Opinion vs fact, and predicting what comes next

By B1, you can read sentences and follow stories. The next layer of reading skill is critical reading — reading like a detective, not like a sponge.

This lesson covers two related skills:

1. Telling fact from opinion — and spotting opinion that’s been disguised as fact.
2. Predicting what comes next — using text structure to anticipate the writer’s next move, which makes you a much faster reader.

Both skills protect you from being manipulated by writing that wants you to believe something.

Fact vs opinion — the basic difference

A fact is something that can be verified — checked against the world. Even if it turns out to be wrong, it has the shape of a fact: it makes a specific, checkable claim.

An opinion is a personal judgment, evaluation, or feeling. It depends on the writer’s perspective.

Compare:

• Fact: The film made 280 million dollars at the US box office in its opening weekend.
• Opinion: The film is a worthy successor to the original.

The first can be checked against box office data. The second is a judgment — worthy is a value word.

Markers of fact

Sentences that present facts often include:

• Specific numbers, dates, percentages. 15 percent, 2024, 47 states.
• Named sources. According to the Bureau of Labor Statistics… A 2025 Pew Research study found…
• Attributable claims. The mayor said… The report concluded…
• Verifiable specifics. Born in Chicago in 1972, she moved to Boston after college.
• Past-tense reporting verbs. announced, confirmed, reported, found, showed.

Even when a fact turns out to be incorrect, the form is factual: it points to the world and invites verification.

Markers of opinion

Sentences that present opinions often include:

• First-person markers. I think, I feel, in my view, personally, I’d argue.
• Hedges of belief. It seems, it appears, arguably, possibly.
• Modal verbs of obligation. should, must, ought to, need to (when applied to behavior or policy).
• Evaluative adjectives. great, terrible, brilliant, disappointing, beautiful, ugly, important, trivial.
• Generalizations. Everyone knows, most people agree, it’s clear that.
• Unsupported claims. A statement with no evidence and no source attached.

Tip: when an article uses I or we, you’re almost certainly reading opinion writing. American newspapers usually keep first-person out of news reporting.

The tricky case: opinion presented as fact

The most dangerous reading trap at B1 is opinion that wears the costume of fact. Some examples:

• Everyone knows that streaming has destroyed the movie industry. — Everyone knows is not evidence; it’s just rhetoric. The claim itself (streaming destroyed the movie industry) is contested.
• It is well established that homework improves academic outcomes. — Well established is doing a lot of work; the actual research is mixed.
• Most experts agree that the new policy will fail. — Most experts? Which experts? This is opinion in factual clothing.
• Naturally, the company chose profits over employees. — Naturally signals the writer’s attitude, not a fact.

When you see vague crowd-claims (everyone, most people, all experts), suspect opinion. Real facts come with specific sources. A 2024 Brookings Institution report found that 62 percent of… is a fact-shaped claim. Most experts agree is not.

Bias and source stake — who’s writing, and why?

Even a true fact can be selected, framed, and presented in a biased way. To read critically, ask:

1. Who wrote this? A journalist? A company press release? A think-tank policy paper? A blogger?
2. What’s their stake in the topic? A car company writing about EV regulations has a different stake than an environmental group writing about the same regulations. Both can be technically accurate but emphasize different facts.
3. What facts are missing? A piece praising a policy might leave out the costs. A piece criticizing it might leave out the benefits.
4. What’s the publication? The New York Times news section, The Wall Street Journal opinion section, a Reddit comment, a corporate blog — each comes with a different default lens.

You don’t have to distrust everything. You just have to know who’s talking.

News section vs opinion section

A specific American convention worth knowing: US newspapers strictly separate news from opinion.

• News articles (the front page, the metro section, business news) are supposed to be factual reporting. They name sources, attribute claims, and avoid first-person.
• Op-eds, editorials, and columns (usually labeled Opinion at the top) are explicitly arguments. They use I, take sides, and try to persuade.

A B1 reader should always check what kind of article they’re reading before deciding how much of it to take as fact. The same newspaper can publish a news report saying the law was passed and an op-ed saying the law is a disaster on the same day. Both are doing their jobs.

Predicting what comes next — text structure as a tool

Good writing is structured. Once you learn the common structures, you can predict what’s about to happen on the page. This makes you faster and helps comprehension.

Common signal phrases and what they predict

When you see one of these openers, you can almost always anticipate the next move:

• On the one hand, … → expect on the other hand, … (a contrast is coming).
• First, … / The first reason is … → expect second, third, or a list.
• There are three main reasons … → expect three numbered or signaled reasons.
• However, … / But … → the writer is about to challenge or reverse the previous claim.
• For example, … / For instance, … → an illustration of the previous abstract claim.
• In other words, … → a restatement of the previous sentence in clearer form.
• As a result, … / Therefore, … / This means … → a consequence of the previous claim.
• Despite this, … / Nevertheless, … → contrast — the writer concedes something but holds their position.
• In conclusion, … / To sum up, … → the writer is closing.
• Critics argue that … / Some have suggested that … → expect however, then a counter-argument from the writer.

