There are 4 columns associated with child communication codes: id for communication, child vocalization column, child utterance type column, and a comments column for communication.
Column name: com_id
Code: (<lab_id>, <com_coder>, <childcom_date>, <childcom_mins>, <relchildcom_coder>, <relchildcom_date>, <momcom_date>, <momcom_mins>, <relmomcom_coder>, <relmomcom_date>)
Code the com_id_column according to the definitions for coding the id column.
Column name: childutterancetype
Code: (<language_pw>, <langlike_bv>,<crylaughgrunt_clg>, <unintell_x>)
You will receive video files that have been transcribed by the PLAY team. These transcriptions have been divided into two columns: childvoc and momspeech. Cells in these columns include the transcribed content of each speaker’s utterances and vocalizations. For the Child communication pass, the focus is on the childvoc column. Each utterance and vocalization in the childvoc column will have it’s own corresponding, time matched cell, in the childutterancetype column.
A childvoc cell can contain full words or codes that denote vocalizations by the child that are not words.
The code ‘b’ in a cell in the childvoc column denotes instances in the transcript where the child emitted a babble or a vowel-consonant sound, or a vocalization where the transcriber was unsure if the child produced a word or a babble and was conservative and coded the utterance as a babble.
The code ‘v’ in a cell in the childvoc column denotes instances in the transcript where the child produced a vocalization such as crying, screaming, grunting, cooing, consonant alone sounds, or an audible gasp.
The code xxx in a cell in the childvoc column denotes instances in the transcript where parts of a child’s utterance were unintelligible or due to ambient noise or the child whispering the transcriber could not make out the child’s vocalization.
These codes can help you code for communication in the childutterancetype column. However, do not solely rely on the transcript contents to code for communication as the operational definitions for codes used during transcription ARE NOT the same as the definitions for communication coding.
In the childutterancetype column, coders categorize the utterances in the childvoc column as a specific type of speech form. Read the utterance transcribed in the childvoc column and categorize each utterance based on codes detailed under Value List below.
Codes are mutually exclusive. The prompts/arguments in the code are designed to speed the coder through the easiest to detect and easiest to code categories (language, language-like sounds, etc.) down through the more nuanced and time-consuming codes. Once the proper code has been found, enter it into the prompt you are at, then code all of the rest as periods ”.”. For instance, if the child didn’t speak in full speech, or speech-like sound, but did cry/scream, then code <.,.,c,.,.>.
Do not rely solely on the transcript.. Always watch and listen to the video as you read the transcript because tone, context, and other factors can help when coding.
After completing the coding pass for the entire video, coders should rewatch the first 5 minutes in order to recheck their codes to ensure that they did not miscode any vocalizations. Coders must do this because there are individual differences in children’s vocalizations and intonation, which coders become more attuned to after extensive experience coding for that child.
p = phrase or proto sentence
w = word
b = babble
v = vowel
c = cry
l = laugh
g = grunt
x = unintelligible
Phrase or proto sentence (p)
An utterance in which the infant combined two or more words in a single phrase. These can take many forms to express many relations between words, such as actor-action (e.g. “Daddy run”), actor- object of action (e.g., “Baby spoon”), possession (e.g., “Mommy shoe”, “my turn”), recurrence request (e.g. “More cookie”), noun-adjective (e.g., “big yard”). Generally, utterances that relay more than one verbal concept can be coded as a ‘p’. For example “uhoh down” includes two concepts where uhoh is one concept and down is another concept.
The following are exceptions where an utterance may have more than 1 word, but are not considered a phrase: a proper noun where a name is composed of two words (e.g., “Cookie Monster”), a common expression that is composed of two words (e.g., “clean up”, “bye bye”, “oh no”, “uhoh”), an article and a noun (e.g., “a drawing”), or any utterance where the child repeats the same word more than once (e.g., “mama mama mama”). The child making an utterance of items in a list (“and truck” “and car”) as separate utterances would not count as a phrase, but using “and” to mean “get ready” or something of the like would count as a phrase (for example, “and go”). It is important to make this distinction, as simply counting any string of two or more words as a phrase or proto sentence would not capture the complexity of the child’s utterance.
An utterance in which the speaker utters a single word, such as “dolly” or “ball.” By a single word we mean that the child relays only a single concept (e.g., yum yum).
