The J sound in jump is CH with the voice switched on, and 90% of children produce it by around age 5. It is a voiced affricate, written /dʒ/ in the International Phonetic Alphabet (IPA): a D-like stop that releases straight into a buzzy ZH, all in one sound. So a child who can already say CH is one voicing switch away from J, and a child who says “dump” for jump is missing the second half of the sound.
One notation trap before going further: in IPA, the symbol /j/ is the Y sound in yes, not the J in jump. This guide writes the jump sound as /j/, the plain-letter style our sound pages use, and glosses it as /dʒ/ where the difference matters.
This guide shows SLPs and parents how to teach /j/ and fix its three errors (stopping, deaffrication and devoicing), with shaping routes, word lists, J vs D minimal pairs, norms and sample IEP goals.
It is written primarily for SLPs, with a clearly marked home-practice section for parents. Two tables are worth bookmarking: the spelling table, for parents who search by letter, and the error-triage table, which maps each error to its fix.
What the /J/ Sound Is (and Why It Is Spelled Four Ways)
/j/ is one of English’s only two affricates, and the only voiced one: a stop released into a fricative, with the voice on throughout. The tongue makes a brief closure just behind the gum ridge, a little further back than for D, then releases the air as a buzzy ZH while the lips round slightly forward. That is three jobs at once (stop, fricative and voice), and in Kent’s (1992) grouping of consonants by motor complexity, as McLeod & Crowe (2018) summarize it, the affricates fall in the most demanding set.
Phonetically, /j/ is a voiced palatal (post-alveolar) affricate, and it is CH’s voiced partner: same place, same affricate manner, with the voice as the only difference. The two share identical norms and tend to emerge together. Because only the voice differs, most of the shaping ladder in our CH guide carries straight over; for a single-sound reference, see the /j/ sound page.
On timing, /j/ sits in Shriberg’s Middle-8 (ng, t, k, g, f, v, ch, j), and McLeod & Crowe’s own regrouping of the English data also places /dʒ/ in their middle group. It is mid-pack, not late, so a J error that outlasts its mastery age deserves attention earlier than a TH or R error would.
Four spellings, one sound
Parents often search by letter, and the letter misleads: the J sound has four main spellings, plus a few rare ones.
| Spelling | Rule of thumb | Examples |
|---|---|---|
| j | starts words; English words almost never end in j | jump, jet, juice |
| g | before e, i or y (a “soft g”) | magic, engine, energy |
| ge | ends words after a long vowel, a consonant or an unstressed -age | cage, change, cabbage |
| dge | ends words right after a short vowel | bridge, badge, edge |
| rare: dg, gg, di | uncommon, mid-word | badger, veggie, soldier |
Sound Safari’s own /j/ practice lists show the pattern clearly. All 30 initial words start with the letter j, but not one of the 29 final words ends in it: every one is spelled -ge or -dge (cage, huge, bridge). Only 2 of the 29 medial words, major and project, use a j at all, and across the three lists, spellings with a g outnumber the letter j.
When Should a Child Say the /J/ Sound?
About half of children produce /j/ by 3;6, and 90% by 5;0. The 90% figure comes from Crowe & McLeod’s 2020 review, which pooled 15 U.S. studies of 18,907 children and places /dʒ/ in the 4;0–4;11 band at the 90% criterion. In the review’s own words, “all affricates were acquired by 4;11.”
Its cross-linguistic companion, McLeod & Crowe (2018), reached the same broad conclusion across 27 languages: most consonants are acquired by 5;0.
| Sound | 50% (customary) | 90% (mastery) | Why it is here |
|---|---|---|---|
| J /dʒ/ | 3;6 | 5;0 | the target: a voiced affricate |
| CH /tʃ/ | 3;6 | 5;0 | its voiceless twin, with the same ages |
| D /d/ | — | 3;0 | the stop a stopping child falls back on |
| ZH /ʒ/ | — | 6;0 | the fricative a deaffricating child lands on |
| /s/ | — | 5;0 | for comparison |
| /r/ | — | 6;0 | for comparison: J comes earlier |
90% ages from Crowe & McLeod (2020), using the review’s “by age” framing; companion cross-linguistic data in McLeod & Crowe (2018). The 50% ages are the values in Sound Safari’s norm set.