When you see Critics argue that the plan is too expensive, you can predict almost word-for-word: the next sentence will be something like However, supporters point out… or In response, the administration says…. The writer is setting up a counter so they can knock it down.

Paragraph-level prediction

Inside a single paragraph:

• A topic sentence at the start predicts what supporting details are coming.
• A comparison phrase (unlike X, Y is…) predicts the rest of the paragraph will contrast X and Y.
• A chronological cue (it began in 1995, …) predicts a timeline of events.

Reading with prediction is reading actively. You’re not just absorbing words — you’re constantly asking what’s the writer going to do next? and checking your guess.

Worked example — a product review paragraph

Read this short review paragraph (the sort of thing you’d find on a US tech site):

The Vortex 14 laptop weighs in at 3.2 pounds and ships with a 14-inch OLED display, 16 GB of RAM, and a 1 TB SSD. Battery life clocked in at 11 hours of mixed use during our testing, beating the manufacturer’s claim of 10 hours. At 1,499 dollars, however, it costs 300 dollars more than last year’s model with only modest hardware upgrades. For students or general users, that premium is hard to justify; for creative professionals who need the OLED panel and extra storage, it’s still one of the better mid-range options on the market.

Let’s label every claim as fact or opinion and notice the bias.

Fact claims (verifiable specifics)

• Weighs 3.2 pounds. — Fact, checkable on a scale.
• 14-inch OLED display, 16 GB of RAM, 1 TB SSD. — Fact, on the spec sheet.
• Battery life clocked in at 11 hours of mixed use during our testing. — Fact, but tied to the reviewer’s specific testing methodology.
• Beating the manufacturer’s claim of 10 hours. — Fact, comparing two specific numbers.
• Costs 1,499 dollars. — Fact.
• 300 dollars more than last year’s model. — Fact, two prices subtracted.

Opinion claims (judgments)

• Modest hardware upgrades. — Opinion. Modest is an evaluation; another reviewer might call them significant.
• That premium is hard to justify (for students or general users). — Opinion. Hard to justify is a value judgment.
• Still one of the better mid-range options on the market. — Opinion. Better is comparative judgment, not a specific spec.

Mixed / fact-shaped opinion

Notice the phrase for students or general users, that premium is hard to justify. The reviewer is using a confident tone, but hard to justify is not a fact about the laptop — it’s a verdict. They’re packaging an opinion in factual clothing.

Source bias

Who wrote this? A tech reviewer. What’s the stake? Tech reviewers generally test loaner units provided by manufacturers, often need access for future reviews, and depend on advertising on the same site. A truly hostile review is rare. The hedged structure (here are the specs + but it’s expensive + still a good option for some) is the standard balanced-but-friendly tech-review format.

That doesn’t mean the review is wrong. It means you should know what kind of writing this is — a tech-press review, not a consumer-protection report.

Practice approach — how to drill this skill

Daily fact/opinion sort (10 minutes):

1. Pick any article — news, opinion, or review.
2. Take a paragraph. Number each sentence.
3. Label each: F (fact), O (opinion), or F/O (mixed — fact-sounding opinion).
4. For each opinion, find the value words that gave it away.

Bias hunt (weekly):

Read two articles on the same news event from different outlets. Don’t read for content — read for what’s emphasized and what’s left out. Where do they agree on facts? Where do they differ in framing?

Prediction drill (5 minutes):

Pick any article. Read the first sentence of a paragraph, then stop and predict what the rest of the paragraph will say. Read on and check. Over time you’ll get scarily accurate.

Signal-word notebook:

Keep a running list of the signal words you spot in your reading (however, on the other hand, critics argue, in conclusion). Note what each one predicts. After two weeks you’ll start spotting them automatically.

Common Russian-speaker mistakes

1. Trusting any claim that sounds confident. Russian academic and journalistic writing often uses confident, declarative tone for opinion. Russian-speaking B1 readers can carry that habit into English and accept it is well known that… as fact. Always ask for the source.
2. Missing hedges that signal opinion. It seems, arguably, possibly, in my view are flags that what follows is opinion. Don’t read past them — they change the status of the claim.
3. Treating opinion sections as news. US newspaper opinion pages look similar in layout to news pages. Always check the word Opinion or Editorial at the top.
4. Equal weight to all sources. A paid blog post and a peer-reviewed study are not the same. Source matters.
5. Skipping the structural signals. On the one hand, however, in conclusion are gold. Russian learners sometimes treat them as filler. They’re navigation tools — use them.

Summary

• Facts are verifiable; opinions are judgments. Look for specific numbers, named sources, and attributable claims for facts.
• Beware opinion in factual clothing: everyone knows, it is well established, most experts agree — these phrases hide opinion.
• Check the source and its stake. Who wrote this? What’s their angle?
• News vs opinion section is a strict line in US newspapers — know which you’re reading.
• Predict what comes next using signal phrases — it makes you faster and more comprehending.

Next lesson: Common text types — news, reviews, blogs, and instructions.