De-identifying name-spelling activities in CHAT

Mon May 18 17:46:39 UTC 2026

Hi Sarah!

We decided internally that any reference to sounds or syllables were
treated more like phonological fragments (&+ne), even if it was an explicit
teaching context with a lexical goal. But if a parent was spelling out the
letters of a word we did use the @k:

*MOT: sat at k that's the word sat .
OR
*MOT: um, I've seen it spelled Zoe at k, and I've seen it spelled Zoi at k .

Again, this was rare for us, so we did not really think through the
implications for our research questions beyond what gets counted as a type
or token...

Janet

On Fri, May 8, 2026 at 4:00 PM Sarah Surrain <sarahsurrain at gmail.com> wrote:

> Thanks, Janet! We will follow your lead and use a pseudonym with the same
> number of letters to capture the letter usage (but use Child_name when the
> full name is spoken).
>
> I like your postcode convention for cloze prompts, but I think what is
> going on here is different. The dad is producing isolated syllables from
> the child's name to scaffold her writing, not to prompt her to complete the
> word. I wonder if the best fit for this is from p. 49 of the current CHAT
> manual: "In Japanese, many letters refer to whole syllables or “kana” such
> as ro or ka. To represent this as well as strings of letters in English,
> use the @k symbol, as in ka at k or jklmn at k."
>
> If we did that, the pseudonimized version would be as follows:
>
> *MA1: v at l.
> *MA1: va at k.
> *MA1: ne at k.
> *MA1: tan rápido?
> *MA1: &=s_sound.
> *CHI: es s at l.
> *MA1: s at l vanes at k +...
> *MA1: s at l.
> *MA1: a at l, Child_name.
>
> Any thoughts on whether this is how the @k is intended to be used?  Or a
> better way to represent the dad pronouncing the /s/ sound to prompt child
> to write the letter S?
>
> -Sarah Surrain
> Assistant Professor of Early Childhood Education
> Early Childhood, Multilingual, and Special Education
> University of Nevada, Las Vegas
>
> On Thursday, May 7, 2026 at 4:41:47 PM UTC-7 janet... at sjsu.edu wrote:
>
>> Hi Sarah,
>>
>> De-identifying a name-spelling activity - This has come up for us too
>> before. The decision for us came down to how we were going to analyze the
>> data. We ultimately decided we wanted to capture the letter-usage, and we
>> were ok with using a pseudonym or replacement letters. For example, for us
>> with a name like 'Hannah', we decided to keep the repetition and use a
>> nonsense set of letters (e.g., abccba - using @l with each letter, so it
>> could retain the number of letters and any repeated letters). Each time
>> they used the child's name we replaced it with something like "child_name".
>>
>> For any syllable teaching or teaching with a partial word, we marked them
>> as a cloze utterance with our own postcode convention (below). This did
>> mean that the partial words did not count for our purposes when we analyzed
>> tokens/types, but it was a small portion of the data so ultimately we let
>> that go but retained it in the transcript and had the postcode in case we
>> wanted to go back to it.
>> *MOT:       &+tur? [+ cloze]
>> *CHI:        &+tle. [+ cloze]
>>
>> When things were in Spanish, we also added the precode when the whole
>> utterance was in Spanish, or we added the @s for each word.
>>
>> Our conventions are here:
>> https://talkbank.org/childes/access/Eng-NA/FMB_home.html.
>>
>> Feel free to email me separately if you want me to share our internal lab
>> manual.
>>
>> Janet
>>
>>
>>
>> On Thu, May 7, 2026 at 2:50 PM Sarah Surrain <sarahs... at gmail.com> wrote:
>>
>>> Hi all,
>>>
>>> I'm working on a new corpus of child language data drawn from randomly
>>> selected 5-minute clips of daylong home recordings. We're transcribing in