In some instances, an utterance may appear to have multiple words but refer to a single concept, and would still be coded as ‘w’. Examples include “Cookie Monster”, “bye bye”, or any utterance where the child repeats the same word (e.g., “mama mama mama”). Animal sounds (e.g., mooing, roaring, barking) or other sound effects by the child should be coded as ‘w’.
An utterance in which the infant utters a consonant and a vowel together, but it does not form a word. The syllables may be repeated, such as “ba-ba-ba”, or “ga-ga-ga”, or they can be a single utterance such as “ba” or “ga”. In the childvoc column, cells marked as “b” may contain a babble. However, always listen to the audio to confirm.
If an utterance only has consonant sounds, code as a grunt (see definition below).
An utterance in which the speaker utters a vowel sound alone (e.g, /a/, /i:/) such as cooing (e.g. “ooo”). It is a babble like sound, but without consonants. The vowel sound can be repeated or uttered only once. In the childvoc column, cells marked as “b” may contain a vowel. However, always listen to the audio to confirm.
An utterance that is composed of a non-linguistic vocal sound indicating distress, fussing, yelling or whining. If there is any doubt whether an utterance is a cry or a grunt, be conservative and code as a grunt. Crying is not every negative vocalization. If the child is expressing a negative vocalization but producing a babble, word, or phrase/proto sentence, then the utterance should be marked as a babble, word, or phrase/proto sentence respectively. The goal is to capture the vocal category of crying because it does not fall into any of the other coding categories, but it is easy to recognize.
An utterance that is composed of a non-linguistic vocal sound indicating joy or finding something comical. A laugh is not every positive vocalization. If the child is expressing positive vocalization but producing a guttural sound, babble, word, or phrase/proto sentence then the utterance should be marker as a babble, a word, or a phrase/proto sentence respectively. The goal is to capture the vocal category of laughter because it does not fall into any of the other coding categories, but it is easy to recognize. If coder can’t tell the difference between a cry and a laugh while the baby is turned away, try to identify if the baby is distressed. If the baby is not distressed code as laugh.
An utterance in which the child produces a low, short, inarticulate, guttural sound often used to express effort or exertion. Vegetative sounds, such as coughing and sneezing will not transcribed and do not fall under this code. Consonant sounds alone (with no accompanying vowel sound) that are not a cry or a laugh should be marked as grunt as well. Grunt is a catchall category for any vocalization that does not fall into another code – not babble or vowel and not laugh or cry. If you are between one code and a grunt, consider coding as grunt.
Either what the child said was not intelligible to the coder, or after listening the coder was not able to discern what the infant said even with the transcript to properly code it. Transcription content in the childvoc column will use “xxx” for words within utterances where the transcriber could not decipher a word or the child’s utterance was completely inaudible. Double check these instances when coding for in the child utterance type column. Even if the original transcription indicates that a word is unintelligible (“xxx”), sometimes you as a language coder may be able to tell that the unintelligible sound was a word because of the child’s intonation etc. and can code the utterance type as a “p” or “w”.
Protophrase and word codes supersede unintelligible codes (e.g. baby says “throw ball xxx would be coded as ‘p’, “ball xxx” would be coded as ‘w’ unless it is clear the ‘xxx’ definitely refers to a word that is difficult to identify).
Set the “JUMP-BACK-BY” to 2 s.
Press control and t to space out the cells on the Datavyu spreadsheet temporally. This will align the childvoc cells with their corresponding childutterancetype cells and will make coding easier.
Hit “FIND” on the controller to go to the onset of each utterance, which was populated in childvoc column. JUMP-BACK-BY 2 s so the utterance can be viewed in context.
Play in real time to code each utterance, which is coded in mutually exclusive categories. TAB to between each argument/prompt inserting period “.” until you reach the appropriate code. Then insert periods to the end of the cell.
You can use the content in the childvoc column to help guide you as you code in the childutterancetype column. However, do not rely on the childvoc column alone. Listen to the audio and watch the video when you move into each cell to code.
When finished coding, run the script to check for typos. The script will flag any occurrence of typos that arise from an error due to any invalid codes (any characters besides ‘p’, ‘w’, ‘b’. ‘v’, ‘c’, ‘l’. ‘g’, ‘.’ or ‘x’). Fix errors and run the script again. When all the errors have been resolved the script will inform that no errors are found.
There are 4 columns associated with mother communication codes: id for communication, mom speech column, mom utterance type column, and a comments column for communication.
