GED Free

Making Inferences and Drawing Conclusions

Use clues in texts, charts and graphs to infer, conclude, predict and apply ideas to new situations.

Practice this skill 25 min Inferences and conclusions Leer en español
Key idea

What you'll learn

Authors don't always say everything directly. Good readers fill in the gaps by reasoning from clues. In this lesson you will learn to:

  • make inferences: logical conclusions based on evidence in the text;
  • draw conclusions that are supported by several details together;
  • make predictions about what is likely to happen next;
  • apply ideas from a text (such as a policy) to new situations;
  • combine information from a passage with a chart, graph or table.

How it shows up on the GED. Inference questions are among the most common on the RLA test. You'll see them with every kind of passage: stories, articles, workplace policies and texts with graphics. Typical stems:

  • "Based on the passage, it can be inferred that..."
  • "Which conclusion is best supported by the passage?"
  • "Based on the passage and the graph, ..."
  • "Which employee would qualify under this policy?"
  • "What will the character most likely do next?"

The key word in all of them is support. The right answer is never just possible. It is the one that the text gives you the best evidence for.

Words to know

Key terms

  • inference — a logical conclusion based on clues plus what you already know (wet umbrella by the door → it's probably raining).
  • conclusion — a judgment reached by putting several details together (sales fell three years in a row → the product is losing popularity).
  • evidence — the specific words, facts or data that support an idea ("Many plot holders live in apartments with no yard").
  • prediction — a logical guess about what will happen next, based on the text (a character promises to call → she will probably call).
  • apply — use a rule or idea from the text in a new situation (use a policy's rules to decide if a new employee qualifies).
  • explicit — stated directly ("The meeting is at 3:00").
  • implicit — suggested but not stated ("Bring an umbrella" implies rain is expected).
  • unsupported — not backed by the text, even if it might be true in real life.
  • legend / key — the part of a graph that tells what each color or symbol stands for.
  • axis — the scale on the side or bottom of a graph that shows what is measured and the units.

All RLA words in the glossary

Inference = clue + reasoning

An inference is like detective work. You notice a clue, add common-sense reasoning, and reach a conclusion that the author did not state.

Formula: Text clue + what I know = inference

Example. "Marcus checked the clock for the fifth time and tapped his foot. The interviewer was twenty minutes late."

  • Clue: checks clock repeatedly, taps foot, interviewer late.
  • What I know: people do these things when nervous or impatient.
  • Inference: Marcus is anxious or impatient.
Inference formula: text clue plus what I know equals inference. Text clue: Marcus checks the clock a fifth time and taps his foot; the interviewer is 20 minutes late. What I know: people act this way when they are nervous or impatient. Inference: Marcus is anxious or impatient. Test: can you point to the exact words that support it?Text clueChecks the clock afifth time, taps hisfoot; the intervieweris 20 minutes late.+What I knowPeople act this waywhen they arenervous orimpatient.=InferenceMarcus is anxiousor impatient.Test: can you point to the exact words that support it?

The test of a good inference: can you point to the exact words that support it? If you need to imagine facts the author never gave, it's not an inference; it's a guess.

Stay close to the text. GED wrong answers are often things that could be true but go too far. If a story says a character "slammed the door," you can infer she is upset. You cannot infer she will move to another city.

Inference levels:

Too weakJust rightToo far
"She closed the door." (just restates)"She was angry.""She hates her family and will never come back."
Inference levels from staying on the page to going beyond the evidence. Too weak: 'She closed the door', which just restates. Just right: 'She was angry.' Too far: 'She hates her family and will never come back.' The story says she slammed the door; pick the middle box.stays on the pagegoes beyond the evidenceToo weak“She closed the door.”(just restates)Just right“She was angry.”Too far“She hates her familyand will never comeback.”The story says she “slammed the door.” Pick the middle box.

Drawing conclusions and making predictions

A conclusion pulls several details together. Ask: What do all these facts point to?

Example. A report says (1) bus ridership rose 20 percent, (2) the city added late-night routes, and (3) riders' top request in a survey was "more evening buses." Conclusion: the new late-night routes likely helped increase ridership. Each detail alone is not enough; together they support the conclusion.

Umbrella organizer. The conclusion on top, 'The new late-night routes likely helped increase ridership', is held up by three details: bus ridership rose 20 percent; the city added late-night routes; riders' top survey request was 'more evening buses.' Each detail alone is not enough; together they hold up the conclusion.CONCLUSIONThe new late-night routes likely helpedincrease ridership.Detail 1Bus ridership rose 20percent.Detail 2The city addedlate-night routes.Detail 3Top survey request:“more evening buses.”Each detail alone is not enough; together they hold up the conclusion.

