Do Idioms Make You Sound Fluent in English? The Evidence
By The Dobi Apps Team9 min read
Set phrases help English learners sound fluent, but colorful idioms are only a small part of that. Corpus work puts ready-made word combinations at about 55% of English, and a 38-learner study found learners who used more of them got higher oral proficiency ratings. Idioms matter mostly for understanding native speakers, so learn frequent ones first.

Do Idioms Make You Sound More Fluent in English?
Ready-made phrases make English learners sound more fluent, but colorful idioms are a small slice of those phrases. The evidence for fluency gains comes from studies of formulaic language as a whole: collocations, fixed expressions, and chunks like "as far as I know". Idioms such as "spill the beans" matter more for understanding native speakers than for sounding fluent yourself.
The clearest link between set phrases and how fluent a learner sounds comes from a 2022 peer-reviewed study by Marco Cancino and Jonathan Iturrieta at Andres Bello University in Chile. They split 38 adult EFL learners into two groups of 19, gave one group 38 hours of teaching built around lexical chunks, kept the other on regular lessons, and had judges rate both groups' speaking. The chunk group scored higher on every measure, and across all learners, the more formulaic sequences a learner used, the higher the judges rated them.
"These findings indicate that mastering ready-made structures when learning a new language will likely increase perceived oral proficiency."
Note the word "perceived". These studies measure how fluent listeners judge a speaker to be, which is what most learners care about, but they are small classroom studies with dozens of learners, not large trials. None of them isolates idioms. In the Chilean study's direction-giving task, the formulaic sequences were ordinary combinations like "turn left" and "go straight ahead", nothing like "it's raining cats and dogs".
What Is the Difference Between Idioms, Collocations, and Lexical Chunks?
An idiom is a phrase whose meaning cannot be worked out from its words, like "pull someone's leg". A collocation is a pair of words that simply go together by habit, like "prominent role". Lexical chunks, also called formulaic sequences or prefabs, is the umbrella term that covers both, plus fixed expressions and phrasal verbs.
The linguist Joan Bybee, writing in the peer-reviewed journal Language in 2006, draws the line this way: idioms "have extended meaning (usually of a metaphorical nature)", while prefabs "are word sequences that are conventionalized, but predictable in other ways", with examples such as "mixed message", "beyond repair", and phrasal verbs like "finish up".
| Type | Example | Meaning from the words? | What it does for a learner |
|---|---|---|---|
| Idiom | spill the beans | No, figurative | Comprehension of native speech, color in casual talk |
| Collocation | prominent role, mixed message | Yes, but the pairing is fixed by habit | Sounds natural instead of "correct but odd" |
| Phrasal verb or verb-preposition pair | finish up, interested in | Partly | Everyday speech, very frequent |
| Conversational formula | as far as I know, long story short | Mostly | Buys time and links ideas while speaking |
The distinction matters because it changes what "learn idioms to sound fluent" should mean. Most of the phrases that make English sound native are collocations, the area where learners also mix up near-synonyms, covered in why English learners confuse similar words. Idioms sit on top of that layer. "Do a mistake" and "make a mistake" are both grammatical, but only the second is what English speakers say, and no idiom list will teach that.
How Much of Everyday English Is Made of Set Phrases?
About half. A corpus study by Britt Erman and Beatrice Warren, published in the journal Text in 2000, counted conventionalized word combinations in samples of spoken and written English. Joan Bybee summarizes the result in Language (2006): prefabricated word combinations "constitute about 55% of both spoken and written discourse".
About 55%
That figure needs two caveats. First, it was counted with Erman and Warren's own definition of a prefab, based on which combinations native speakers prefer, so a different definition gives a different number. The 55% here is a secondary report of their result, not a figure read from the original paper. Second, the 55% is mostly not idioms. Bybee's examples of prefabs are plain phrases like "to need help" and "interested in", which no learner would put on an idiom flashcard.
What the figure does show is why grammar plus single words is not enough. Speakers "recognize prefabs as familiar", Bybee writes, which is evidence those sequences are stored in memory as units. A learner who builds every sentence word by word produces English that is grammatical but often not the way anyone says it. Knowing roughly how many English words you need is one part of vocabulary size. Knowing which words travel together is the other, and it is the part that listeners notice.
What Does Research Say About Lexical Chunks and Speaking Fluency?
