Social Work & Human Services · Communication
Active Listening
On this page 9 sections
In 30 seconds
Active listening is a set of trainable behaviors, not a personality trait. You attend, you use minimal encouragers, you Paraphrase A condensed restatement of the content someone just gave you, in your own words and offered tentatively so they can correct it. Full entry → content, you reflect feeling, you summarize, and you ask open questions instead of stacking closed ones. Thomas Gordon cataloged the responses that stop people talking: advising, reassuring, warning, moralizing, and eight more. The hardest habit to break is answering a feeling before the person has finished having it.
Why this matters
Every assessment, safety plan, and referral you write rests on what somebody chose to tell you. People stop telling you things for reasons that usually trace back to a single response: advice offered too early, reassurance that contradicted them, a run of closed questions that turned a conversation into a form. Learning to catch those moves and replace them is the difference between a record built from what a person said and a record built from what you assumed. Field instructors assess these skills directly, often from a recording or transcript. The research picture matters too, because trained listening behaviors are measurably changeable while several benefits routinely claimed for them are not established.
The college version
A behavior set, not a temperament
The useful claim behind active listening is narrow and testable: whether a person feels heard depends partly on things the listener visibly does, and those things can be taught, rehearsed, and rated. Carl Rogers and Richard Farson published a short work titled Active Listening in 1957 through the University of Chicago Industrial Relations Center, applying counseling method to ordinary workplace conversation and insisting that a listener attend to emotional undertone as well as words. Thomas Gordon's organization gives a complementary account: Gordon learned to reflect clients' meanings and feelings at the University of Chicago Counseling Center in 1946-47, and adopted the name after Farson, a former student of his, suggested it. Communication research separates hearing, the physiological reception of sound, from listening, the cognitive work of receiving, interpreting, recalling, evaluating, and responding. Reception draws on both the auditory and visual channels, which is why a phone call is not simply a quieter version of an office interview. This topic covers the moves themselves. The relationship those moves are meant to build is the subject of the helping relationships lesson, and the specialized use of reflection to guide someone through ambivalence belongs to motivational interviewing.
Attending, without inventing rules
Attending The visible, physical side of listening — body orientation, gaze, stillness, note-taking choices — calibrated to the person rather than applied as a fixed posture. Full entry → is the part of listening a camera would catch: how you orient your body, what you do with your hands and your phone, whether you take notes and how, whether you let a silence sit. It matters because it is the first evidence the other person has about whether talking to you is worth the trouble. What attending is not is a fixed posture. Introductory checklists that instruct students to maintain eye contact are wrong as universals, and confidently wrong. Communication texts describe high-context settings, where silence and nonverbal signal carry much of the message, and low-context settings, where meaning is expected to be stated outright; a listener who imports one set of expectations into the other reads as cold, or as pushy. Individual variation runs just as wide. Looking away often means a person is thinking rather than disengaging. Autistic people, people in acute distress, and people describing something shameful may all find a steady gaze harder to bear than helpful. The workable version of the skill is calibration: begin at a moderate setting, watch what happens, adjust, and when it seems to matter, ask. "I can take notes or I can just listen, whichever is easier for you" is better practice than any rule about eye contact, because it hands the person a decision they are qualified to make.
The micro-skills
Minimal encouragers, sometimes called back-channel cues, are the small signals delivered while the other person is still talking: mm-hm, right, oh, a nod, a lean forward. Their job is to keep the floor with the speaker. Delivered automatically they become Pseudo-listening Producing the outward signals of attention while not processing the message; observers usually detect it, and it costs the listener credibility. Full entry →, which people generally notice and resent. A paraphrase restates the content of what someone said, condensed, in your own words, and offered tentatively enough that they can correct it. A Reflection of feeling A statement naming the emotion you believe you heard, whether or not the speaker labeled it themselves, offered for them to confirm or revise. Full entry → names the emotion you think you heard, whether or not the person named it. The two often ride together, and both can miss: overshoot and the person backs away, undershoot and they conclude you did not get it. Missing is recoverable, because the correction is information. A summary gathers several minutes of talk into a few sentences. It closes one section of an interview and opens the next, and it exposes gaps you would otherwise carry silently into the record. Questions come in two shapes. A closed question fixes one variable, and there are moments that need one: which agency, what date, whether there is a firearm in the house. An Open question A prompt whose form lets the speaker choose what to say and how much, rather than fixing a single variable such as a date or a yes-or-no fact. Full entry → invites the person to choose what matters. The 2018 Cochrane review of communication skills training found that use of open questions is one of the few behaviors that training reliably shifts. Silence is a skill, not an absence of one. Speakers produce roughly 125 to 175 words per minute while a listener can process 400 to 800, and that surplus is what fills a pause with your next clever question. Letting three seconds pass after a hard sentence often produces the sentence that actually mattered.