Column name: com_id
Code: (<lab_id>, <com_coder>, <childcom_date>, <childcom_mins>, <relchildcom_coder>, <relchildcom_date>, <momcom_date>, <momcom_mins>, <relmomcom_coder>, <relmomcom_date> <gest_date>, <gest_mins>, <relgest_coder>, <relgest_date>)
Code the com_id_column according to the definitions for coding the id column.
Column name: momutterancetype
You will receive video files that have been transcribed by the PLAY team. These transc riptions have been divided into two columns: childvoc and momspeech. Cells in these columns include the transcribed content of each speaker’s utterances and vocalizations. For the Mom communication pass, the focus is on the momspeech column. Each utterance and vocalization in the momspeech column will have it’s own corresponding, time matched cell, in the momutterancetype column.
Momspeech cells will include the utterances produced by the mother. A momspeech cell can contain full words or a code that denotes that a part of an utterance or a full utterance was unintelligible to the transcriber. The momspeech column does not include laughs, sighs gasps, vegetative sounds, and some sounds effects.
The code xxx in a cell in the momspeech column denotes instances in the transcript where part or all of a mother’s utterance were unintelligible to the transcriber. This xxx code can help you code for communication in the momutterancetype column. However, do not solely rely on the transcript contents to code for communication as the operational definitions for codes used during transcription ARE NOT the same as the definitions for communication coding.
Some utterances include the symbol //. This indicates that an utterance had a pause, and was divided into two. Utterances with // and the following utterance go together, and should be coded as the same code.
In the momutterancetype column, coders categorize the utterances in the momspeech column as a specific type of speech form. Read the utterance transcribed in the momspeech column and categorize each utterance based on codes detailed under Value List below.
Codes are mutually exclusive. The prompts/arguments in the code are designed to speed the coder through the easiest to detect and easiest to code categories (directives to look, do , and communicate, prohibitions, etc. ) down through the more nuanced and time-consuming codes. Once the proper code has been found, enter it into the prompt you are at, then code the rest as periods ”.”. For instance, if the mother didn’t direct the child to look, act or communicate, but did prohibit them from performing a certain behavior, code <.,p,.,.,.>.
Context matters! Always consider the communicative intent of an utterance.
Do not rely solely on the transcript to determine the category of mother language. Always watch and listen to the video as you read the transcript because tone, context, and other factors can help discern how to code the mother’s statement (e.g., what appear to be questions can be intended to direct the child’s actions, or the same phrase could have different meanings in different contexts). If you have narrowed down an utterance to two codes and can’t decide which to code as, choose the higher level code (e.g., if you are between coding an utterance as maintenance and provide information, code as provide information).
After completing the coding pass for the entire video, coders should rewatch the first 5 minutes in order to recheck their codes to ensure that they did not miscode any utterances. Coders must do this because there are individual differences in mother’s intonation and communication styles, which coders become more attuned to after extensive experience coding for that mom.
l = directive to look
d = directive to do
c = directive to communicate
p = prohibit/correct
i = provide information/referential
m = maintenance and engagement
r = reading
s = singing
f = filler or affirmation
x = unintelligible
z = not directed to child
Directive to look (l)
Utterances that function to direct the child’s attention, which may include naming the object or activity of attention. Utterances such as “Look”, “See what we did?”, “Look at the truck”, “Hey”, or calling the child’s name are examples of directives to look. The main function of the utterance is to encourage a child to attend to something in the environment.
Additional Guidelines and Rules:
Words like “here” or “over there” are sometimes used to direct the child’s attention, but they should only be coded if looking appears to be the ultimate goal of mothers’ utterance (e.g., instances where the mother tells the child “here” to get the child to change location are NOT attention directives; in such instances code as directive to do).
Some utterances may include the words such as look or see, but their primary function is NOT to call the child’s attention (e.g., “Let’s see if it will fit” would be coded as action directive if the mother is suggesting inserting a shape in a shape sorter, “She looks sad” would be provide information). Mothers may say utterances such as “Pay attention to me”. If the function of the utterance is for the child to look at the mother then code as directive to look. However these are different from instances where the mother might say “Pay attention to what you’re doing”, which would count as an action directive.
Directive to do (d)
Utterances with a main function of directing the child to engage in a specific action or instructing the child about how to do something. Action directives can involve the mother prompting the child to engage in a specific behavior (e.g., play with an object, retrieve an item). Examples include “Put it there”, “Flip it over”, “Turn the page”.
