SALT Reference Databases
Compare your transcript to a database of normative language measures. The individual may be matched on age, grade, gender, sampling context, and transcript length. Matched records are selected from the database and mean, range and standard deviation statistics are given for many of the analysis variables.
Databases of English-Fluent Speakers
• Play Database
The Play database consists of play-based conversational samples from 69 typically developing children, ages 2;8 - 5;8, grades Pre-K and K,located in Wisconsin. These play-based samples were elicited between the child and an examiner. Age, grade, and gender data is available for all participants.
• Conversation Database
The Conversation database consists of conversational samples from 584 typically developing students, ages 2;9 - 13;3, grades Pre-K, K, 1, 2, 3, 5, and 7, located in Wisconsin and California. Conversation samples were elicited between the student and an examiner using a defined set of conversational topics including classroom activities, holidays, family activities and family pets. There are students from a variety of economic backgrounds and ability levels. Age, grade, and gender data is available for all participants.
• Narrative SSS Database
The Narrative SSS database (student selects subject) database consists of narrative samples from 330 typically developing students, ages 5;2 - 13;3, grades K, 1, 2, 3, 5, and 7, drawn from preschools in Madison, Wisconsin or from the Madison Metropolitan Public School System. Additional 3, 5, and 7-year-old subjects were drawn from rural areas in northern Wisconsin . There are children from a variety of economic backgrounds and ability levels. "Typically developing" was determined by normal progress in school and absence of special education services. Economic background was determined by eligibility for the free lunch program. Ability level was determined by teacher rating. Age, grade, and gender data is available for all participants. Participants were asked to retell a favorite movie or game.
• Narrative Story Retell Database
The Narrative Story Retell database database contains samples from 500 typically developing English-fluent children from preschool through 6th grade, ages 4;4 - 12;8, located in Wisconsin and California. Age, gender, and grade data is available for all participants. The examiner read or told the story to the student while they looked at the book together. Then the examiner gave the book to the student and asked him/her to retell the story using the book. Different stories were used depending on grade in school.
• Expository Database
The Expository database consists of expository samples from 242 typically developing English-fluent students, ages 10;7 - 15;9. These middle and high school students are from public schools located in Wisconsin. The samples were elicited following a defined script which asked the student to explain a game or sport of his/her choosing. The protocol included giving the student a few minutes to take notes on a planning sheet which contained the following eight topics: object, preparations, start, course of play, rules, scoring, duration, and strategies. The student was then asked to explain the game or sport using the planning sheet. Age, gender, and grade data is available for all participants. The expository project was presented at the ASHA 2008 convention (click here for a summary of the presentation) and was written up in the Feb 23, 2009 edition of the Advance magazine (click here to go to the Advance web site).
Databases of Bilingual (Spanish/English) Speakers
• Bilingual Spanish/English Story Retell Databases
The Bilingual Spanish/English Story Retell databases consist of English and Spanish story-retell narratives from over 2,000 native Spanish-speaking bilingual (Spanish/English) children. These English language learners (ELLs) were drawn from public school ELL classrooms in urban Texas (Houston and Austin), border Texas (Brownsville), and urban California (Los Angeles). The children reflect the diverse socio-economic status of these areas. Age, grade, and gender data is available for all children and mother's education is available for many. Samples were elicited from each child in both Spanish and English. Using one of four wordless frog story books by Mercer Mayer, the examiner told the story to the child in the target language while they looked at the book together. Then the examiner gave the book to the child and asked him/her to retell the story using the book.
• Bilingual Spanish/English Unique Story Databases
The Bilingual Spanish/English Unique Story databases consist of English and Spanish story tell (not retell) narratives from 475 native Spanish-speaking bilingual (Spanish/English) children. These English language learners (ELLs) were drawn from public school ELL classrooms in urban Texas (Houston and Austin), border Texas (Brownsville), and urban California (Los Angeles). The children reflect the diverse socio-economic status of these areas. Age, grade, and gender data is available for all children and mother's education is available for many. Samples were elicited from each child in both Spanish and English. Using the wordless picture book One Frog Too Many by Mercer Mayer, the examiner silently looked through the book with the child. Then the examiner gave the book to the child and asked him/her to tell the story using the book.
Databases Contributed by Colleagues
• ENNI Database
The ENNI database consists of 377 samples from children ages 3;11 - 10;0. These samples were collected using the Edmonton Narrative Norms Instrument (ENNI), an assessment tool for collecting language information from children aged 4 to 10 through storytelling. Pictures that portray a story are presented to a child, who then tells the story to the examiner. Picture sets were drawn for the ENNI by a professional cartoonist; they range from a simple story with 2 characters to a complex story with 4 characters. These pictures and detailed administration and transcription instructions may be downloaded from the ENNI web site at www.rehabmed.ualberta.ca/spa/enni. The ENNI is copyrighted, including the pictures and all other materials. You are welcome to download, print, and use any of the materials for clinical, educational, or research purposes. None of the ENNI materials may be altered in any way or included in publications without permission from the authors.
