JCPSLP Vol 14 No 1 2012

for each section and for 15 out of 18 ALDeQ questions. The questionnaire had good specificity (96%) in the classification of ELL with typical language, but lower sensitivity (66%) in the classification of ELL clinically identified as language impaired. The ALDeQ was found to be a useful tool to obtain L1 information, particularly when direct L1 examination was not possible. The aim of the current study was to investigate use of the ALDeQ within an Australian ELL population. The hypotheses were: Hypothesis 1. ALDeQ scores for typically developing Australian ELL would be similar to the Canadian norming sample. Hypothesis 2. ALDeQ Total Scores would differentiate between typically developing ELL and ELL with language difficulty, with lower scores for the language difficulty group. Method Participants Five Perth metropolitan primary schools thought to have a high ELL population were approached for involvement in the study and two schools expressed interest in the research. High ELL populations were identified through the Schools Online website, and indicated by high percentages of ELL enrolments, and ELL specialist programs at the school (Department of Education, 2010). All 36 pre primary and year 1 ELL students were sent consent forms and those who met the selection criteria and agreed to participate in the study were selected (N = 17). Children with primary speech difficulties in the absence of language difficulties were excluded from the study. Children with pre-existing cognitive, neurological, psychological, or sensory impairment, as identified through parent report, were excluded from the study. Participants were 17 parents or caregivers of ELL recruited from these two schools. The Child Language and Medical Questionnaire (see Appendix B) was completed by children’s parents to collect background information. Parents of three children expressed concern about their child’s language development. Two children were reported to have been clinically identified with a language difficulty and one child was reported to have a current referral to language therapy within this group. The remaining 14 parents expressed no concerns about their child’s language development (question 12). Parent report allowed children to be placed in two groups – possible language difficulty (those who were reported to have been clinically identified with a language difficulty or had a current referral to language therapy), and those for whom no concerns regarding language development were reported. Children were aged between 5 years, 3 months and 8 years, 7 months (M = 79.38 months). Parent report revealed that the majority of children (n = 16) were sequential ELL and spoke Vietnamese, Romanian, or Macedonian as a L1, while one child was a simultaneous ELL. Prior to preschool or daycare entry at age 2 to 3, children’s English language exposure ranged from none to moderate. See Table 1 for a summary. Procedure Data were collected by the primary researcher (speech pathologist) and one Vietnamese interpreter using parent and caregiver report (parents n = 16, caregiver n = 1). Parents will be used to refer to all interviewees. Interviews took place at the family home ( n = 9) or by telephone ( n = 8), according to parent preference. The language of the

misinterpreted (Guiberson, Barrett, Jancosek, & Yoshinga Itano, 2006). These varying proficiency levels highlight the need for the speech pathologist to consider both languages of the ELL. One possible solution to the assessment in both L1 and L2 is use of non-standardised assessment of L1 language ability, which includes parent report. Research has shown parent report of a child’s current communicative behaviours to be valid and to decrease context sensitivity by gaining information on the child over longer time periods, within a range of communicative contexts (Marchman & Martinez- Sussmann, 2002; Ring & Fenson, 2000). Furthermore, parent report is less time-consuming and cheaper than extensive behavioural assessments (Marchmann & Martinez-Sussmann, 2002). Research using assessments such as the MacArthur Communicative Development Inventory (CDI; Fenson et al., 1994) and the Language Development Survey (LDS; Rescorla, 1989) has provided reliable and valid parent reports of children’s early language skills. Research on retrospective parent report is limited, but there is some evidence that it is useful in identifying conditions such as autism spectrum disorder (ASD). Goldberg, Thorsen, Osann, and Spence (2008) found substantial concordance between videotape observation and parent report of expressive language onset and loss with typically developing children and children with ASD. Sivberg (2003) used retrospective parent report to describe early symptoms of ASD and atypical development for children. The results showed consistency between similar research reports of early ASD symptoms using direct observation and parent report. These two studies suggest retrospective parent report may be a valid and reliable means of assessment, and may be applicable for language assessment. The ALDeQ (Alberta Language Development Questionnaire) is a norm referenced questionnaire that uses retrospective parent report to obtain information on the L1 of ELL (Paradis, Emmerzael, & Duncan, 2010). The ALDeQ aims to reduce difficulties in the assessment of L1 through use of non-culture specific information obtained from parents. It comprises 18 questions organised in four subsections: A) Early Milestones, B) Current L1 Abilities, C) Behaviour Patterns and Activity Preferences, and D) Family History (see Appendix A for example questions). Questions are given a numerical score using rating scales and added to form a subsection score total. Numerator and denominator scores are added from each section to obtain a proportion (0–1) which forms the ALDeQ Total Score (Paradis et al., 2010). The ALDeQ Total Score is compared to norming sample characteristics where scores of –1.25 standard deviations or lower indicate a language profile more consistent with that of a child with language impairment than typical development. Paradis and colleagues’ (2010) preliminary Canadian study used the ALDeQ with 129 typically developing ELL and 29 ELL with language impairment aged 5 to 7 years. Children were identified as language impaired through face-to-face assessment and had a range of language backgrounds: Chinese, South Asian, Arabic, and European-Romance. There were no significant ALDeQ Total Score differences between cultural groups (Paradis et al., 2010). Findings revealed the ALDeQ to be a significant discriminator between the typically developing and language impaired groups. There were significant between-group differences

26

JCPSLP Volume 14, Number 1 2012

Journal of Clinical Practice in Speech-Language Pathology