At what age should children acquire speech sounds?


Phonetic versus phonological acquisition



The ability to physically produce a single sound (or 'phone') is known as 'Phonetic acquisition'. This may be different from when a child is able to use that sound in combination with other sounds to form words. Single sounds have no meaning in isolation, and imitation of a sound is a motor skill.

A child may be able to imitate a single sound on request, such as [s] - Say 'sss' , but still not yet use /s/ in words such as 'saw', 'bus' and 'kissing'. In this case, the child would be said to be stimulable for [s], but not yet using the phoneme /s/ at word or utterance level.


What is an age norm and why are there different age norms?



A 'norm' (short for normative data) is when a child typically acquires a new skill. There is often a great range due to individual differences, and so criteria should be stated to help the person reading the normative data to help them understand the range. For example, Dodd et al. (2003) lists phonological error patterns used by 10% of the normal population (p. 634), and defines acquisition of single sounds (phones) by 90% of children.

If another group of researchers use other / different criteria, then different ages may be reported. Similarly, the population assessed needs to be balanced for other factors such as gender, and socioeconomic status.


UK Speech Sound Acquisition Data



The largest research study to date in the United Kingdom was undertaken by Barbara Dodd and colleagues (2003). This study assessed 684 monolingual English speaking British children aged 3;0 to 6;11 years across a range of geographical areas, balanced for a gender and socioeconomic status. An additional 32 children aged 2;0-2;11 were assessed to supplement normative data on error patterns in this age range (pp. 627-632).

Data often cited from other sources may be less accurate when considering children growing up in the UK due to being:
  • Collected from fewer children
  • Collected in a way which does not account for geographical, gender or socio-economic differences
  • Collected using different criteria for acquisition
  • Collected from another community, such as the United States, Australia etc.

Any data set is a snapshot of the population at that time. Speech realisations continue to change and evolve due to changes in society such as the influence of social media. The Dodd et al. (2003) study is now more than 20 years old and we cannot know if norms will have changed during that time. However, until another large-scale study is undertaken, these data are more reliable than other sources we are aware of.


Bilingual children and children speaking languages other than English (LOTE)



The data below was collated from children who are monolingual speakers of English growing up in the United Kingdom (UK).

These data are NOT applicable if the child has receptive and/or expressive language skills in another language. Acquiring two (or more) languages results in separate phonological systems for each language.

This means that data from the SAME speech community should be used to compare the child's age of acquisition of speech sounds and/or suppression of phonological processes.

A diagnosis of a SSD based on UK monolingual English speakers is not evidence-based if the child is exposed to two or more languages and the data below should not be cited.

There is a growing body of research into languages other than English, including monolingual and bilingual acquisition. Please see the Royal College of Speech and Language Therapists' pages here (Member log-in required).


Phonetics / Articulatory / Motor skills / Stimulability of Phones (Single Sounds)

Ability to say a single sound ( or 'phone') in isolation - 90% of children
Adapted from Dodd et al. (2003) p. 635
Age
Manner
Present
Manner
Absent
3;0-3;5
[rowspan="3"]
Plosives
[ p, b, t, d, k, g ]
Plosives
All plosives acquired
3;0-3;5
[null]
Nasals
[ m, n, ŋ ]
Fricatives
[ θ, ð, ʃ, ʒ ]
3;0-3;5
[null]
Fricatives
[ f, v, s, z, h ]
Affricates
[ t͡ʃ, d͡ʒ ]
3;6-3;11 [rowspan="3"]
Affricate [rowspan="3"]
[ t͡ʃ ] [rowspan="3"]
Fricatives
[ θ, ð, ʃ, ʒ ]
3;6-3;11 [null]
Affricate [null]
[null]
Affricate
[ d͡ʒ ]
3;6-3;11 [null]
Affricate [null]
[null]
Approximant
[ ɹ ]
4;0-4;5 [rowspan="2"]
Fricative
[ ʒ ]
Fricatives
[ θ, ð, ʃ ]
[null]
Affricate
[ d͡ʒ ]
Approximant
[ ɹ ]
4;6-4;11 [rowspan="2"]
[rowspan="2"]
No new phones added [rowspan="2"]
Fricatives
[ θ, ð, ʃ ]
[null]
[null]
[null]
Approximant
[ ɹ ]
5;0-5;5 [rowspan="2"]
Fricative [rowspan="2"]
[ ʃ ] [rowspan="2"]
Fricatives
[ θ, ð ]
[null]
[null]
[null]
Approximant
[ ɹ ]
5;6-5;11 [rowspan="2"]
[rowspan="2"]
No new phones added [rowspan="2"]
Fricatives
[ θ, ð ]
[null]
[null]
[null]
Approximant
[ ɹ ]
6;0-6;5
Approximant
[ ɹ ]
Fricatives
[ θ, ð ]
6;6-6;11
No new phones added
Fricatives
[ θ, ð ]
7;0 and above
Fricatives
[ θ, ð ]
Acquisition complete

Note that some accents/dialects do not employ dental fricatives. If the phone is not present in the speech of the parent(s)/carer then this is likely to be an accentual difference and not a disorder and therefore should NOT be the target of speech and language therapy.