Ages use the standard speech-pathology years;months format, not decimals: 3;6 means 3 years, 6 months (not 3.6 years), and 5;0 means 5 years, 0 months. Read the mastery age as around age 5, not as a hard cutoff; the “by age” framing is why the review’s 4;0–4;11 band appears here as 5;0.
That frames the IEP decision: a child past about 5;0 who still substitutes /j/, with an intelligibility or participation impact, is typically a reasonable target. Being under age does not rule out service (a highly unintelligible 4-year-old may still need it), and being over age does not mandate it (a mild, fully intelligible residual error may not meet eligibility).
The error patterns in the next section run on their own clocks, which our phonological processes guide lays out. Stopping of /ʃ/, /tʃ/ and /dʒ/ is typically gone by 4;6, with concern past 5;6, and deaffrication is typically gone by 4;0, with concern past 4;6.
ASHA’s table of selected phonological patterns points the same way (stopping of these sounds likely gone by age 5, deaffrication by 4), and our speech sound milestones by age has the full consonant timeline.
The Three /J/ Error Patterns
Each of the three common /j/ errors drops exactly one of the sound’s three parts, and each runs on its own clock. Stopping is typically gone by 4;6 and deaffrication by 4;0; the published tables give devoicing no J-specific age, so this guide uses about 4;0 as a working threshold (explained below). Which part is missing tells you which fix to reach for.
| Pattern | Sounds like | What is missing | Process | Typical by / concern past | Fix with |
|---|---|---|---|---|---|
| j→d | “dump” for jump | the fricative release | stopping | 4;6 / 5;6 | J vs D minimal pairs + the add-the-release route |
| j→zh | “zhump” for jump | the stop | deaffrication | 4;0 / 4;6 | the add-the-stop route, from a held ZH |
| j→ch | “chump” for jump | the voice | devoicing | about 4;0, this guide’s working threshold (no J-specific published age) / past about 4 | voicing cues + a CH vs J contrast |
j→d is stopping. The child keeps a stop they mastered early (90% of children have /d/ by 3;0, per McLeod & Crowe) and drops the harder fizzy release, so jump becomes “dump” and jeep becomes “deep.” Caroline Bowen’s elimination table uses that very example and gives stopping of J as gone by about 4;6, so a j→d that lingers past 5;6 is a phonological target rather than a stray slip.
It often travels with ch→t and sh→t, which is why the stopping goal later in this guide is written as a pattern goal.
j→zh is deaffrication. The child keeps the buzzy friction and drops the stop: “zhump” for jump. For J this error has a twist, because the sound the child lands on, ZH (/ʒ/, the sound in the middle of measure), develops later than J itself: Crowe & McLeod place /ʒ/ in the 5;0–5;11 band at the 90% criterion, a full band after /dʒ/.
A child who says “zhump” already owns that late fricative, which is exactly what the add-the-stop route builds on. A variant drops the same stop but lands on /z/ (“zump”), the voiced twin of the ch→s variant in our CH guide. On the CH side, the classic version is ch→sh (“ship” for chip).
j→ch is devoicing. Place and manner are right, and only the voice is off: “chump” for jump. The published tables give no J-specific devoicing age: Bowen’s generic word-final devoicing (pig said as “pick”) is gone by about 3;0, and ASHA’s table has no devoicing row.
Devoicing as a general pattern is gone early: Bowen’s table puts word-final devoicing at 3;0. But half of children do not produce /j/ until 3;6 and 90% not until 5;0, so a “chump” for jump from a 3-year-old is usually a sound that has not arrived yet rather than a pattern to treat. This guide therefore uses about 4;0 as its working threshold: a “chump” for jump past about 4 is worth a closer look, and the fix is the voicing cue rather than a wait-and-see. One rarer slip to know: if the stop and the friction both go, J slides to the Y glide (“yump” for jump).
