SALT Spanish and Bilingual Transcripts

General Information

Acknowledgements: The implementation of Spanish transcription using SALT Software is a result of the collaboration with Aquiles Iglesias, Raúl Rojas and Brian Goldstein from Temple University. We would like to express our appreciation to them for their insights into the many issues involved in analyzing Spanish transcripts and their willingness to work with us to add transcription conventions and analyses designed specifically for Spanish and bilingual transcripts. The work at Temple University has been partially funded by research contract N01-DC-8-2100 from the National Institute on Deafness and Other Communication Disorders, National Institute of Health.

There isn't a separate "Spanish SALT" program. SALT for Windows® may be used with English, Spanish, and bilingual language samples.

Because of the increasing number of Spanish and bilingual speakers in the United States, we were motivated to expand the SALT program to handle non-English language samples. Working with researchers from Temple University (see acknowledgements), it quickly became apparent that it was not enough to simply allow the Spanish character set to be used. Specific transcript-entry conventions and analyses were needed. In developing the transcript-entry/coding system for Spanish and bilingual transcripts, the following important factors were considered in order to respect non-biased assessment:

  • Dialectal variations exhibited in phonology, morphology, syntax and pragmatics;
  • English influence depending on different degrees of exposure and contact with English-speakers;
  • Language loss.

The following sections describe some of the issues, specific to Spanish and bilingual language samples, that have been addressed. Also, view a Bilingual Case Study.

Spanish Word Lists

There are fifteen standard word lists built into the SALT Software (e.g., questions, conjunctions, negatives, personal pronouns). There are two different sets of standard word lists built into the program: one for English transcripts and one for Spanish transcripts. The "language setting" determines which word lists to use.

Bound Pronominal Clitics

Spanish has great word order flexibility, and is not as dependent on it as English. One syntactical phenomenon that we were interested in marking to control for possible MLU (mean length of utterance) inflation and dialectal variation, was the use of pronominal clitics in Spanish. Pronominal clitics must occur with a verb because they are directly verb related as direct or indirect objects. They can be located preceding the verb as a proclitic, or positioned after the verb as an enclitic. Pronominal clitics can move within an utterance, changing the form but not necessarily the content of the utterance. It is precisely these morphemes in Spanish that can occur as free- or as bound morphemes. The freedom of movement that pronominal clitics possess is an important aspect in individual differences across language development and dialect. For example, a child could say:

  1. C me lo das.    Give it to me.

or instead he/she, or another child, could choose to say:

  1. C damelo.    Give it to me.

Regardless of which utterance the child produces, the content of the two utterances is the same. Due to strict rules of spelling convention, (a) is written as three separate words, and (b) is written as one word. Herein lies the temptation to assign three words and three morphemes to (a), and only one word of one morpheme to (b). However, it is important to remember that SALT transcription analysis is based on oral language, not on written language. To address this problem, SALT added a "clitic" marker, the plus sign, to mark bound pronominal clitics. Utterances (a) and (b) are transcribed in SALT in the following manner:

  1. C me lo das.    Give it to me.
  2. C da+me+lo.    Give it to me.

The "+" symbol indicates the use of bound pronominal clitics. Thus, utterances (a) and (b), which include clitics, are given equal weight in the analysis. Both utterances possess the same verb and object pronouns, having equal morphological value.

Tables have been added to SALT that summarize the use of clitics in the language sample.

Root Identification

The highly inflected morphology of Spanish can significantly affect the post-inflected root word/stem. The word root identification convention, vertical bar "|", was developed in order to credit Spanish-speakers for exhibiting use of morphological forms as well as to avoid over-inflation of the number of different words (NDW) used. If a child produces a variety of inflected forms of the same word within a transcript (e.g., "es", "son", "eran", "éramos"), each production is coded to identify the root word (e.g., "es|ser"). The child would be given credit morphologically for producing different words, but each inflected form would be considered an inflected variation of the same root word, "ser". For example:

C Había|haber una vez un niño que tenía|tener una rana.

Root identification instructs SALT to consider the word immediately preceding the "|" symbol as the word that was actually said (i.e., "había" and "tenía"), and the word immediately following to be the root word (i.e., "haber" and "tener").

Automating Root Identification

To simplify transcription, lookup files of words with their corresponding root forms are available. These root identification files (RIFs) are used to automatically identify a different word root than the one that was produced.

Diminutives

Diminutives are bound morphemes intended to indicate diminution. However, the use of diminutives may be so pervasive in some Spanish dialects, that the morphological inflection of the diminutive may no longer indicate diminution Thus we decided not to count diminutives (i.e., perrito) as bound morphemes, to prevent potential MLU (mean length of utterance) inflation for speakers who use a high frequency of diminutives without strictly indicating diminution. Instead, we mark diminutives as separate main body words derived from the corresponding root words. For example:

C Mira los perritos|perro/s.    Look at the dogs.

Known Spanish diminutives: -ete, -eta, -ico, -ica, -ito, -ita, -illo, -illa, -uco, -uca, -ucho, -ucha, -uelo, -uela

SALT Demo

Try SALT 2008 before you buy.

SALT Brochure

Request or download our product brochure.

SALT Databases

Review the reference databases available with SALT.

Case Studies

See how SALT can be used for assessment.

Adobe Reader

Download Adobe Reader®, which is needed to view the case studies and other documents.