Predictions use the same skill, pointed toward the future. Good predictions are:

  • Based on clues, often in the last paragraph (promises, plans, a character's habits).
  • Small and logical, the next likely step, not a dramatic new event.
  • Consistent with character. A careful, rule-following character is unlikely to suddenly do something reckless.

Example. "Lena circled the job ad, then looked at her old résumé and sighed. She opened her laptop." Prediction: she will update her résumé and apply. Not supported: she will start her own company.

Applying ideas to new situations. The GED may give you a policy, a rule or a principle and ask how it applies to a case the text doesn't mention. Turn the rule into a checklist, then test the case against every item. One failed condition is enough to change the answer.

Integrating text with charts and graphs

Many GED passages come with a graph, table or chart. Some questions ask you to use both sources together.

Step 1: Read the graphic carefully.

  • Title: what is it about?
  • Axes: what is measured, and in what units (number, percent, dollars)?
  • Legend: which color or symbol stands for which group?
  • Time span: what years or periods are shown?

Step 2: Find the pattern. Is something rising, falling, staying the same? Are two series moving apart or together?

Step 3: Connect it to the text. The text often explains why the graph looks the way it does. For example, a graph shows a line going flat; the text says a site reached its maximum size.

Example.

YearLibrary visitorsComputer sessions
202250,0008,000
202348,00011,000
202447,00015,000

The text says: "The library added 20 new computers in 2023 and began free job-search classes." Conclusion supported by both: even though total visits dipped slightly, computer use grew quickly after new computers and classes were added. Not supported: "People stopped reading books" (the table doesn't measure books).

Here is the table drawn as two bar charts:

Two bar charts from the library table. Library visitors: 2022, 50,000; 2023, 48,000; 2024, 47,000. Computer sessions: 2022, 8,000; 2023, 11,000; 2024, 15,000. A red dashed line marks 2023, when 20 new computers and free job-search classes were added. Visits dipped slightly while computer use grew quickly.Library visitors0k20k40k60k50,000202248,000202347,0002024Computer sessions0k6k12k18k8,000202211,000202315,0002024Red dashed line: 2023, when 20 new computers and free job-search classes were added.Visits dipped slightly; computer use grew quickly.

Watch out: a "both sources" answer must be supported by the text and the graphic. An answer supported by only one is a trap.

Worked example

Worked example 1: Inference about a character

Ms. Patel read the email twice. Then she folded her reading glasses, set them on the desk, and stared out the window for a long time. When her assistant knocked, she said, "Not now, Jen," without turning around.

Question: Which inference about Ms. Patel is best supported? A. She is excited about a promotion. B. She received upsetting news. C. She dislikes her assistant. D. She is about to retire.

Think-aloud:

  1. Clues: reads the email twice (surprise or shock), stares out the window a long time (deep thought), says "Not now" without turning (wants to be alone).
  2. These are common signs of someone absorbing bad or troubling news. That is B.
  3. Eliminate A: excitement would look different (smiling, calling someone). C: "Not now" shows she's preoccupied, not that she dislikes Jen; one short reply is weak evidence. D: nothing mentions retirement. That goes too far.

Answer: B.

Worked example

Worked example 2: Applying a policy

Overtime Policy. Hourly employees who work more than 40 hours in a week are paid 1.5 times their regular rate for each hour over 40. Overtime must be approved in advance by a supervisor. Salaried employees are not eligible.

Question: Which employee should receive overtime pay? A. Luis, salaried, worked 46 hours. B. Tasha, hourly, worked 44 hours with her supervisor's advance approval. C. Omar, hourly, worked 39 hours. D. Beth, hourly, worked 45 hours without approval.

Think-aloud:

  1. Turn the policy into a checklist: (1) hourly? (2) more than 40 hours? (3) approved in advance?
  2. Luis: not hourly → fails (1). Omar: 39 hours → fails (2). Beth: no approval → fails (3).
  3. Tasha: hourly ✓, 44 hours ✓, approved ✓.
Overtime policy checklist with three conditions: hourly, more than 40 hours, approved in advance. Luis: no, salaried; 46 hours; approval not stated; no overtime. Tasha: hourly yes; 44 hours yes; approved yes; overtime yes, 4 hours. Omar: hourly yes; 39 hours, no; approval not stated; no overtime. Beth: hourly yes; 45 hours yes; approved no; no overtime. One failed condition is enough: only Tasha passes all three.Employee1. Hourly?2. Over 40 hours?3. Approved ahead?Overtime?LuisNo, salariedYes, 46not statedNOTashaYesYes, 44YesYES: 4 hoursOmarYesNo, 39not statedNOBethYesYes, 45NoNOOne failed condition is enough: only Tasha passes all three.