>>> ELAN using CHAT conventions. Our de-identification protocol replaces the
>>> child's spoken name with "Child_name" throughout. I have two related
>>> questions that I don't think the CHAT manual addresses directly, and I'd
>>> love the community's input.
>>>
>>> 1.* De-identifying a name-spelling activity:* In this clip, a father
>>> and daughter dyad engage in a name-spelling activity. He guides her through
>>> the letters and sounds of her name as she writes the letters. Transcribing
>>> this verbatim would effectively reveal the child's name through the
>>> sequence of letters and sounds, even if we never write the name itself.
>>>
>>> Here is a pseudonymized version of the exchange (the child's name has
>>> been replaced with "Vanessa"). MA1 = Male Adult 1 (the father):
>>>
>>> Original Spanish:
>>> *MA1: v at l.
>>> *MA1: /va/.
>>> *MA1: /ne/.
>>> *MA1: tan rápido?
>>> *MA1: /s/.
>>> *CHI: es s at l.
>>> *MA1: s at l Vanes +...
>>> *MA1: s at l.
>>> *MA1: a at l, Vanessa.
>>>
>>> English Translation:
>>> *MA1: v at l.
>>> *MA1: /va/.
>>> *MA1: /ne/.
>>> *MA1: so fast?
>>> *MA1: /s/.
>>> *CHI: it's s at l.
>>> *MA1: s at l, Vanes...
>>> *MA1: s at l.
>>> *MA1: a at l, Vanessa.
>>>
>>> I've considered replacing the actual letters and sounds with the
>>> corresponding letters/sounds of a pseudonym, or just using a placeholder
>>> like [Child_name spelled] or &=spells_child_name.
>>>
>>> *2. Transcribing deliberate, isolated letter sounds and syllables*. This
>>> exchange also raises another transcription question. The father produces
>>> isolated phonemes (/s/) and syllables (/va/, /ne/) as deliberate
>>> pedagogical acts — he is teaching his daughter how to spell her name by
>>> modeling its component sounds. This is distinct from letter naming, which
>>> CHAT handles with @l. Is there a CHAT convention for this? (Because it is
>>> in Spanish, we have the added complication that sometimes the letter name
>>> and letter sound are the same, as in the final A of Vanessa.)
>>>
>>> Any guidance on either question would be much appreciated!
>>>
>>> Thanks,
>>> Sarah Surrain
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "chibolts" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to chibolts+u... at googlegroups.com.
>>> To view this discussion visit
>>> https://groups.google.com/d/msgid/chibolts/d895d935-a90e-45c3-ba1e-f1f26d703c12n%40googlegroups.com
>>> <https://groups.google.com/d/msgid/chibolts/d895d935-a90e-45c3-ba1e-f1f26d703c12n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>
>>
>> --
>> Janet Y. Bang, Ph.D (she/her/hers)
>> Assistant Professor
>> Child and Adolescent Development
>> Lurie College of Education, San José State University
>> janet... at sjsu.edu | 408-924-3714 <(408)%20924-3714>
>> https://www.sjsu.edu/education/faculty/janet-bang.php
>>
> --
> You received this message because you are subscribed to the Google Groups
> "chibolts" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to chibolts+unsubscribe at googlegroups.com.
> To view this discussion visit
> https://groups.google.com/d/msgid/chibolts/bb1902dd-12ae-459d-a805-d0e6cb219000n%40googlegroups.com
> <https://groups.google.com/d/msgid/chibolts/bb1902dd-12ae-459d-a805-d0e6cb219000n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
>

-- 
Janet Y. Bang, Ph.D (she/her/hers)
Assistant Professor
Child and Adolescent Development
Lurie College of Education, San José State University
janet.bang at sjsu.edu | 408-924-3714
https://www.sjsu.edu/education/faculty/janet-bang.php

-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/chibolts/CAL7GuZrfioMx8sGKW9N3-fMAhY4uBGp6W3GCy09Mk4Ufxhyq_w%40mail.gmail.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20260518/e09154d3/attachment-0001.htm>