Phonological error patterns used by 10% of the normal population (n=684)

Error patterns in the context of words NOT caused by Articulation Disorder
  • The child MUST be stimulable for both the target phone for the phonological process to be confirmed
  • One off examples do NOT indicate the presence of a phonological process. A pattern should be observed (usually 5 examples in a 50 word sample).
  • If in doubt, further probe words containing the target phone(s) / cluster / syllable should be transcribed.

Age group
Age range
Definition
Example
Assimilation [rowspan="2"]
<3;0
(normal pattern identified in the speech of 2-year-olds)
[rowspan="4"]
Influence of another phoneme in the target word
[rowspan="2"]
/ ˈje.ləʊ / →[ leləʊ ]
[null]
[null]
[null]
/ bɹed / →[ beb ]
Final Consonant Deletion (FCD0
[null]
Deletion of word final (WF) consonants (most commonly plosives, l, s and z).
Note: Do not confuse with WF cluster reduction. FCD results in a word ending in an open vowel.
/ dʊk / →[ dʊ ]
Note, the presence of a glottal stop word final (common in many accents as an allophone of /-t/) would be counted as a C and would NOT therefore be FCD.
Reduplication
[null]
Complete or partial duplication of a stressed syllable
/ ˈtaɪ.gə / →[ taɪtaɪ ]
Context Sensitive Voicing (CSV)
[rowspan="2"]
<3;0
[rowspan="2"]
Prevocalic (BEFORE the vowel) voicing
AND
Postvocalic (AFTER the vowel) devoicing
[rowspan="2"]
/ pɹæm / →[ bæm ]
Error of voice.
[null]
[null]
[null]
/ pɪg / →[ pɪk ]
Error of voice.
Stopping [rowspan="3"]
3;0-3;5 [rowspan="3"]
Replacement of fricatives with stops [rowspan="3"]
/ væn / →[ bæn ]
Error of manner.
[null]
[null]
[null]
/ ðɪs / → [ dɪs ]
Error of manner.
[null]
[null]
[null]
/ ˈze.bɹə / → [ debɹə ]
Error of manner.
Fronting [rowspan="3"]
3;0-3;11 [rowspan="3"]
Place of articulation is moved to a more anterior position
Velars:
/ ˈmʊŋ.ki / → [ mʊnti ] - voiced velar nasal to voiced alveolar nasal
Error of place.
[null]
[null]
Note: Stopping labiodental fricatives / f, v / to [ p, b ] respectively is NOT also fronting. Despite the fact that the place is further forward in the vocal tract, there is no analogue dental plosive in English phonology, making the bilabial position the nearest default place of articulation.
/ eg / → [ ed ] - voiced velar plosive to voiced alveolar plosive.
Error of place.
[null]
[null]
Fronting of the velars / k, g / was not present after 3;11.
More than 10% of the sample fronted / ŋ / to [ n ] in 'fishing' until the age of 5;0 despite being able to produce it correctly in other test items.
Fricatives:
/ ʃɪp / → [ sɪp ]
Error of place.
Weak syllable deletion (WSD) [rowspan="4"]
3;0-3;11 [rowspan="4"]
Deletion of an unstressed syllable [rowspan="4"]
/ tə.ˈmɑ.tə / → [ mɑtə ]
Error in word structure.
[null]
[null]
[null]
/ ʊm.ˈbɹe.lə / → [ bɹelə ]
Error in word structure.
[null]
[null]
[null]
/ d͡ʒə.ˈɹæf / → [ dɹæf ]
[null]
[null]
[null]
/ ˈe.lə.fənt / → [ efən ]
Error in word structure.
Deaffrication [rowspan="2"]
3;0-4;11 [rowspan="2"]
Moderation of the affrication feature [rowspan="2"]
/ wɒt͡ʃ / → [ wɒʃ ]
Error in manner.
[null]
[null]
[null]
/ bɹɪd͡ʒ / → [ bɹɪd͡z ]
Error in manner.
Di-cluster reduction - Word initial OR word final (WI / WF)
[rowspan="3"]
3;0-3;11 [rowspan="3"]
Deletion of one consonant from the cluster [rowspan="3"]
/ ˈsbaɪ.də / → [ baɪdə ]
Error in word structure.
[null]
[null]
[null]
/ bɹed / → [ bed ]
Error in word structure.
[null]
[null]
[null]
/ ˈe.lə.fənt / → [ ˈeləfən ]
Error in word structure.
Note that Final Consonant Deletion leaves only an open vowel, and the above example is V.CV.CVCC to V.CV.CVC, hence WF di-cluster reduction.
Tri-cluster reduction
3;0-4;11
Deletion of one or two consonants from the cluster
/ sgweə / → [ sweə ]
Error in word structure.
Gliding [rowspan="3"]
3;0-5;11 [rowspan="3"]
Replacement of liquids / l, ɹ / with glides [ w, j ] [rowspan="3"]
/ ɹabIt / →[ wabɪt ]
Error in manner.
[null]
[null]
[null]
/ lam / → [ jam]
Error in manner.
[null]
[null]
[null]
/ ˈd͡ʒe.li / → [ d͡ʒewi ]
Error in manner.
Adapted from Dodd et al. (2003), pp. 642-643 and Table 12. Phonological error patterns used by 10% of normal population (n-684), p. 634.
  • Vowels have been changed to reflect the standard Northern accent.
  • Syllable boundaries and some primary stress markers have been added.