Answer: B. She would get overtime pay for 4 hours (44 − 40).

Worked example

Worked example 3: Text plus graph

A city added protected bike lanes on Oak Avenue in 2023. The table shows average daily bike trips on Oak Avenue.

Year2021202220232024
Daily bike trips310320540690

Question: Which conclusion is best supported by the text and the table? A. Bike trips fell after the lanes were built. B. Bike trips rose sharply after the lanes were added. C. Car traffic on Oak Avenue dropped by half. D. Most residents now bike to work.

Think-aloud:

  1. Read the table: 310 → 320 (small change before the lanes), then 540 and 690 after 2023 (big jumps).
  2. Connect to text: the lanes were added in 2023, right where the jump begins. So B is supported by both.
  3. Eliminate A (opposite of the data). C: car traffic isn't measured. D: "most residents" goes far beyond 690 trips on one street.
Bar chart of average daily bike trips on Oak Avenue: 2021, 310; 2022, 320; 2023, 540; 2024, 690. A red dashed line before the 2023 bar marks when protected bike lanes were added. There is a small change before the lanes, an increase of 10, and big jumps after, to 540 and then 690.Average daily bike trips on Oak Avenue02004006008003102021320202254020236902024Protected lanes added in 2023trips per daySmall change before the lanes (+10); big jumps after (540, then 690).

Answer: B.

Worked example

Worked example 4: Prediction

Every Saturday for three months, Andre had walked past the community theater's audition sign. This week he stopped, read it twice, and took a photo of the date with his phone. That night, he practiced a monologue in front of the bathroom mirror until his roommate pounded on the door.

Question: What is Andre most likely to do next?

Think-aloud:

  1. Clues build in steps: walks past for months (interest), stops and photographs the date (planning), practices that night (preparing).
  2. Each clue moves him closer to one action: going to the audition.
  3. Possible traps: "He will become a famous actor" (too far), "He will move out because of his roommate" (one knock on the door doesn't support that), "He will ignore the sign again" (contradicts his new actions).
Staircase of clues leading to a prediction. Step 1, for three months: Andre walks past the audition sign, showing interest. Step 2, this week: he stops, reads it twice and photographs the date, showing planning. Step 3, that night: he practices a monologue at the mirror, preparing. Top: prediction, he will most likely go to the audition. Too far: 'He will become a famous actor.'For threemonthswalks past theaudition sign(interest)This weekstops, reads ittwice, photographsthe date(planning)That nightpractices amonologue at themirror (preparing)PREDICTIONHe will mostlikely go to theaudition.Too far:“He will become afamous actor.”

Answer: He will most likely go to the audition.

Common mistake

Common traps

  • Going too far. Wrong: "The garden will close because it is full." Right: Being full limits growth; nothing says it will close. Choose the smallest conclusion the evidence fully supports.
  • Taking characters literally. Wrong: A bus driver tells dispatch he "had a mechanical issue," so "the bus really broke down." Right: Compare words with actions. If he was watching the street for a late rider, not fixing the engine, his words may be an excuse.
  • Using outside knowledge instead of the text. Wrong: "Fees are too high, as they usually are." Right: Only the passage counts. If it isn't there, it isn't supported.
  • Checking only one condition. Wrong: "Mei works 40 hours a week, so she qualifies." Right: Test every rule. If the policy also requires 6 months on the job and Mei was hired 4 months ago, she does not qualify.
  • Misreading the graph. Wrong: mixing up the two colors or the units. Right: Read the legend and axes first.
  • Choosing an answer supported by only one source when the question says "both the passage and the graph."
Test strategy

Test-day strategy

  1. Read the question first. Words like infer, suggest, most likely, best supported signal an inference question.
  2. Find the evidence before looking at the choices. Locate the lines (or bars) the question points to, and form your own answer.
  3. Use the "prove it" test. For each choice, ask: Which words or numbers prove this? No proof → eliminate.
  4. Beware of extreme words: all, never, always, completely, most people. Inferences on the GED are usually moderate.
  5. For policies, make a checklist on your noteboard and test each person or case against every condition.
  6. For graphics, read title, axes, units and legend before the data. Write down key numbers you need.
  7. For predictions, look at the last paragraph and the character's habits. Pick the next logical step, not a dramatic twist.
  8. Manage time: if two choices seem close, pick the one closer to the exact words of the text, flag the question, and move on.
Check yourself

Check your understanding

Inferences and conclusions
Riverside Community Garden: Growing Demand

(1) When the Riverside Community Garden opened in 2021 on a former parking lot behind the public library, organizers worried that no one would sign up. They had 40 raised-bed plots, each about the size of a parking space, and they rented them to local households for $25 a season.

