Linguistic Detectives: Systemic Phonological Analysis of Child Speech

So far in our discussion of childhood speech sound disorders, we’ve looked at the differences between relational and independent analyses and taken an in-depth look at the place-voice-manner analysis, a relational analysis.

In this article, part three of our four-part series, we’ll be taking a closer look at the systemic phonological analysis of child speech, or SPACS. We will then conclude the series by putting it all together in a discussion that will help you choose the right analysis for the right situation.

SPACS combines independent and relational analysis to provide both a description of the child’s sound system as well as an explanation of their sound system relative to the adult sound system. SPACS analyzes children’s speech as if the disordered sound system is the child’s native language—something parents have been saying for years.

The major difference between SPACS and PVM (and other relational analyses) is the broader system-to-system comparison as opposed to the more narrow sound-to-sound comparison of the error analyses. This difference is reflected in the following ways:

1. The child’s production of one sound for multiple adult target sounds and clusters is identified. This shows the one-to-many correspondence that is characteristic of a phonological disorder.

2. A more holistic description of the child’s sound system is possible because the one-to-many correspondence includes all the target sounds and clusters produced as a single sound by the child, regardless of place, manner, or voicing.

Because there is not a pre-determined or finite number of error categories or labels used to describe the child’s errors, the identification of error patterns is not based on a set of separate, independent listings of errors by place, manner, or voicing.

Phoneme Collapse Worksheets

Similar to the PVM analysis form, the SPACS Phoneme Collapse Worksheet organizes the consonants according to manner and then by place within manner categories. The voiceless consonant cognates are listed first. There are two Phoneme Collapse Worksheets: one for word-initial and one for word-final. It is important to use the correct form because not all English consonants are produced in both positions. For example, /ŋ/ occurs word-finally, but not word-initially.

The worksheet also separates the six English manner categories of consonant production into the major class distinctions of obstruent and sonorant:

  • Obstruents—stops, fricatives, and affricates
  • Sonorants—nasals, liquids, and glides

Notice that there are columns along the top that specify “Child” and “Adult.” This represents the mapping of the child:adult sound systems. There are two sets of these columns so that two different phoneme collapses can be completed on a single worksheet.

How to Complete a SPACS Analysis

1. Begin with the word-initial Phoneme Collapse Worksheet.

2. Identify a sound that was used by the child frequently for a number of different target sounds in the word-initial position. Write that sound on the line under the “Child” column.

3. For each target sound and cluster that the child produced with their error, write the target sound (not the child’s error) on the line next to the target sound. For example:

Continue through all the singleton adult targets that were produced by the child with their error. Include the clusters that were produced by the child’s error.

4. Continue diagramming the predominant phoneme collapses identified in word-initial and word-final positions.

5. After you’ve mapped the child’s one-to-many correspondences to the adult sound system, count the number of adult phonemes and clusters that were produced as a single sound by the child, for example, a 1:17 phoneme collapse.

6. Put on your linguistic detective hat to identify the organizing principles the child uses in their language/sound system.

There will be a phonetic resemblance between the phonetic properties of the child’s production in relation to the adult targets. Beginning with the singletons, what is the phonetic resemblance of all the adult target sounds? Do they fall within one manner, such as stops or fricatives? More likely, the adult targets will go across manner categories and fall either into the major class division of obstruents or sonorants. Is there anything else that is similar across the adult target sounds? For instance, are they all voiced obstruents?

Now look at the child’s error substitute. Does it resemble the phonetic characteristics of the adult targets? That is, if the adult targets are voiced obstruents, the child’s error would also be a voiced obstruent.

Next, look at the clusters that were collapsed to the single child error. What is the phonetic resemblance across the adult clusters? Do they all have at least one consonant in common, for example, a stop (/dr/, /gl/, /kw/, /st/)? This will resemble the child’s error as well. For instance, the child’s error production of /g/ is a stop.

A data set and example of a SPACS analysis is provided in the MedBridge course, “Assessment and Differential Diagnoses of Speech Sound Disorders in Children” for a child, “Adam.”

Comparing PVM and SPACS

Like the PVM analysis, SPACS is also a simple and quick analysis to complete. The form provides a nice visual representation of the child’s predominant error patterns.

Unlike the PVM, however, the child’s errors will not be fragmented into a number of different phonological processes. For example, in the course mentioned above, Adam had a 1:18 phoneme collapse of obstruents and clusters to [g]. That one phoneme collapse required five different phonological processes to capture the same error productions: stopping, backing, voicing, deaffrication, and cluster reduction. This highlights the difference between:

(a) the broader system-to-system comparison of SPACS over the narrow sound-to-sound comparison of PVM, and
(b) the child-based analysis of SPACS that does not use a pre-determined and finite number of labels to describe the child’s errors as incorporated in the adult-based PVM analysis.

The visual diagram of phoneme collapses makes it easy to select treatment targets and describe the child’s sound system to parents and teachers. Finally, SPACS not only provides a more economical description of the child’s error patterns (for example, one phoneme collapse compared to five phonological processes), but it also provides an explanatory account for the error patterns. In the example of Adam, he was not able to produce all the target sounds in word-initial position, but he recognized the similarity across sounds and grouped all the adult target obstruents and stop clusters and produced them in his language, i.e., an obstruent stop [g]. What might have looked like idiosyncratic errors (e.g., [g] for /f/) are now clearly accounted for by his phoneme collapse.

It may take some time to hone your linguistic detective skills in identifying the phonetic resemblance between the child and adult sound systems. The forms will help. You may want to brush up on your phonetic skills and refer to the IPA consonant chart for English consonants.

  1. Dodd, B. (2005). Differential Diagnosis and Treatment of Children with Speech Disorder. London: Whurr.
  2. Grunwell, P. (1987). Phonological assessment, evaluation and explanation of speech disorders in children. Clinical Linguistics & Phonetics, 2(3): 221-252.
  3. Skahan, S., Watson, M., & Lof, G. (2007). Speech-language pathologists’ assessment practices for children with suspected speech sound disorders; Results of a national survey. American Journal of Speech Language Pathology,16: 246-259.
  4. Williams, A. L. (2001). Phonological assessment of child speech. In D. M. Ruscello (Ed.), Tests and Measurements in Speech-Language Pathology (pp. 31–76). Woburn, MA: Butterworth-Heinemann.
  5. Williams, A. L. (2005). Assessment, target selection, and intervention: dynamic interactions within a systemic perspective. Topics in Language Disorders, 25(3), 231-242.
  6. Williams, A. L. (2006). A systemic perspective for assessment and intervention: a case study. International Journal of Speech-Language Pathology, 8, 245-256.
  7. Williams, A. L. (2015). Assessment and intervention from a systemic perspective. In C. Bowen (2nd edition), Children's speech sound disorders (pp. 199-203). Oxford: Wiley-Blackwell.