A reflection is not a question
Students blur these constantly, and the blur is worth taking seriously because the two moves distribute work differently. "So you feel like nobody at the school listens to you?" is an interrogative: it hands the sentence back and asks the person to certify it. "Nobody at that school has actually listened to you" is a statement: you have taken a position, and you carry the risk of being wrong. The second invites elaboration; the first invites yes or no. Watch your intonation, because a reflection delivered with a rising tone converts itself into a question on the way out of your mouth. The practical cost of question-shaped listening is that the person stops volunteering and waits to be asked, which puts you in charge of a story you know less about than they do. Gordon lists probing and questioning among his roadblocks for exactly that reason. None of this makes questions bad; it makes an unbroken chain of them bad. During the exploratory part of an interview, aim for more reflections than questions, and ask one at a time. Motivational interviewing develops reflection much further than this lesson does, into complex, double-sided, and amplified reflections used deliberately to work with ambivalence; that machinery, and the OARS acronym that organizes it, belongs to the motivational interviewing topic.
Gordon's roadblocks and the reassurance problem
Gordon's twelve roadblocks came out of his dissertation research, which built on studies by Carl Rogers' students of responses that shut clients down, and they entered his first leadership training program in 1957. The list is usually given as ordering or directing; warning or threatening; moralizing or preaching; advising or giving solutions; persuading with logic or arguing; judging, criticizing, or blaming; praising or agreeing; name-calling or ridiculing; analyzing or diagnosing; reassuring or sympathizing; probing and questioning; and diverting, sarcasm, or withdrawal. Gordon's own publications word the list differently in places, so treat the categories as a diagnostic vocabulary rather than a fixed canon. What unites them, in his organization's account, is effect: the other person stops talking, the problem is quietly transferred to the listener, and the response communicates that the person as they currently are is not acceptable. Two entries surprise students. Praise is a roadblock because it is still an evaluation, and evaluation invites performance. Reassurance is the more important one. When you tell someone it will be fine, or that plenty of families go through this, you have contradicted a claim they were in the middle of making, answered a feeling before they finished expressing it, and signaled that the topic is uncomfortable for you. The topic usually closes. This is not an argument against ever giving advice or information. Case management is largely advice and information, and withholding a benefits deadline in order to stay non-directive would be malpractice of a different kind. The argument is about sequencing: advice offered before you understand the problem is usually advice about the wrong problem, and it costs you the disclosures that would have corrected it.
Content, feeling, meaning, and the record
It helps to hear three layers in the same sentence. Content is what happened. Feeling is the affect carried with it. Meaning is what the event signifies to this person, which is the layer you cannot deduce and must check. "I had to ask my mother for the rent" carries content anyone could transcribe, a feeling that could be relief or humiliation, and a meaning that depends entirely on a family history you have not heard yet. Assuming the meaning is the most common way an interview goes wrong while appearing to go well. Checking is cheap: name what you think you heard, mark it as provisional, and let the person edit. Documentation is the last stage of listening rather than a separate chore. Recall is the lossy stage of the process, so the interval between the conversation and the note is where accuracy is lost. Write soon, and keep three registers separate in the record: what the person said, with short quotations where the exact words carry weight; what you observed; and what you inferred, labeled as inference. "Client appeared overwhelmed" is your inference wearing the costume of an observation. "Client said she has not slept more than four hours since the eviction notice and stopped mid-sentence twice" is what you actually have. The reflections you offered during the interview are also accuracy checks, so the corrections a person made to them belong in the record too. Agency format and jurisdictional requirements govern what a record must contain and who may see it.