How to Teach the /J/ Sound: Shape It From CH, D or ZH
J differs from CH by one feature (the voice), from D by one part (the fricative release) and from ZH by the other part (the stop). The fastest route is to shape it from whichever neighbor the child already owns. Probe first, because the error picks the route; then lock the voicing and generalize.
Step 1 — Probe stimulability and sort the error
Model /j/ in isolation and in jump, and listen for which part is missing: a D means the release is gone, a ZH means the stop is gone, and a CH means the voice is gone. Then check the neighbors on their own: can the child say CH, D and ZH? Most children with a J error already own at least one of them, and that sound is your starting point.
Step 2 — Route A: from CH, add the voice
Best for devoicing, or for any child with a solid CH. Set the shared placement: lips rounded and pushed slightly forward, teeth close together, tongue blade up just behind the gum ridge.
Have the child say a crisp CH, then say it again with the “motor on,” hand on the throat to feel the buzz, while nothing else moves. If you have used the shaping ladder in our CH guide, this rung comes almost free.
Step 3 — Route B: from D, add the release
Best for stopping. Start from the child’s own /d/ with the tongue tip a touch further back than usual, just behind the gum ridge; hold the contact for a beat, then let it go slowly into a buzzy “zhh” instead of a clean pop. Say “d… zhh,” then “d-zhh,” then faster until the two fuse into J, with the cue “a D that pushes the buzz out” and a mirror to show the lips rounding.
Step 4 — Route C: from ZH, add the stop
Best for deaffrication. Have the child hold a long, buzzy ZH, then add a quick tongue-tip closure behind the ridge that blocks the air for an instant, and release straight back into the buzz: “zhhh… d-zh.” That brief block is the missing stop (stop-then-release is what makes any affricate), and for the /z/ variant, cue the tongue back and the lips round before the closure.
Step 5 — Lock the voicing with contrast
Whatever the route, the voice has to stay on, and two cues check it. The tactile cue is a hand on the throat: buzz for J, quiet for CH. The listening cue is vowel length: the vowel before a final J is longer than the vowel before CH, so ridge holds its vowel longer than rich does.
Phoneticians call this pre-fortis clipping: vowels are clipped short before voiceless consonants. In word-final position the buzz on J itself is often faint, even in adult speech, so vowel length carries much of the contrast there. Keep the lips rounded throughout, since rounding is part of the J shape.
Step 6 — Generalize through the hierarchy
Once J is reliable in isolation, climb the standard hierarchy: isolation, syllable (jo, jay, jee), word, phrase, sentence and story. Work word-initial J first, then word-final J, then J between vowels, the order the word lists and IEP goals below follow.
/J/ Word Lists by Position
Sound Safari ships 88 /j/ practice words: 30 initial, 29 medial and 29 final. Here is a working sample from each position; the /j/ sound page linked above has the full lists.
| Initial /j/ | Medial /j/ | Final /j/ |
|---|---|---|
| jump | magic | bridge |
| juice | pigeon | cage |
| jet | engine | page |
| jam | angel | badge |
| jelly | ginger | judge |
Every word in the medial and final columns spells J with a g (g, ge or dge), which is the usual answer when a parent asks why magic counts as a J word.
Start word-initial, where the lip rounding and the CH or D shaping are easiest to cue. Move to final J next, where the vowel-length cue is available, and leave medial J for last, because J between two vowels is the easiest to collapse to D or ZH.
Sound Safari generates these position-sorted /j/ lists for you, so you pick from a set rather than build one.
J vs D Minimal Pairs: Fixing Stopping
As few as 3–5 well-chosen minimal pairs can drive generalization to untreated words (Elbert et al., 1991, as Caroline Bowen cites it), and for j→d the pairs to choose contrast J with D. Dump and jump mean different things, so the child needs the fricative release to be understood.