Action directives may specify information about the object or action or may contain generalized statements (e.g., “That will fit if you press it”, “Markers go in the box”, “We pet the dog gently”), but the main function is to guide children’s actions and thus would not be categorized as provide information. Action directives may also function to tell a child how to do something (e.g., “Walk slowly, “Play nicely”, “Be careful”, “Watch your head”). Note, the child does not need to comply with the action being prompted for an utterance to be coded as an action directive.
Any invitations to act—either together (“we” or “let’s”) or alone—count as action directives (e.g., “Wanna build bridges?”, “Let’s clean up”, “Now let’s try this one”, “Why don’t we drive the cars?”). However, if the mother uses the word “we” to describe or narrate her own action, do not count as action directive, code as provide information.
Instances where the mom says “here”, “here you go”, or other word(s) and the mom extends out an object for the child to grab likely fall under the directive to do code as the mother likely intends for the child to grab the object.
Additional Guidelines and Rules:
Syntactically, action directives may be framed as questions (e.g., “Can you feed the teddy?”, “Would you go over there?”; “Should we go read?”, “Should we find the basketball?”). Action directives framed as questions function to indirectly prompt specific behaviors in children. A helpful rule of thumb is if you remove the “can you”, “would you”, “could you”, or “should we” from the utterance, what is left over is a directive for the child to act in some way.
Directives to inform the child about a correct way to do something, such as saying “The block doesn’t go in there”, “Be gentle” would instead be coded as prohibit ‘p’. Generally, if the mom is inviting the child to do a new action, code as ‘d’, but if the mom is trying to get the child to stop and action or redirect them then ‘p’.
Directive to communicate (c)
Utterances that function to prompt the child to communicate or vocalize. Communication can include speaking or gesturing. Many such utterances by the mother are formulated as questions (e.g., “What color is that?”, “Can you count?”, “What letter does ball start with?”). Other examples include the mother explicitly eliciting words from the child (“Say bye-bye”). Mothers offering a choice (e.g., mother holding out two toys, mother asking if child wants to play on floor or couch) for the child to respond by selecting the object or location would also count as directives to communicate (the action choice is the intended communicative action). Sometimes moms say “Are you ready?/ Ready?”, because the mom is asking the child about their state to perform code as directive to communicate. Situations where mom is asking child if they “remember” how something works or how to act should be coded as directive to communicate.
Mothers may also repeat a word the child says in the form of a question (e.g., child says “water” and mom responds “water?”), code as a directive to communicate. In these instances mothers are likely looking for clarification or elicit more communication from the child.
Note, the child does not need to provide a verbal or gestural response for an utterance to be coded as elicit verbal, but often if the child DOES respond verbally or through gesture or action to the mother’s utterance then it is likely a directive to communicate.
Utterances that function to stop, curb, slow down, or correct the child’s actions (e.g., “Stop that”, “No”, “Don’t do that”, “Stop crying”, “Don’t pick that up”, “No hitting”, “Put it down”). Prohibitive utterances often include the words no or don’t (but always pay attention to the context). Other examples can include “Be quiet”, “Leave that on the ground”, “The square doesn’t fit there”, “Hold on”, “Wait”, where the goal is to curb or correct the child’s actions. Short words such as “Uhuh” or “Mnm mnm” count as prohibitions/corrections.
Additional Guidelines and Rules:
Sometimes mothers use prohibitions/corrections in indirect ways (e.g. “We don’t play like that”,
“You can’t fit under there”, “No thank you”, “That’s yucky”, “Mommy’s brush is not a toy”). Such utterances should be coded as prohibitions (not as provide information), because their main function is to stop or prevent a behavior.
Prohibitions can be phrased as questions as another indirect way of stopping or correcting child action (e.g., “Can you stop hitting that?”, “Could you quit crying?”, “Would you not eat that?”). Whenever the context and function of a question is to stop the child from engaging in a behavior, the utterance should be coded as prohibition.
Utterances where mother corrects the child’s speech or gesture while using “no” (e.g., “No that’s not a triangle” in response to the child calling a circle a triangle, “No that isn’t dad”, “That’s not blue” if the child points to green after the mother asks “Where is the blue one?”) are coded as provide information ‘i’ utterances, and do not count as a prohibition.