• Gillam Narrative Tasks Database
The Gillam Narrative Tasks database consists of narrative samples from 500 children, ranging in age from 5;0 to 11;11, including 50 five-year olds, 100 six-year olds, 100 seven-year-olds, 100 eight-year-olds, 50 nine-year-olds, 50 ten-year-olds, and 50 eleven-year-olds. Children came from four US regions (Northeast, South, Midwest , and West). Their primary language was English and they had not been identified with a disability and were not receiving any special education services. Examiners collected data on children's ability to tell stories in three formats: (a) a script retell (no picture cues), (b) a story about five sequenced pictures, and (c) a fictional narrative based on a single picture. The examiner scripts and picture stimuli which were used to elicit the narratives are available in the Test of Narrative Language (Gillam & Pearson, 2004).
• New Zealand Reference Databases
The New Zealand Conversation database contains 248 spoken language samples collected from New Zealand children aged 4;5 – 7;7. The language samples were collected from the participants in a conversational context. The children were randomly selected from schools in Auckland, Hamilton, Christchurch (major urban areas in New Zealand) as well as secondary urban areas surrounding Christchurch. Approximately 80% of the participants were from the Auckland/Hamilton region to reflect New Zealand's population density in these areas. Children with diagnosed disabilities were excluded from the sample. The schools reflected a range of socio-economic areas and English was the first language of all children included in the database. The protocol aimed to elicit 50 complete and intelligible utterances from the child in 10 minutes of conversation. The protocol was adapted from interview procedures described by Evans and Craig (1992). The child was asked to bring an object from the classroom to discuss with the examiner. The examiner encouraged the child to talk about the object. The child was then asked to talk about his or her family, school, and after-school activities. The Westerveld & Gillon Language Sampling Protocol guide, by Marleen Westerveld and Gail Gillon, contains additional information on the protocol used to elicit these samples.
The New Zealand Story Retell database contains 264 spoken language samples collected from New Zealand children aged 4;0 – 7;7. The language samples were collected from the participants in a story retelling context using a story format and vocabulary that is familiar to children in New Zealand. The initial data were collected in 2000/2001 from 188 children, ages 4;6 to 7;7, who had been randomly selected from kindergartens and schools in Auckland, Hamilton, Christchurch (major urban areas in New Zealand) as well as secondary urban areas surrounding Christchurch. Approximately 80% of the participants were from the Auckland/Hamilton region to reflect New Zealand's population density in these areas. Children with diagnosed disabilities were excluded from the sample. The schools reflected a range of socio-economic areas and English was the first language of all children included in the database. A second set of data was collected in November 2009 from 76 children aged 4;0 to 4;11. All children attended their local kindergarten in Christchurch, New Zealand. The kindergartens reflected a range of socio-economic areas and English was the first language of all children. For this protocol, the child was required to listen to two audio-recordings of an unfamiliar story (while looking at pictures in the story book). Following the first listening of the story, the child was asked eight questions about the story, to evaluate story comprehension. All children were provided with the correct information after answering the questions. Following the second listening of the story the child was asked to retell the story without the use of the pictures. The Westerveld & Gillon Language Sampling Protocol guide, by Marleen Westerveld and Gail Gillon, contains additional information on the protocol used to elicit these samples.
The New Zealand Personal Narrative database contains spoken language samples collected from 228 New Zealand children, ages 4;5 – 7;7. The children were randomly selected from schools in Auckland, Hamilton, Christchurch (major urban areas in New Zealand) as well as secondary urban areas surrounding Christchurch. Approximately 80% of the participants were from the Auckland/Hamilton region to reflect New Zealand's population density in these areas. Children with diagnosed disabilities were excluded from the sample. The schools reflected a range of socio-economic areas and English was the first language of all children included in the database. The protocol was adapted from a conversational technique developed by Peterson and McCabe (1983), called the Conversational Map. In adapting this technique, the examiner related a brief personal experience related to a photo prompt in order to encourage the child to share one of his or her personal experiences. A pocketsize photo album with a series of carefully selected photos was used for the stimulus items. Each photo was presented individually in separate sleeves of the photo album. The examiner provided a short prompting narrative with each photo followed by the question: "Did anything like that ever happen to you?" If the child responded "no", the examiner turned the page of the photo album to the next photo. If the child responded "yes" a follow-up question was asked, "Can you tell me about it?" The Westerveld & Gillon Language Sampling Protocol guide, by Marleen Westerveld and Gail Gillon, contains additional information on the protocol used to elicit these samples. The photos that were used to collect the personal narrative language samples are available for download.The file containing the photos is a self-extracting zip file containing the eleven photos. Run this file and you are prompted to select the folder where you want the photos stored. Browse to select this folder and then select Unzip.
The New Zealand Expository database contains contains samples from 65 six- and seven-year-old participants, ages 6;1 – 7;11, who were recruited from three primary schools located in suburban Auckland, New Zealand (NZ). The schools were awarded mid socio-economic status based on the Ministry of Education ranking system. These children had no known history of hearing disorder, neurological disorder, or speech-language therapy, spoke English as their first language, and were progressing normally at school. Expository language generation samples were elicited using the Favorite Game or Sports (FGS) task, developed by Nippold, Hesketh, Duthie, and Mansfield (2005). In this task, the examiner carefully follows script. First, the child is asked what his or her favorite game or sport is and why. The examiner then asks the child to explain the game or sport, using the pragmatically felicitous prompt "I am not too familiar with the game of [..]". Finally, the child is asked what a player should do to win a game of [..]. For specific information regarding the protocol, see Nippold et al. (2005). The child should be allowed as much time as necessary to finish the explanation. The examiner needs to make sure to show interest in the child's explanation and only use neutral responses as needed to encourage the child to continue.