Interpreters, telephones, and cameras
An interpreted conversation is a three-party conversation with rules. The NCIHC national standards of practice direct the interpreter to render all messages accurately and completely, without adding, omitting, or substituting; to replicate the speaker's register, style, and tone; to advise everyone that everything said will be interpreted; to manage the flow of talk, for instance by asking a speaker to pause or slow down; to correct their own errors; and to be transparent about any exchange the others cannot follow. Two further standards shape your behavior: the interpreter promotes direct communication, telling the parties to address each other, and limits their activity to interpreting, so an interpreter asked for advice redirects the question. In practice: speak to the person, not about them; keep segments short enough to be rendered whole; drop idioms; and never ask the interpreter what the client really meant, or ask them to condense. If two minutes of speech comes back as one sentence, pause and ask for the full rendering. Who interprets is not your discretion either: the National CLAS Standards require language assistance at no cost, notice that it is available, and verification that the provider is competent, and they discourage using untrained people and minors. A family member as interpreter is a confidentiality problem, an accuracy problem, and sometimes a safety problem. Remote contact removes different things. On an audio-only call your visual back-channel cues do not exist: nods buy you nothing, and encouragers have to be audible without becoming interruptions. Silence, which is a tool in the room, becomes ambiguous on a phone line and needs labeling, and video latency does the same thing in a smaller way. Checking understanding Explicitly testing your interpretation with the speaker and inviting correction, instead of proceeding on an assumption about what an event meant to them. Full entry →, always useful, becomes the main instrument you have left when you cannot read a face. Treat these as reasoned adjustments rather than tested findings, because the experimental literature here is much thinner than the practice advice built on it.
What the evidence actually shows
Active listening is taught almost universally in the helping professions, and its component parts have uneven experimental support. Saying so is better than manufacturing confidence. The strongest results concern listening quality as a bundle. A 2020 study in the Journal of Experimental Social Psychology defined high-quality listening as conveying empathy, interest-taking, and unconditional regard, and across a pilot and three studies with 952 participants reported greater speaker self-insight, greater openness to attitude change, and lower prejudiced attitudes, with self-insight carrying the effect. The authors themselves flag the limits: short-term comparisons, self-report measures, and no test of durability. Component-level evidence is thinner. In a 2010 experiment with 180 undergraduates, receiving a paraphrase rather than a plain acknowledgement made participants rate the listener as more socially attractive but did not make them more satisfied with the conversation or more likely to feel understood, which is precisely the outcome training curricula promise. Training evidence sits between the two. The 2018 Cochrane review pooled 17 randomized trials with 1,240 healthcare professionals and found that trained clinicians asked more open questions, showed more empathy, and gave facts-only responses less often, at moderate to low certainty. It found no significant change in eliciting concerns, giving appropriate information, clarifying and Summarizing Gathering several minutes of talk into a few sentences to close one section of an interview, open the next, and expose gaps before they reach the record. Full entry →, or negotiation, no evidence of benefit for professionals' burnout or for patients' satisfaction, perceptions, or health, and no way to tell whether the gains persist. Read together the picture is coherent: listening quality appears to do real work, specific techniques are how people learn to produce it, and no single technique is the mechanism. That is a good reason to practice the moves and a bad reason to perform them, since a paraphrase produced without interest is a paraphrase people can hear coming.

Eli explains
The same idea, in plain words
Explain it like I’m 10
Imagine telling somebody about the worst afternoon you have had all year. Halfway through, they say, "Don't worry, it'll be fine." You stop. Not because they were unkind, but because they answered before you had finished, and now finishing feels like arguing. Listening well is mostly about not doing that. You make small noises so the person knows you are still there. You say back what you heard, in your own words, so they can fix it if you got it wrong. You say what you think they are feeling, and let them correct you. Every so often you gather it all up and check it. You ask questions that cannot be answered with one word, and you leave gaps of silence, because the sentence somebody says after a pause is usually the one they were working up to. None of this is a personality you either have or lack. It is a short list of moves that people learn, practice badly, get corrected on, and eventually do without thinking.
Picture it like this
It works like a good server repeating your order back to the table. They do not decide what you should have ordered, and they do not tell you the fish is a mistake. They say the order back, and you get the chance to say "actually, no onions" before anything reaches the kitchen.
Where the picture stops working
The comparison breaks in three places. An order has a correct answer that can be checked against a ticket, while what somebody means by their own situation has no ticket and only they can confirm it. A server wants the exchange over quickly, whereas listening well often means slowing down and leaving space. And a server needs no genuine interest for the read-back to work, while the research suggests it is listening quality — real empathy, interest, and non-judgment — that produces the effect, and a paraphrase delivered as pure technique is one people can hear coming.