Provide information (i)
Utterances that function to provide information about objects and activities or contain explicit referential information. These can include labeling objects (“That’s your bottle”, “The blocks fell”), narrating actions by the child or by the mother (“You’re eating”, “I’m cooking”), or commenting on states or emotions (“He’s sad”, “I’m tired”, “You’re cranky” , “You hurt your foot”). Utterances that include comparisons (e.g., “Looks like a square”), reference a past experience related to the object or event (“It looks like the one from grandma’s house”) or describe objects or events (“The little one”), should be coded as provide information. Sometimes mothers will provide information as questions (“Is that your baba?”). Usually these rhetorical questions are about objects or events the mother is naming to provide information to the child.
A general rule of thumb for deciding between a code of ‘i’ or maintenance/engagement ‘m’: if an utterance includes a noun, adjective, verb, or location (and does not fall under a directive or prohibit code), it likely will fall under the provide information code. Examples include “You want a bigger train?”, “How about this one here?”)
Additional Guidelines and Rules:
Often mothers narrate or describe their own ongoing or future actions when speaking to children (e.g., “I’m going to get your bottle”, “Can I wipe your nose?”). These types of utterances should be coded as provide information/referential, as the mother is trying to let the child know what is going on or what will happen soon.
Sometimes mothers will discuss events that have just happened (i.e., immediate past activities), or actions the child is already or just finished doing. For example, child walks across the room with a cup of water, and the mother says “Get the water” describing what the child is already doing. These types of utterances should be coded as ‘i’ or ‘m’ depending on context and the other rules for coding.
Mothers may also discuss their own knowledge or mental state as part of their utterances (“I know where it is”). These types of utterances should be coded as provide information.
When a mother recasts or repeats a child’s speech but adds additional information (Child says “pumpkin”, mother says “Yeah, the pumpkin it’s under the bed”) count as ‘i’. If the mother just repeats the utterance without adding new information (Child says “blue ball”, mother says “Mmhm it’s blue”) count instead as maintenance/engagement ‘m’.
Utterances that contain the word “no” but are meant to offer correct information (e.g., “No that’s not a triangle”, “No that isn’t dad”, “he’s not blue”) count as provide information ‘i’ rather than prohibition.
Utterances where the mother is counting “One, two, three…” (e.g., during hide and seek, while counting blocks) should be coded as “i”. Utterances where the mother is saying letters of the alphabet (“a@ b@ c@”) but NOT singing them fall under provide information.
Maintenance and engagement (m)
An utterance that functions to continue the flow of conversation or as a nonspecific response to a child’s vocalization or action. Often maintenance and engagement utterances serve to further comment on the ongoing events after another type of utterance (after providing information or directive to communicate). For example, mother says “You’re gently moving it” (coded as ‘i’), then she says “I noticed” (would be coded as ‘m’). Maintenance and engagement utterances also include instances when the mother repeats what the child said without adding more information (e.g., repeating the child’s utterance; child says “hug” mother repeats and says “hug”) EXCEPT in situations where the mother’s repetition of the word is intended for the child to repeat the word as well (code as ‘c’).
Additional Guidelines and Rules:
Sometimes maintenance utterances are framed as questions such as “What should we do next?”, “Where are you going?”, or “What are you doing?”. Discern whether the mother's statement appeared to direct the child’s behavior or communication in these instances, if her statement did not function to direct action or elicit communication, it should be coded as maintain engagement.
Sometimes mothers will discuss events that have already happened (e.g., past tense activities). These types of utterances should be coded as ‘i’ or ‘m’ depending on context and the other rules for these codes.
Sound effects alone (e.g., “choo choo”, “mwah”, “woof woof”, “boo”), without any additional communicative information should be coded as maintenance and engagement. This includes sound effects made in the context of pretend play (e.g., mom says “mmm” as if to imitate someone enjoying eating). If an utterance with a sound effect has additional information (e.g., “The train goes choo choo”) then code as provide information.
Instances where the mother says the child’s name (or a pet name such as “sweetie, honey, sweetheart, etc.”) and nothing else, and the utterance does not fall under a directive or prohibition, code as maintenance.
An utterance in which mother reads from a book or other source, thus the sequence of words is not self-generated/created. If based off of mothers tone/intonation and content you can tell the mother is not reading verbaitum, do not code as reading. If the mother uses some of the text verbatim but adds additional words, code as provide information (“Look the fish is saying “Bloop bloop”).
If you are unsure if the mother is reading, and the pages of the book are visible on the video, try to see if you can identify the text on the book and if it matches what the mother is saying.