Small classroom studies point the same way, some more weakly than their abstracts suggest. In Cancino and Iturrieta's 2022 study, judges rated chunk-taught adult learners higher on fluency, accuracy, and range. A 2022 study of Saudi teenagers showed gains within the taught group, but its comparison with a control group was not statistically significant.
In Cancino and Iturrieta's study, published in the Vigo International Journal of Applied Linguistics, the chunk-taught group's median scores on the specific measures ranged from 9.75 to 11.25, against 3.75 to 4.5 in the control group. The count of formulaic sequences each learner used correlated with fluency ratings at 0.497 (Spearman's rho), a moderate correlation. The same paper describes the study it follows, Boers and colleagues' 2006 experiment, in which 22 hours of phrase-noticing work also produced higher judged oral proficiency.
The 2022 study by Hani Hamad M. Albelihi in Frontiers in Psychology is weaker evidence. It taught lexical chunks to 30 intermediate Saudi EFL learners aged 12 to 17 (its abstract says 13 to 17) over 18 sessions and measured fluency as words per T-unit (one main clause plus its attached clauses) in an interview.
| Group | Pretest mean | Posttest mean |
|---|---|---|
| Lexical chunk group (n = 30) | 9.98 | 12.02 |
| Control group (n = 30), posttest only reported | 10.05 |
The chunk group improved significantly from pretest to posttest, but the posttest difference between groups was reported as p = 0.397, not significant. The author lists small groups, an all-male sample, and uncontrolled teaching variables as limitations, so this study supports "chunk teaching did no harm and may help" more than it proves a fluency gain.
Why Are English Idioms So Hard for Learners?
English idioms are hard mainly because their meaning cannot be worked out from the words. In a 2026 survey of 60 fourth-year English majors at the University of Baghdad, 53% agreed idioms are difficult because the individual words mean something different from the whole phrase, the top difficulty of three surveyed. Missing first-language equivalents came next.
The survey, by Yasamine Falih Ahmed in the Journal of the College of Basic Education, asked students to rate statements on a five-point scale, so the results are self-reported attitudes rather than test scores.
| Reason idioms are difficult (Baghdad survey, n = 60) | Agree or strongly agree | Neutral |
|---|---|---|
| The meaning of the individual words differs from the whole phrase | 53% | 32% |
| The idiom does not exist in my first language | 43% | 32% |
| Idioms are not well taught in English class | 41.6% | 35% |
The same students still valued idioms: 75% agreed it is important to encourage EFL students to learn idioms for effective communication. The gap between wanting idioms and being able to use them has a name. Sulaiman Alnujaidi's 2025 study in the Journal of Language and Education describes it: "second-language learners tend to avoid them or use them inaccurately", a pattern the literature calls the "idiomatic paradox".
A learner who meets "break the ice" can picture literal ice and get no closer to the meaning. The phrase has to be learned whole, which is why the break the ice page on the Dobi Apps idioms guide gives a plain-English meaning and example sentences rather than a word-by-word gloss.
Which English Idioms Are Worth Learning First?
Learn the idioms that turn up most in everyday speech, and skip rare or dated ones until you meet them. The best-known list is Dilin Liu's 2003 corpus study in TESOL Quarterly, which searched about 6 million words of spoken American English and found 302 idioms frequent enough to rank, sorted into three frequency bands.
According to Yuji Tanaka's 2012 review of that study, Liu's corpora covered professional meetings, academic speech from the Michigan Corpus of Academic Spoken English, and unscripted talk from US TV news, debate, and talk shows. The bands were 50 or more, 20 to 49, and 2 to 19 occurrences per million words. Liu himself called the bands "rather arbitrary" and "intended merely as a reference, not a guide". The lists also converged: 283 of the 302 idioms appeared in all three corpora, which suggests a core of idioms that cross settings.
Three practical filters follow from this:
- Frequency over color. An idiom you will hear every week, such as on the same page, is worth more than a vivid one you will hear once a year.
- Register. Check whether an idiom is informal or work-safe before using it. The communication idioms and work and business idioms topic pages carry a usage note on each entry.
- Meaning group. Learning idioms by topic lets near-synonyms like let the cat out of the bag and spill the beans reinforce each other.
The full English idioms guide groups its entries into 13 topics, from feelings to money.
What Is the Best Way to Learn English Idioms?