Sound Safari’s J vs D set, which the app files under stopping, includes these pairs:
| J | D |
|---|---|
| jeep | deep |
| jig | dig |
| jog | dog |
| jolly | dolly |
| jump | dump |
| June | dune |
| just | dust |
| hedge | head |
Why the contrast works: it adds back exactly the part stopping drops, giving the child a communicative reason to release the stop into friction. Hedge versus head carries the same contrast to the end of a word, and the logic matches the CH vs T set in our CH guide.
If the child’s stop is voiceless (“toy” for joy), the app also carries a J vs T contrast, filed under stopping too, with pairs such as joy/toy and jug/tug. Because j→d often travels with ch→t and sh→t, treat it as a pattern; browse every contrast in our minimal pairs library.
CH vs J: The Voicing Contrast
CH and J differ in exactly one feature, voicing, so devoicing (j→ch) is fixed by making the voice meaningful. Neither the app’s minimal-pairs game nor this site ships a CH vs J set, so the pairs below are hand-picked examples:
| CH (voice off) | J (voice on) |
|---|---|
| cheap | jeep |
| chug | jug |
| chunk | junk |
| choke | joke |
| rich | ridge |
| etch | edge |
| batch | badge |
| lunch | lunge |
Teach the listening side with two cues. Hand on the throat: buzz means J, quiet means CH. Then vowel length: the vowel before a final J is longer, so ridge stretches where rich is clipped, and at the ends of words that length difference is often the stronger of the two cues.
This is the contrast behind Route A in the elicitation steps above. The list also runs backwards: for a child who voices CH (“junk” for chunk), keep the same pairs and flip the target to the CH word.
Sample IEP Goals for the /J/ Sound
A /j/ goal either targets correct /j/ by word position (for devoicing or deaffrication) or eliminates stopping as a pattern (for j→d), typically at 80% accuracy across 3 consecutive sessions.
Sound Safari’s goal bank ships 22 built-in templates across 8 categories, all defaulting to an 80% accuracy criterion across 3 consecutive sessions. Once /j/ is confirmed as a target (past about 5;0, with an intelligibility impact), two of those templates cover it.
The sample goals below say “20 trials in 3 of 4 consecutive sessions”; those are common clinician conventions layered on the app’s 3-session default, not app values.
Devoicing and deaffrication go through the articulation “Sound in Word Position” template, because the bank has no voicing or deaffrication template and the goal is simply a correct /j/. Suggested baseline 0–40%, 80% accuracy, 12–36 weeks:
Goal 1 (initial /j/, word position): Given a verbal model and tactile cues, [Student] will produce /j/ in the initial position of single-syllable words (e.g., jump, jet, jam) with 80% accuracy across 20 trials in 3 of 4 consecutive sessions by the end of the IEP year. Suggested baseline: 0–40%. Timeframe: 12–36 weeks.
Final position makes a natural second goal, since the vowel-length cue is available there:
Goal 2 (final /j/, word position): Given a verbal model and auditory cues, [Student] will produce /j/ in the final position of single-syllable words (e.g., bridge, cage, page) with 80% accuracy across 20 trials in 3 of 4 consecutive sessions by the end of the IEP year. Suggested baseline: 0–40%. Timeframe: 12–36 weeks.
Stopping is different. Because stopping usually spans several sounds, j→d belongs to the phonology “Stopping” template (suggested baseline 0–30%, 80% accuracy, 20–40 weeks). Its text is written for fricatives, so for J you adapt it:
Goal 3 (stopping, phonology, adapted): [Student] will eliminate the process of stopping by producing fricatives and affricates, including /j/ and /ch/, correctly with 80% accuracy across 3 consecutive sessions by the end of the IEP year. The template supplies the 80% criterion, suggested baseline (0–30%) and timeframe (20–40 weeks); its text says fricatives, so “and affricates, including /j/ and /ch/” is your edit.