An utterance in which the mother is singing from memory or along to a song. If the intonation of the voice is singing for any utterance, and the utterance does not fall under any prior codes (e.g., directives, prohibitions, etc.), code as singing. If the mother uses some of the words from the song verbatim but adds additional words, code as provide information (“GIVE EXAMPLE”)
Singing the alphabet counts towards the singing code, but if mother is saying letters without singing, count as provide information.
Filler or affirmation (f)
An utterance in which mother acknowledges the child’s behavior, uses words as conversational fillers, or provides a short affirmation. For instance, when the mother says “There you go” when the child successfully completes a puzzle, or when she says “Yeah”, or “Uh huh”. A list of common conversation fillers is provided on page 13.
Incomplete thoughts by the mother that are never finished, and cannot be coded due to lack of further detail can be coded as filler/affirmation.
Filler and affirmation words or phrases are usually said alone with no other words. Note that words that otherwise could be coded as fillers and affirmations can be part of longer utterances (e.g., “Oh thats your dolly”, “Yeah the duck is yellow”, “Good job petting gently”). Just because an utterance includes a filler or affirmation word/phrase it does not automatically mean the utterance should be coded as a filler/affirmation.
Any utterance that contains an affirmation plus the child’s name or a pet name (“sweetheart, sweetie, honey, love”) would also just count as a filler. However, that filler must not contain any other information or a directive (in those cases code as ‘i’ or a directive, respectively).
Instances where the mother repeats a filler/affirmation that the child said should be coded as maintenance NOT as filler.
An utterance was not intelligible to the communication coder.
Transcription content in the momutterancetype column will use “xxx” for words or full utterances where the transcriber could not decipher the content of the utterance (e.g., “Why don’t you xxx?”). Double check these instances when coding in the momutterancetype column. Even if the original transcription indicates that a word or full utterance is unintelligible (“xxx”), sometimes coder can detect if the utterance was an action directive, prohibition, etc. based on intonation, other words, or other aspects of the utterance and video context.
The unintelligible code is NOT used as an “I’m not sure” code. If a coder is unsure of a code, they must use their best judgement and process of elimination to narrow down the code.
Not directed to child (z)
This code is used when the mother is speaking to a voice assistant (e.g., Alexa, Google Home, etc), on the phone, or to a pet. These would count as words that would be overheard by the child, but are not intended as speech directed to the child.
If the mother is narrating her actions or talking about ongoing things in a performative way with the child this does not count as ‘z’. Usually this is the mom talking to the child by talking to herself about what she is doing, or maybe talking for the camera and so codes will usually follow the rules for other codes ( ‘i’ or ‘m’).
HOWEVER, situations where mom is talking to herself because of something that happened (e.g., mom dropped something and swears, stubs her toe, etc.) that isn't child directed speech/there is a change in tone, can be code as ‘z’.
Set the JUMP-BACK-BY key for 2 s.
Press control and t to space out the cells on the Datavyu spreadsheet temporally. This will align the momspeech cells with their corresponding momutterancetype cells and will make coding easier.
Hit “FIND” on the controller to go to the onset of each utterance, which was populated in the momspeech column.
JUMP-BACK-BY 2s so the utterance can be viewed in context.
Play in real time to code each utterance, which is coded in mutually exclusive categories. Tab between each argument/prompt until you reach the appropriate code. Use the up and down arrows to navigate between cells.
Run the “add_periods_momutterancetype.rb” periodically to add periods to empty arguments in cells the coder has already coded.
You can use the content in the momspeech column to help guide you as you code in the momutterancetype column. However, do not rely on the momspeech column alone. Listen to the audio and watch the video when you move into each cell to code.
Filler words and affirmations are coded as ‘f’ for the momutterancetype communication pass. These are words that provide limited information beyond simple agreement or are used commonly to keep the flow of conversation going. When used as a filler or affirmation they are usually said alone with no other words. Note that sometimes these words or phrases are part of longer utterances that should be coded as another code. (“Oh thats your dolly”, “What color is the duck?”, “Play nice”. Just because an utterance includes one of these words it does not automatically mean the utterance should be coded as a filler/affirmation.
Good job / good work
Here we go
Hello/ Hello (name)
Mmm (EXCEPT when “mmm” is said as part of pretend play/saying something tastes good)
Nice/ nicely done/ nice job
Oh my/ Oh my goodness
Oh my god
Okay/ it’s okay
There you go
Uhoh (EXCEPT when tone indicates prohibition)
Way to go
You’re fine/ everything is fine