(2) Those worries did not last. Every plot was claimed in the first month, and by the end of the season, a dozen households were on a waiting list. The garden added 8 plots in 2022 by using a strip of lawn next to the building. In 2023, the city allowed the garden to take over the rest of the old lot, bringing the total to 60 plots. That is as many as the site can hold.

(3) Gardeners say the benefits go beyond fresh tomatoes. Many plot holders live in apartments with no yard. Retirees mentor first-time growers, and a local elementary school uses two plots for science lessons. Last season, gardeners donated more than 900 pounds of produce to the food pantry down the street.

(4) The garden's volunteer board is now considering two proposals. The first would lease a vacant lot on Elm Street, about half a mile away, which could hold roughly 30 new plots. The second would limit each household to two seasons in a row so that people on the waiting list get a turn. Some longtime gardeners oppose the second idea. "It takes three years to really learn your soil," one said at a recent meeting.

(5) The graph below shows how many plots were in use and how many households were waiting for a plot at the end of each season.

Grouped bar chart titled 'Riverside Community Garden: Plots in Use and Waiting List, 2021-2025'. Vertical axis: number (0 to 80). For each year, the blue bar is plots in use and the red bar is households on the waiting list. 2021: 40 plots in use, 12 households waiting; 2022: 48 plots in use, 25 households waiting; 2023: 60 plots in use, 41 households waiting; 2024: 60 plots in use, 58 households waiting; 2025: 60 plots in use, 73 households waiting. Riverside Community Garden: Plots in Use and Waiting List 0 20 40 60 80 Number 40 12 2021 48 25 2022 60 41 2023 60 58 2024 60 73 2025 Plots in use Households on waiting list

Graph data: 2021, 40 plots in use and 12 households waiting; 2022, 48 and 25; 2023, 60 and 41; 2024, 60 and 58; 2025, 60 and 73.

Based on the graph, which statement best describes what happened between 2023 and 2025?

Answer choices
Hint

Look at the blue bars and the red bars separately for 2023, 2024 and 2025.

Check yourself

Check your understanding

Inferences and conclusions
Riverside Community Garden: Growing Demand

(1) When the Riverside Community Garden opened in 2021 on a former parking lot behind the public library, organizers worried that no one would sign up. They had 40 raised-bed plots, each about the size of a parking space, and they rented them to local households for $25 a season.

(2) Those worries did not last. Every plot was claimed in the first month, and by the end of the season, a dozen households were on a waiting list. The garden added 8 plots in 2022 by using a strip of lawn next to the building. In 2023, the city allowed the garden to take over the rest of the old lot, bringing the total to 60 plots. That is as many as the site can hold.

(3) Gardeners say the benefits go beyond fresh tomatoes. Many plot holders live in apartments with no yard. Retirees mentor first-time growers, and a local elementary school uses two plots for science lessons. Last season, gardeners donated more than 900 pounds of produce to the food pantry down the street.

(4) The garden's volunteer board is now considering two proposals. The first would lease a vacant lot on Elm Street, about half a mile away, which could hold roughly 30 new plots. The second would limit each household to two seasons in a row so that people on the waiting list get a turn. Some longtime gardeners oppose the second idea. "It takes three years to really learn your soil," one said at a recent meeting.

(5) The graph below shows how many plots were in use and how many households were waiting for a plot at the end of each season.

Grouped bar chart titled 'Riverside Community Garden: Plots in Use and Waiting List, 2021-2025'. Vertical axis: number (0 to 80). For each year, the blue bar is plots in use and the red bar is households on the waiting list. 2021: 40 plots in use, 12 households waiting; 2022: 48 plots in use, 25 households waiting; 2023: 60 plots in use, 41 households waiting; 2024: 60 plots in use, 58 households waiting; 2025: 60 plots in use, 73 households waiting. Riverside Community Garden: Plots in Use and Waiting List 0 20 40 60 80 Number 40 12 2021 48 25 2022 60 41 2023 60 58 2024 60 73 2025 Plots in use Households on waiting list

Graph data: 2021, 40 plots in use and 12 households waiting; 2022, 48 and 25; 2023, 60 and 41; 2024, 60 and 58; 2025, 60 and 73.

Which conclusion is best supported by both the passage and the graph?

Answer choices
Hint

Find one fact in the text and one pattern in the graph that point to the same conclusion.

Summary

  • An inference = text clue + reasoning. You must be able to point to the evidence.
  • Conclusions combine several details; predictions extend clues to the likely next step.
  • To apply a policy or idea, turn it into a checklist and test every condition.
  • With graphics, read title, axes, units and legend; then connect the pattern to the text.
  • "Both sources" answers need support from the text and the graphic.
  • Avoid answers that go too far, rely on outside knowledge, or take characters' words at face value.
Practice this skill →