A sentence-level goal is a common next step you write yourself. The app has no dedicated sentence-level template, so the baseline and timeframe below are illustrative planning conventions, not app values:
Goal 4 (sentence level, illustrative): Given a sentence frame and picture cues, [Student] will produce /j/ in all word positions within structured sentences with 80% accuracy across 20 trials in 3 of 4 consecutive sessions by the end of the IEP year. Illustrative baseline: 35–55%. Illustrative timeframe: 16–24 weeks.
Two scaffolds operationalize these goals. Trials are scored Correct / Approximate / Incorrect (Approximate is an opt-in setting, so the default is Correct/Incorrect), and four prompting levels run from Maximum (Picture + Word + Model) through Moderate (Picture + Word) and Minimal (Word Only) to Independent (Picture Only). Fading from Maximum to Independent is how a goal’s cue condition tightens over the year.
Every goal here is a starting point, not a substitute for clinical judgment, baseline data or your district’s conventions. For the full copy-paste set, see the CH progression in our IEP goal templates, which you can adapt for /j/ by swapping /tʃ/ for /dʒ/. In Sound Safari, session scoring links straight to the goals you have written, so the data behind each one stays current without manual re-entry.
Home Practice for Parents
Most children master the J sound by around age 5 (90% of them by 5;0), so a 3- or 4-year-old who says “dump” for jump is often right on track. This section is for parents and caregivers, and no clinical background is needed; an “SLP” is a speech-language pathologist, the professional who treats speech sounds.
It is worth checking with an SLP if your child is clearly past about 5 and still swapping J, if they are past about 4 and still saying “ch” for “j” (chump for jump), or if unfamiliar listeners struggle to understand them. Our guide on when to see a speech therapist walks through that decision by age.
One spelling note, because parents often search by letter: the J sound hides inside words spelled with g, ge or dge, like magic, cage and bridge, and they all count as J words.
A few simple things help at home:
- Buzz check. Put your child’s hand on your throat, then on their own. CH is quiet and J buzzes: same mouth shape, different switch.
- D plus buzz. Have your child say “d,” then let it fizz out slowly instead of popping: “d… zhh.” Speed it up and it turns into J.
- Model, don’t correct. If your child says “dump” for jump, don’t ask for a redo. Say the word correctly two or three times in natural talk (“Yes, jump! A big jump!”); hearing the right version matters more than drilling it.
- Jump game. Lay out a few pillows and have your child say “jump” before each hop. One clear J earns one jump, and three to five minutes is plenty.
- Picture pairs. Line up pictures of jump/dump and jeep/deep and let your child hear, and see, that they are different words. A guessing game keeps it light.
If your child’s SLP uses Sound Safari, they can send you a homework link that opens right in your web browser, with no app to download. On privacy, no third party ever receives your child’s clinical or student data: no ads, no behavioral tracking, no data brokers. Account sign-in and subscriptions are handled by standard providers that see only account and billing details, never session data; see what parents get.
Sound Safari is a clinical tool, not a medical device. It does not provide diagnoses or treatment recommendations, and clinical decisions should always be made by qualified professionals.
From Practice to Progress: Tracking /J/ Therapy
The hard part of J therapy is knowing which of the three errors a child is making, because each one has a different fix. In Sound Safari’s screener, you note what the child said instead (the quick-picks for /j/ are d, ch and y, and you can type another sound such as zh). The app then labels the likely pattern on the spot: stopping for d, devoicing for ch, deaffrication for zh.
You can use that label, together with your own judgment, to choose between J vs D pairs, the ZH-to-J route or the voicing contrast.
The built-in articulation screener covers 24 sounds across initial, medial and final positions, and is included in Pro, with no separate assessment app. For J it samples initial, medial and final position (jump, magic, bridge), offers d, ch and y as one-tap substitutions, and groups /j/ in its Middle band (4–5 years).