Worked example
A worker at a community agency meets Ms. R, who came in about a rent arrears letter. Names and details are invented; this is a craft demonstration, not case direction.
Ms. R: "They cut my hours in March and I've been juggling since. Now the school keeps calling me at work about Deion, and I can't keep taking those calls."
Weak response 1 — "Have you applied for emergency rental assistance yet? You should call 211 today." Diagnosis: advising and giving solutions, Gordon's fourth roadblock, plus a closed question. It is also premature. The worker has heard about hours, a school, and a son, and has no idea yet which of those is the actual problem. Rewrite: "Two things landing at once — the hours dropping in March, and now the school calling you at work." That is a paraphrase in statement form. It does not ask her to certify anything; it lets her pick which thread to pull.
Ms. R: "He's been getting into it with other kids. I know he's angry. I'm the one who's not there." (She looks away.)
Weak response 2 — "Oh, don't worry, you're a great mom. Plenty of families go through this." Diagnosis: reassuring and sympathizing, tangled with praising — roadblocks ten and seven. It contradicts a claim she is still in the middle of making, evaluates her when she did not ask to be evaluated, and tells her this subject makes the worker uncomfortable. She will almost certainly change the subject. Rewrite: "Being the one who isn't there is the part that's hardest to say out loud." Then nothing. Three seconds of silence, held on purpose.
Ms. R: "I never wanted him raised the way I got raised."
Weak response 3 — "So you feel like you're letting him down?" (rising intonation). Diagnosis: this is a question dressed as a reflection. It hands the sentence back for a yes or no, and it substitutes the worker's word — letting him down — for the comparison she actually made, which was to her own childhood. Rewrite: "You're measuring this against how you were raised, and that's a hard measure." Statement form, her frame, and she can correct it if the worker has it wrong.
Closing the section — "Let me check I've got this. Hours cut in March, arrears since, the school calling during shifts, and underneath it Deion and how he's doing. What's the piece that's worrying you most this week?" That is a summary followed by one open question, and it moves the interview toward the practical work without deciding for her what the practical work is.
In the record afterward: quote her comparison to her own upbringing rather than writing "client feels guilty," and log "looked away, paused" as observation rather than as evidence of shame.
Key takeaway
Active listening is a short list of learnable moves — attend, encourage, paraphrase, reflect feeling, summarize, ask open questions, and let silences sit — and most of the skill lies in withholding the roadblock, especially reassurance, until the person has finished saying what they came to say.
Quick check
3 questions here, of 5 in this lesson’s practice set. Answers stay hidden until you check.
A student says to a client, with rising intonation, "So you're feeling like nobody at the school listens to you?" What is the most accurate critique of this response?
In Weger, Castle, and Emmett's 2010 experiment, participants who received a paraphrase rather than a simple acknowledgement:
Study tools & related lessonsYou’ll learn to · Common mistakes · Easily confused · Key vocabulary · Related
You’ll learn to
- Define active listening as an observable behavior set and attribute its origins to Rogers and Farson (1957) and to Thomas Gordon's work.
- Distinguish a reflection from a question, and a paraphrase from a reflection of feeling, in both form and function.
- Apply minimal encouragers, paraphrase, reflection of feeling, summarizing, open questions, and silence to a practice conversation.
- Analyze a practitioner response for Gordon's communication roadblocks and rewrite it.
- Explain how listening changes in interpreter-mediated, telephone, and video contact.
- Evaluate what the experimental evidence does and does not establish about active listening and listening training.
Common mistakes
Turning every reflection into a question by adding a rising intonation or a tag — "so you're feeling overwhelmed?"
Deliver reflections as statements. A statement lets the person elaborate or correct you; a question asks them to certify your wording with a yes or no, and a chain of them makes you the interview's driver.
Reassuring someone early, because the silence is uncomfortable and reassurance feels kind.
Premature reassurance is one of Gordon's roadblocks: it contradicts a claim the person is still making and signals that the topic bothers you. Name the feeling instead, let a pause sit, and let them finish.
Parroting the speaker's exact words back and calling it a paraphrase.