Learning idioms through combined pictures, audio, video, and interactive practice worked best in the one study that compared methods head to head. Sulaiman Alnujaidi's 2025 study of 66 Saudi university students found that a multimodal group outscored both a chunk-memorization group and an idiom-origin group, with large effect sizes, after ten weeks and 40 idioms.

All three groups learned the same 40 idioms, two per session, and took the same 40-item test before and after.
| Method (Alnujaidi, 2025) | Students | Pretest mean (of 40) | Posttest mean (of 40) |
|---|---|---|---|
| Lexical: idiom taught as a whole chunk with examples | 23 | 15.34 | 20.13 |
| Etymological: idiom taught with its historical origin | 22 | 14.04 | 21.22 |
| Multimodal: text, pictures, audio, video, online quizzes | 21 | 13.61 | 25.14 |
The multimodal group beat the lexical group (Cohen's d = 1.23) and the etymological group (d = 1.10). The lexical and etymological groups did not differ significantly. The author is candid that 25.14 is only 62.5% of the test, that there was no delayed posttest, and that the sample was one all-male group of intact classes.
Origin stories gave no clear edge here, and the same paper's review notes that earlier results on etymology are mixed: Boers (2001) found it improved retention, while other studies found little benefit. Origins are also often uncertain. The study's handout traced spill the beans to an ancient Greek voting practice and said it "likely" came from there; the Dobi Apps idiom pages only give an origin when it is well attested.
How Do You Start Using Idioms Without Sounding Unnatural?
Start by understanding idioms before using them, then use the few you know well in the right setting. An idiom used slightly wrong stands out more than a plain phrase, and in Alnujaidi's 2025 classroom study even the best group answered only 62.5% of idiom questions correctly after ten weeks. Comprehension comes first; production follows.
Meeting an idiom in context is what makes it usable. Stories and audiobooks put idioms in the situations where they are actually said, which is the case for learning English through stories more generally. Once a phrase is familiar, saying it aloud along with a narrator, the technique covered in shadowing audiobooks for English pronunciation, trains the rhythm of the whole chunk rather than word by word.
A short routine that fits the evidence:
- Pick a few frequent idioms a week. The 2025 study taught four a week, and even its best group missed over a third of the final test.
- Read the meaning, usage note, and examples on a page such as beat around the bush.
- Review them on a schedule, the method explained in how spaced repetition builds English vocabulary retention.
- Use each one once in speech or writing, in a setting where its register fits.
Do not ration collocations the same way. Phrases like "make a decision" or "interested in" are the 55% of English that makes speech sound native, and they deserve more study time than any idiom list.
Frequently asked questions
Is it bad to use idioms in an English exam or job interview?
How many idioms should I learn at a time?
Are phrasal verbs idioms?
Frequently asked questions
Is it bad to use idioms in an English exam or job interview?
How many idioms should I learn at a time?
Are phrasal verbs idioms?
Sources
- Assessing the impact of the Lexical Approach on EFL perceived oral proficiency: What is the role of formulaic sequences?, Marco Cancino and Jonathan Iturrieta
- Intermediate English as a Foreign Language learners' formulaic language speaking proficiency: Where does the teaching of lexical chunks figure?, Hani Hamad M. Albelihi
- Enhancing EFL Students' Idiomatic Competence: A Comparative Analysis of Lexical, Etymological, and Multimodal Approaches, Sulaiman Alnujaidi
- Exploring the Challenges and Strategies in Mastering English Idioms Among Baghdad University Students, Yasamine Falih Ahmed
- From Usage to Grammar: The Mind's Response to Repetition, Joan Bybee
- Second Language Vocabulary Research: 2003, Yuji Tanaka
Published
The Dobi Apps Team
Editorial
The team behind Dobi Apps, makers of English-learning apps like Audiobook and Storybook, posting release notes, learning tips, and app updates.
Related posts
How Many English Words Do You Need to Read and Speak?
Nation's corpus study puts unaided reading at 8,000 to 9,000 word families and speech at 6,000 to 7,000. What that means per CEFR level, and how to get there.
Why English Learners Confuse Similar Words and How to Fix It
Learners mix up affect and effect, make and do, or borrow and lend because of first-language transfer, one-to-many translations, and look-alike word forms.
How to Learn English Effectively Through Stories
Learning English through stories works through extensive reading: meeting words again in real context. How to pick the right level, plus 5 apps to try.