It also generates SOAP notes inside the articulation app: session data flows into the standard four-section structure (Subjective, Objective, Assessment, Plan), so the note is largely written by the time the session ends. Practice, screen, document, and share — all in one app. For the screening side, see our articulation screening checklist; for documentation, see writing SOAP notes in speech therapy.
Sound Safari is a clinical tool, not a medical device. It does not provide diagnoses or treatment recommendations. The screener is informational, and clinical decisions should always be made by qualified professionals; to put the full loop to work, see Sound Safari for professionals.
Frequently Asked Questions
At what age should a child say the J sound?
About half of children produce the J sound by 3;6 (three years, six months), and 90% by about age 5: Crowe & McLeod’s (2020) review of U.S. studies places /dʒ/ in the 4;0–4;11 band at the 90% criterion. J errors before age 5 are usually developmentally typical. Past 5, with an intelligibility impact, J is often a reasonable therapy target.
Why does my child say “d” for “j” (dump for jump)?
That is a pattern speech therapists call stopping: the child keeps the D, a sound 90% of children have by age 3, and drops the fizzy release that turns D into J. Stopping of J is typically gone by about age 4½, so it is worth a closer look past about 5½, or sooner if your child is hard to understand. The standard fix is J-vs-D minimal pairs such as jump/dump and jeep/deep.
What is the difference between the CH and J sounds?
CH and J use the same mouth position (lips rounded, teeth close, tongue blade just behind the gum ridge) and differ only in voicing. J adds the voice, which you can feel as a buzz with a hand on the throat; CH is quiet. You can hear the difference in vowel length too: the vowel before a final J is longer than the vowel before CH, as in ridge versus rich.
How do you teach the J sound, and where does the tongue go?
Round the lips slightly forward, keep the teeth close, and raise the tongue blade to just behind the gum ridge, the same position as CH. Then shape J from a sound the child already has: say CH and switch the voice on, or say D and let it release slowly into a buzzy zh instead of popping. A hand on the throat checks the buzz, and a mirror checks the lips.
Is J an affricate?
Yes: J is a voiced affricate, a stop (like D) released into a fricative (like the zh in measure), with the voice on throughout. CH is its voiceless twin. In the International Phonetic Alphabet the J sound is written /dʒ/; the IPA symbol /j/ stands for the Y sound in yes, which trips up many readers.
Why is the J sound spelled with a g (giraffe, bridge, cage)?
English spells the J sound four main ways: j (jump), g before e, i or y (giraffe, magic), ge (cage) and dge (bridge), plus rare spellings such as di (soldier). English words almost never end in the letter j, so a J at the end of a word is spelled -ge or -dge. In Sound Safari’s /j/ practice lists, all 29 final words end in -ge or -dge, and only 2 of the 29 medial words use the letter j.
When should we see a speech therapist about my child’s J sound?
Consider an evaluation if your child is clearly past about age 5 and still swaps the J sound, if they are past about 4 and still say “ch” for “j” (chump for jump), or if unfamiliar listeners struggle to understand them. Your school district or a private speech-language pathologist (SLP) can screen or evaluate your child and tell you whether therapy would help. Sound Safari is a clinical tool, not a medical device; clinical decisions should always be made by qualified professionals.
Closing
J is CH with the voice on: a stop, a fricative and voicing in one sound. Each of its three common errors drops one part and has its own route: J vs D pairs and the add-the-release route for stopping, the add-the-stop route for deaffrication, and the voicing cues for devoicing. Gate the decision to treat on the mastery age of around 5, and look sooner at a lingering ch-for-j.
The goals and templates here are starting points, not a replacement for your clinical judgment and baseline data. Adjust them to the student in front of you.
If you want structured J practice with position-sorted /j/ word lists, the J vs D set, automatic error-pattern detection and progress an SLP can review, Sound Safari was built for this workflow. 14-day free trial. Cancel anytime through Apple.