A paraphrase condenses and re-words, which is what shows you processed rather than recorded. Repetition of a whole phrase reads as a technique being applied to them, and people notice.
Treating attending as a set of universal rules, such as maintaining steady eye contact.
Nonverbal expectations vary across cultures and between individuals, and looking away often signals thinking rather than disengagement. Calibrate, watch the effect, and where it matters, ask the person what works for them.
Using a family member as interpreter, or asking a professional interpreter to summarize or explain what the client really meant.
The CLAS Standards discourage untrained people and minors as interpreters, and interpreting standards require complete rendering without omission and limit the interpreter to interpreting. Ask for the full rendering and put cultural questions to the client.
Easily confused
Reflection vs. Question
A reflection is a statement in which the listener takes a position and risks being wrong, which invites elaboration; a question is an interrogative that returns the work to the speaker and invites confirmation or denial.
Paraphrase vs. Reflection of feeling
A paraphrase restates content — what happened, condensed and re-worded. A reflection of feeling names the affect carried with the content, whether or not the speaker labeled it.
Summarizing vs. Paraphrasing
A paraphrase handles one contribution and keeps the exchange moving; a summary gathers several minutes, marks a transition in the interview, and surfaces gaps before they reach the written record.
Empathy vs. Reassurance
Empathy states the feeling as the person is presenting it and leaves the topic open; reassurance asserts that the feeling is unwarranted or the situation is fine, which contradicts them and usually closes the topic.
Hearing vs. Listening
Hearing is the physiological reception of sound; listening is the cognitive process of receiving, interpreting, recalling, evaluating, and responding, which is why listening can be trained and rated while hearing cannot.
Key vocabulary
- Attending
- The visible, physical side of listening — body orientation, gaze, stillness, note-taking choices — calibrated to the person rather than applied as a fixed posture.
- Minimal encourager
- A brief signal such as mm-hm, right, a nod, or a lean forward, given while the other person is still speaking to keep the floor with them.
- Back-channel cue
- The communication-research name for those small verbal and nonverbal signals a listener sends during someone else's turn to show engagement.
- Paraphrase
- A condensed restatement of the content someone just gave you, in your own words and offered tentatively so they can correct it.
- Reflection of feeling
- A statement naming the emotion you believe you heard, whether or not the speaker labeled it themselves, offered for them to confirm or revise.
- Summarizing
- Gathering several minutes of talk into a few sentences to close one section of an interview, open the next, and expose gaps before they reach the record.
- Open question
- A prompt whose form lets the speaker choose what to say and how much, rather than fixing a single variable such as a date or a yes-or-no fact.
- Communication roadblock
- Thomas Gordon's term for a category of response — advising, reassuring, warning, judging and eight others — that predictably stops a troubled person from continuing.
- Pseudo-listening
- Producing the outward signals of attention while not processing the message; observers usually detect it, and it costs the listener credibility.
- Checking understanding
- Explicitly testing your interpretation with the speaker and inviting correction, instead of proceeding on an assumption about what an event meant to them.
Sources & references
- Active Listening (bibliographic record) — Google Books
- Origins of the Gordon Model — Gordon Training International
- What Are the Roadblocks to Helping Your Child? — Gordon Training International
- Do You Use the Dirty Dozen When You Communicate? — Gordon Training International
- Communication in the Real World: An Introduction to Communication Studies, Chapter 5 — Listening — Social Sci LibreTexts
- Active Listening in Peer Interviews: The Influence of Message Paraphrasing on Perceptions of Listening Skill — ERIC (International Journal of Listening, 24(1), 34-49)
- Can high quality listening predict lower speakers' prejudiced attitudes? — Journal of Experimental Social Psychology (via PubMed Central)
- Communication skills training for healthcare professionals working with people who have cancer (Cochrane review, 2018 update) — Cochrane Database of Systematic Reviews (via PubMed Central)
- National Standards of Practice for Interpreters in Health Care — National Council on Interpreting in Health Care
- National Standards for Culturally and Linguistically Appropriate Services (CLAS) in Health and Health Care — U.S. Department of Health and Human Services, Office of Minority Health (Think Cultural Health)
EliExplains lessons are original prose written from the open, credible references above. See Copyright & Licensing.
Researched 2026-08-18
Educational content only. It is not medical, legal or professional advice. Found an error? Tell us.

