Two years in the making: How we developed CELF-5 for Australia & New Zealand
It took a research and development team of four, a panel of eight experts, an illustrator, 184 speech language pathologists, and more than 1,000 students to research, test, trial, and design the adaptation of CELF-5. After two years of rigorous research, testing, and trials, the newest edition of the gold standard assessment tool is finally here: say hello to CELF-5A&NZ!
The brilliant authors, Elisabeth Wiig, Eleanor Semel, and Wayne Secord delivered CELF-5 to American shores around four years ago. The tool needed to be adapted and standardised specifically for Australian and New Zealand students because of differences in language and culture.
Read on to learn more about the research process that went into the adaptation and standardisation of the language assessment tool, and why it’s important to have our own version.
Why we adapted CELF for Australia and New Zealand
There are two main reasons why adaptation is important:
1. There are significant differences across culture and language between America, and Australia and New Zealand
CELF-5 asks students to describe images and to read scenarios. This content can include references to sports, animals, and school environments (among other things).
A student in America, for example, would generally recognise a squirrel or a baseball game but these are not as familiar in Australia or New Zealand. Adapting content might mean changing the squirrel to a possum or the baseball game to cricket.
And it’s not only images - word references and language used to describe actions differ too.
An American student would generally be familiar with the phrase ‘’turn in your paper’’, while in Australia and New Zealand we tend to say ‘’hand in your homework’’.
If students don’t recognise situations they’re asked to describe during the assessment, they may score lower against normative data, leading to a skewed interpretation of results.
And speaking of the normative data...
2. Population demographics differ between the three countries, which affects the normative data used for scoring students
It’s important the normative data accurately represents a cross-section of the Australian and New Zealand population.
During the research phase of the standardisation, we collected sample data based on the latest census results. We used that data to create an average bell curve for scoring students.
Our research and development team collaborated with many organisations including education and health departments, universities, and private practices across Australia and New Zealand to collect data used to standardise the CELF-5A&NZ.
Guidelines used to inform the CELF-5 adaptation
We used the International Test Commission (ITC) guidelines as the basis for our changes. These guidelines are in place to make sure that adaptation of a test stays true to its original purpose.
Interested in learning more about the steps? Read through ITC Guidelines for Translating and Adapting Tests.
The research process for standardising CELF-5A&NZ
Here’s how we brought CELF-5 to Australia and New Zealand:
1. The latest edition of CELF is released in America
Elisabeth Wiig, Eleanor Semel, and Wayne Secord developed CELF-5, and released it to the American market in 2013.
2. The Australian R&D team receives the American edition
Two years later, in 2015, the research and development team within Pearson Clinical began work on the review and adaptation.
The team of four research specialists initially flagged sections and questions which were culturally loaded and required adaptation for Australia and New Zealand.
3. A panel of experts are selected to evaluate the test questions
These experts had different backgrounds — from academia, to government, and private practice — and were from various locations across the two countries.
This panel then worked through the American version and voted to keep or discard questions. For a question to remain on the test, the majority of the panel had to be in agreement.
The panel had to give careful consideration to each question, as language and cultural context can vary greatly depending on a student’s background.
Specifically, the panel took into consideration:
differences in language and culture between Australia and New Zealand,
differences in metropolitan and rural regions, and
indigenous cultural differences (Aboriginal and Torres Strait Islanders in Australia, and M?ori in New Zealand).
4. Recommendations are assessed and modifications are made to images and content as needed
The research and development team then took these results and recommendations, redrafting and re-designing as needed. This included:
Editing and updating the language of certain questions
Removing questions that weren’t able to be appropriately adapted
Creating new questions to replace removed questions
Updating images and contexts, while being mindful to keep the original style and characteristics
5. The draft adaptation of the assessment is handed to examiners to collect standardisation data with a sample of the population
For the standardisation of the CELF-5A&NZ we worked with 184 speech language pathologists — known as examiners. It was important to engage with clinicians for the data collection to ensure the assessment was undertaken correctly and in line with the standards of the profession. Using the draft version, examiners collected data with a sample of more than 1,000 students.
The research and development team selected the cross-section sample of students based on population information contained in the latest census data from Australia and New Zealand. It was important that this sample reflected the diversity of both countries, as the test results taken from these trials make up the normative data that students are scored against in the final edition of CELF-5A&NZ.
Image: representation of cross-section sample across the various regions in Australia and New Zealand.
To cover the full range of language ability, most of the students selected in this sample are considered to be ‘’typically developing’’, while a smaller percentage either had a language disorder or had superior language skills.
6. The draft version is revised based on the standardisation data
The research and development team, with input from expert psychometricians, analysed the data post-standardisation to ensure there were no biased or culturally inappropriate questions. All items were found to be appropriate and no further changes were needed.
7. The final version for Australia and New Zealand is complete
Two years of research and development later, CELF-5A&NZ is complete and ready for Speechies to start using! We are sincerely grateful to the expert panel, the examiners, and all the children and their families who participated in the development of the CELF-5A&NZ. We couldn’t have brought the tool to Australia and New Zealand without all your help!
Are you ready to upgrade from CELF-4 Australian to CELF-5A&NZ? Purchase the latest edition.
Have a further question about the research and development that went into this adaptation? Contact us via phone or email.