperatures, distances, revenue.
- You can apply statistical measures like mean, median, mode, range, and standard deviation.
- It forms the basis of mathematical analysis in Table Analysis, Graphics Interpretation, and Two-Part Analysis.
Data Concepts
Continuos
Quantitative Data
Quantitative Data
- Continuous data can be divided infinitely—there are always possible values between two measurements.
- Examples: Temperature, height, weight.
- It has no mode typically, but you can calculate the range and standard deviation.
- Helps in understanding trends and patterns in Graphics Interpretation (e.g., temperature variations over time).
- Continuous data involves more complex analysis, especially when identifying trends or calculating range or standard deviation.
Data Concepts
**Discrete **
Quantitative Data
Quantitative Data
- Discrete data consists of distinct, separate values—whole numbers only.
- Examples: Number of students, number of courses, prices in currency.
- It can be counted or categorized easily.
- Useful in Table Analysis and Two-Part Analysis where counting or specific comparisons are required.
- Discrete data is easier to count and compare, often requiring basic operations.
Data Concepts
Interval Data
Quantitative Data
- Interval data uses a scale with equal distances between values but no true zero point.
- Examples: Temperatures in Celsius or Fahrenheit, calendar years.
- You can add and subtract interval values (but not multiply or divide them).
- Used for time-related data in Multi-Source Reasoning and Table Analysis, but ratios (e.g., “twice as hot”) don’t apply.
Data Concepts
Ratio Data
Quantitative Data
- Ratio data is numeric data with equal intervals and a meaningful true zero, allowing full mathematical operations.
- Enables accurate comparisons and proportions.
- Validates reasoning like “twice as much,” “half the size,” etc.
- Critical for solving quant-based GMAT questions involving real-world quantities like weight, cost, duration, etc.
Data Concepts
Logaritmic Data
Quantitative Data
- Logarithmic data represents exponential changes in values, where each unit increase represents a multiplicative jump (e.g., a 10 dB increase is 10× more intense).
- it works with e xponential growth: Larger numbers represent greater real differences than smaller numbers.
- Cannot add or subtract raw logarithmic values (e.g., 20 dB + 10 dB ≠ 30 dB)
- Used in decibels (sound), pH levels, Richter scale, etc.
- You’ll need to interpret trends on logarithmic scales (often in Graphics Interpretation).
- Understanding the scale helps you avoid linear assumptions and correctly interpret relative changes (e.g., a 40 dB noise is 10× louder than 30 dB).
Data Concepts
Data Distribution
Distribution definition
- A distribution is the pattern in how frequently different values occur in a data set.
- It helps you see what values are common, rare, or predictable.
- You’ll interpret distributions in charts and graphs to assess variability, central trends, and the likelihood of future values.
