Speech Articulation Survey

About the Speech Articulation Survey

  • The Articulation Survey was developed at the Royal Children's Hospital (RCH) speech pathology department by Neil Atkin and John Fisher (1996) in order to determine the sounds in a child's repertoire in an efficient and uncomplicated manner.

    Description

    A simple, quick way of describing and scoring a child's ability to say sounds in words. The test is easily administered and does not require special training, therefore non-speech pathologists can use the test.

    The child names 88 pictures to make 120 target sounds. A score for the number of correct target sounds can be compared with scores for children in the 3;5 to 7;11 age range.

    Purpose

    • screening children's speech sounds
    • providing an articulation score for research data
    • quantifying a change in articulation over time
    • comparing a child's score against a reference group score

    Contains

    • 7 laminated pages of pictures on A4 sheets 304 x 218mm
    • 13-page manual with response sheets and instructions
    • presentation folder 312 x 220mm

    Features

    • quick - 10-15 minutes to administer and score
    • comprehensive - 24 single consonants and 27 consonant blends (120 target sounds)
    • durable - laminated picture sheets
    • noticeable - the bright red folder with black stripe is easily found
    • widely used - more than 500 copies sold in Australia
    • referenced against 299 children with normal speech
    • data for five one-year age groups from 3;5 to 7;11
    • Reference Group Data are on the Articulation Survey webpage

    The Speech Articulation Survey

    Items

    Single consonants are elicited by naming pictures that have one target sound per word at the start, middle or end of words. All the consonant blends are at the start of words and have optional prompts. Most of the consonant blends have two target sounds; there are five blends with three target sounds. Items that least consistently elicit the target word have alternative pictures - matches-witches, television-treasure, oranges-fire engine.

    Scores for the Speech Articulation Survey

    Raw score

    The raw score is the number of correct target sounds for

    Single consonants 61 items
    Consonant blends 59 items
    Total sounds 120 items

    Raw scores can be used to compare a child's performance on two or more test occasions. Reference group data for mean raw scores by age groups (Table 2) allow the child's performance to be compared with age peers using means and standard deviations.

    Percentage correct of the target sounds

    Raw scores can be expressed at a percentage ie number of correct sounds x 100/ the total possible correct.

    • % single sounds correct x 100/61
    • % blend sounds correct x 100/59
    • % total sounds correct x 100/120

    Percentages allow a comparison between the correctness of single target sounds and blend target sounds. Percentages also allow a comparison of a child's performance on two or more test occasions . Expected percentage correct for an age group can be calculated using the above formulae with the Reference Group raw score data in Table 2.

    Standard scores - percentile rank and standard score

    Raw scores can be converted to percentile rank and age-standardised scores with a mean of 100 and a standard deviation of 15. Tables 3a, 3b, 3c give the conversion of raw scores to percentile rank and standard score equivalents at one, two and three standard deviations above and below the mean. Tables 4a, 4b, 4c give a more detailed conversion of raw scores to age-standardised scores with values extrapolated from the obtained Reference Group scores.

    Reference

    Atkin N.& Fisher J., (1996) Articulation Survey, Royal Children's Hospital, Melbourne