Presence of one or more phonological processes (error patterns) over the age norm would indicate Phonological Delay
UNLESS the errors can be explained by Articulation Disorder
AND/OR if one or more unusual pattern is also observed (Consistent Phonological Disorder).

Unusual / Disordered Phonological Processes / Simplification PatternsThat occurred in less than 10% of English monolingual children in the normative sample or were observed in the clinical population

Backing
[rowspan="2"]
Place of articulation is moved to a more posterior position.
/ tiθ / → [ kiθ ]
Voiceless alveolar plosive → Voiceless velar plosive
Error of place of articulation
[null]
Note: Stopping Dental fricatives / θ, ð / to [ t, d ] respectively is NOT backing. Despite the fact that the place is further towards the back of the vocal tract, there is no analogue dental plosive in English phonology, making the alveolar position the nearest default place of articulation.
/ faɪv / → [ saɪv ]
Voiceless labiodental fricative → Voiceless alveolar fricative
Error of place of articulation
Affrication
Replacement of stops with fricatives or affricates.
/ dɒg / → [ zɒg ]
Voiced alveolar plosive → Voiced alveolar fricative
Error of Manner
Initial Consonant Deletion
Deletion of a word-initial consonant
Note: do not confuse with cluster reduction. ICD results in an open vowel at the beginning of the word.
/ fɹɒg / → [ ɒg ]
Note the CCVC to VC ('open vowel' word structure).
Error of word structure.
Within-Word Consonant Deletion
Previously known as Medial consonant deletion
Deletion or glottalisation of within word syllable initial (WWSI) / intervocalic consonants.
Note: Speech is formed of syllables and so 'Word Medial' is not appropriate for a syllabic model of speech production.
/ ˈfe.ðə / → [ fehə ]
Error of omission or place and manner
Intrusive consonants
Insertion of an extra consonant before another consonant.
/ dʊk / → [ dʊnk ]
Error of word structure: CVC to CVCC.
Denasalisation
Replacement of nasal consonants with a non-nasal sound.
/ naɪf / → [ taɪ ]
Voiced alveolar nasal to voiceless alveolar plosive.
Error of manner.
Favoured sound
Replacement of groups of consonants by a favoured sound.
All initial fricatives marked by /h/
All initial consonants marked by /d/
All final consonants marked with a nasal.

There are NO AGE NORMS for these phonological process because at any age they are classified as unusual.


Presence of ONE or MORE of the above disordered/unusual processes at any age would likely indicate a diagnosis of Consistent Phonological Disorder, provided the child had consistent word production (60% or higher).
The presence of additional delayed phonological processes does not change this diagnosis.
The errors should not be explained by Articulation Disorder.



References



Dodd, B., Holm, A., Hua, Z., & Crosbie, S. (2003). Phonological development: a normative study of British English-speaking children. Clinical Linguistics and Phonology, 17(8), 617-643. https://doi.org/10.1080/0269920031000111348

We do not collect any data or track your activity on this website.