Guidelines for creating a restricted wordlist for cross-linguistic comparison
As the test (target) sounds occur in words, special consideration needs to be given not only to the target sound inventory used, but also to the position and phonetic context of the target sound (Hutters & Henningsson, 2004).
A word list for use in cross-linguistic studies of outcome should be compiled according to the following instructions. It is important to choose words in your own language that fulfils the phonetic requirements but also are represented in young children’s vocabulary. Further, it should be possible to show pictures of the chosen words. One way of creating a word list is to compare the single word lists currently used by the languages below and identify words in the actual language, which include the relevant target sounds and meet the following criteria. They should
- be short (one or two syllables)
- occur in similar phonetic context, that is in strong position (i.e., they occur in a stressed syllable in initial position followed by a vowel)
- include the plosives in initial stressed position and in three different words
- include /s/ in final position in three words and initial position in three words
- include /v, f/ in initial stressed position and each in three different words
- include /n/ in initial stressed position and in three different words
- not include nasal consonants in words with oral target sounds
- not include consonant clusters
- if possible, not include pressure consonants other than the target sound
While the aim is to eliminate the language impact as much as possible when comparing speech outcome after cleft palate treatment, it may not be totally met. Consonant phonemes may not be phonetically identical (which is the reason for narrow phonetic transcription of the target consonants in the restricted word lists) or all selected vulnerable phonemes may not be available in all languages. Consequently, compromises must be made when word lists are developed. For example, some languages do not comprise voiced plosives. If so, an extra set words with the voiceless plosives should be included (see f ex Finnish in the table of word lists on the next page in the left menu).
- include high vowels in 1/3 of the words (the first nine words)
In addition to the target consonants, the first nine words should contain high vowels. These words will thereby be specifically vulnerable for velopharyngeal dysfunction since high vowels require a higher velar position and stronger velopharyngeal force than low vowels for normal resonance. High vowels are therefore more appropriate for assessment of hypernasality. A difficulty to obtain the required velopharyngeal competence in high vowels could by coarticulation influence the articulation of the target consonants. Furthermore, while an overall perceptual rating of velopharyngeal function is included in the ICHOM Standard Set, it does not comprise assessment of hypernasality. However, if such assessment is requested the nine words edited in a string could be used for a valid assessment of hypernasality (Lohmander et al., 2009).
The resulting word list is recommended to include around 30-33 single words.
Put the test consonants and words in random order starting with the first nine words with high vowels (see examples of word lists on the next page in the left menu).
Here is an example of chosen words with the target sound /t/ in five different languages:
Brazilian Portuguese: ‘tatu’
Minimum standard analysis according to ICHOM regarding consonant proficiency in terms of percentage of consonants correct (PCC) based on correct/incorrect production of target sounds (Allori et al., 2017).
Specific analysis based on phonetic transcription, PCC, error categories (cleft speech characteristics, CSCs) including symptoms of velopharyngeal dysfunction (i.e., audible nasal air leakage, reduced pressure on consonants requiring high intra-oral pressure), developmental speech characteristics (DSCs) (Lohmander et al., 2009; Shaw et al., 2019).
* Allori A, Kelley T, Meara J, Albert A, Bonanthaya K, Chapman K, Cunningham M, Daskalagiannakis J, de Gier H,
Guernsey C, Heggie A, Jackson O, Jones Y, Kangesu L, Koudstaal M, Kuchhal R, Lohmander A, Long Jr R, Magee L, Monson L, Rose E, Sitzman T, Taylor J, Thornburn G, van Eeden S, Williams C, Wirthlin J, Wong K. A standard set of outcome measures for the comprehensive appraisal of cleft care. Cleft Palate-Craniofacial Journal, 2017;54:540–554.
* Hutters B, Henningsson G. Speech outcome following treatment in cross-linguistic cleft palate studies: methodologic implications. Cleft Palate-Craniofacial Journal, 2004;41:544–549.
* Lohmander A, Willadsen E, Bowden M, Henningsson G, Persson C, Hutters B. Methodology for Speech Assessment in the Scandcleft Project – an international randomised clinical trial on palatal surgery: experiences from a pilot study. Cleft Palate-Craniofacial Journal, 2009; 46:347-362.
* Shaw W, Semb G, Lohmander A, Persson C, Willadsen E, Clayton-Smith J, Trindade IK, Munro KJ, Gamble C, Harman NH, Conroy EJ, Weichart D, Williamson P. Timing Of Primary Surgery for cleft palate (TOPS): Protocol for a randomised trial of palate surgery at 6 months versus 12 months of age. Br Med J Open, 2019;9 (7) e029780