JCPSLP Vol 17 No 1 2015_lores

for back translation by translators who were accredited by the National Accreditation Authority for Translators and Interpreters (NAATI). The two translations were checked and occasionally the original translators were asked to clarify wording to ensure consistency. Some of the differences have been as a result of discrepancies between words that were accessible to parents versus words that would be correct in an academic context. For example, two versions of the title were possible for the Danish translation: “Vurdering af barnets forståelighed i sine omgivelser” which was easier for parents to understand and “Vurdering af forståelighed i kontekst” which contained the more academic word “kontekst” (context). Eventually, the more parent-friendly translation was chosen and the direct translation is: “Evaluation/assessment of the child’s intelligibility in his/her surroundings.” Some languages that are commonly spoken in Australia and other parts of the world had no available speech pathologists or linguists who could translate the ICS. So the second way that translations were undertaken was by translators from Australian Multi Lingual Services. Most of the translators were NAATI accredited (e.g., for Gujarati, Hindi, Hmong, Karen, Khmer, Polish, and Serbian). However, in a few cases the translators were not accredited (e.g., for Somali and Tongan), so the third method of translation was by non-accredited translators employed by Australian Multi Lingual Services. The final method of translation was undertaken when there were available speech pathologists/linguists, but no available translators at the Australian translation company (e.g., for Irish, Jamaican, Sesotho, Tshivenda, isiXhosa, and isiZulu). In these instances, translations were undertaken (and back translated) by colleagues and other speakers of the language as organized by the translators. For example, in South Africa, Dr Michelle Pascoe organised translations of the ICS in Afrikaans, Sesotho, Tshivenda, isiXhosa, and isiZulu by working with groups of speech pathology students studying at the University of Cape Town to translate, back translate, and check the ICS translations with community members. They used protocols for forward and backward translation of health-related materials from the World Health Organisation (2012). Their research is continuing so that eventually translations will be available in the remaining official languages of South Africa (i.e., isiNdebele, Sepedi, Setswana, SiSwati, and Xitsonga). Research using the ICS in languages other than English To date, the ICS has been used with children with typically developing speech in Slovenia (Kogovšek & Ozbiˇc, 2013), Sweden (Lagerberg, 2013), Hong Kong (Ng, To & McLeod, 2014) and Croatia (Tomi´c & Mildner, 2014). It has been used with typically developing multilingual children who speak Korean and English in New Zealand (Kim, Ballard & McCann, 2014). It has been used with children with speech sound disorders who speak Cantonese in Hong Kong (Ng et al., 2014) and with children with cochlear implants in Iceland (Thoroddsen, 2014). Validiation studies have been undertaken in Traditional Chinese/Cantonese (Ng et al., 2014), Slovenian (Kogovšek & Ozbiˇc, 2013), and Croatian (Tomi´c & Mildner, 2014). For example, Ng et al. (2014) validated the Traditional Chinese version of the ICS in Hong Kong with 72 Cantonese-speaking preschoolers (33 typically developing and 39 with speech sound disorders).

Articulation and Phonology (Dodd, Zhu, Crosbie, Holm & Ozanne, 2002), establishing criterion validity. Subsequently, the ICS was validated and normed on 804 different Australian preschool-aged children (McLeod, Crowe, & Shahaeian, 2014). Each of the children spoke English and in addition, 36.9% spoke at least one of 59 other languages. The mean ICS score for the 804 children was 4.4 ( SD = 0.7). Significant differences in scores were identified based on sex and age, but not socioeconomic

status or the number of languages spoken. There were significant differences in ICS scores between the group of children whose parents had concerns about their child’s speech ( M = 3.9) and those who did not ( M = 4.6), establishing criterion validity. The ICS had high internal consistency and satisfactory test-retest reliability. Sensitivity of .82 and specificity of .58 was established as the optimal cut-off. In another study, McLeod, Harrison, McAllister, and McCormack (2013) studied 109 children with speech sound disorders and found that there was a significant difference between ICS scores for those who had and had not attended speech pathology services. In each of these three studies, the children’s speech was most intelligible to their parent, then their immediate family, and was least intelligible to strangers. Translation of the ICS Over the past three years, the ICS has been translated into 60 languages (see Table 1 and Appendix) with more translations being added regularly. There have been four ways that translations were undertaken. First, most of the translations were undertaken by speech pathologists and linguists in different countries throughout the world. Typically these people worked in university and clinical settings and regularly worked with children with speech sound disorders who spoke the language used in the translation. Their translations were sent to an Australian translation company Translators of the Intelligibility in Context Scale. L:R – Dr Karla Washington (Jamaican), Dr Dana Buntová (Slovak), Professor Martin Ball (Welsh), Professor Vesna Mildner (Croatian), Professor Sharynne McLeod (English)

8

JCPSLP Volume 17, Number 1 2015

Journal of Clinical Practice in Speech-Language Pathology