Leading and Caring for Children

Function Coding

Each statement in a conversation has one of five purposes; some of those purposes are pushy.

I chiseled down other function coding systems found in the literature to five classifications that relate to how one behaves in a conversation. To code the functions, we look at each statement the adult makes, name it, and place it into one of five pigeon holes. After all the birds are sorted, we can look at the percentage of each function, yielding a set of numbers. Then we can see what the numbers tell us about the way we talk in our conversations with children.
Function Codes PDF


We can try this coding system on the Ricky and Paul conversations. We only apply Function Codes to what the adult is saying. Sometimes the coding can be tricky.


Ricky is just three years and two months old. He is with his mother at the Laundromat. Mother is loading the dryer with wet clothes.

  1. Ricky: Let me put some in.
  2. Ricky: I want to put them in.
  3. Mother: Oh, give up. Demanding, for sure, so we look at the lower half of the code only. We can choose D T or Q. Of those three choices, it is a directive. D
  4. Ricky: I want to.
  5. Mother: Go sit down. Another directive. D
  6. Ricky: Let me do it!
  7. Mother I told you no. Non-demanding. No obligation for Ricky to do anything, so it’s S or I. Since it’s not new information, it’s social. S Social is a big category of leftover stuff; it’s talking but not adding anything.
  8. Ricky: (runs down the length of the Laundromat)
  9. Mother: (holds up coins) Look here. D
  10. Mother: Come back and put them in. D
  11. Ricky: (comes back and gets the coins) It won’t go in.
  12. Mother: Here. Watch. D
  13. Ricky: Oh, like this. (the machine starts)
  14. Ricky: Why does it go round and round like that?
  15. Mother: Because it does. S
  16. Ricky: But why?
  17. Mother: Oh, I don’t know. S
  18. Mother: Come sit down. D
  19. Ricky: I’m going outside.
  20. Mother: No, you stay here and keep still. D
  21. Ricky: (sits down but starts sliding away down the bench)
  22. Mother: Come back and sit here. D (Directives are getting kinda old, huh?)
  23. Ricky: (runs off)
  24. Mother: (grabs Ricky and carries him back) only code verbal messages
  25. Mother: (hands a children’s magazine to Ricky and picks up one for herself)
  26. Ricky: (turns the pages and points to a picture) What’s that?
  27. Mother: My feet hurt. informative I. (She said something new finally.)
  28. Ricky: What’s that?
  29. Mother: Stop kicking your feet. D

Now we take the counts of each code and put them into a table like this one. We add up the number of coded statements (3+1+9=13). We compute the percentage by dividing each function count by that total, e.g., 3 ÷ 13 = 0.23, or 23% social.
Here are Ricky’s Mom’s numbers.

Ricky’s mom


Task Six

Function Code Paul’s mom. You can usePaul PDFor write just the number of each Mother statement (skipping the Paul ones) on a piece of paper and put a letter code beside it.


The same Laundromat. Paul, who has just turned three, stands with his mother as she loads the dryer with wet clothes.

  1. Mother: Hold the door for me, will you?
  2. Paul: (holds the dryer door open) Oooh—it’s warm.
  3. Mother: Too hot to hold?
  4. Paul: Not too hot on this part.
  5. Mother: (dumps the clothes into the dryer) Empty all the wet clothes inside.
  6. Paul: That’s a lot, isn’t it?
  7. Paul: We have to put money in, right?
  8. Mother: Four quarters, I think.
  9. Mother: Can you reach that high to put them in?
  10. Paul: I’ve put one in.
  11. Paul: It’s going around fast now.
  12. Paul: If I put in more will it go faster?
  13. Mother: Not faster, it will spin longer.
  14. Paul: How longer?
  15. Mother: We will have to wait longer for it to stop.
  16. Paul: Why does it go around?
  17. Mother: Well, the clothes are tumbling and the water runs out the holes.
  18. Paul: I can’t see any holes.
  19. Mother: Look up at the top.
  20. Mother: When the clothes fall down you can just see them.
  21. Paul: I can just see them!
  22. Paul: And the water goes out through them?
  23. Mother: Well, some of it does, I think.
  24. Mother: But it gets hot, too.
  25. Mother: Feel the door.
  26. Paul: It’s hot.
  27. Paul: Does that make them dry?
  28. Mother: You know what happens when we put wet things in front of the fire.
  29. Paul: My coat went all steaming, didn’t it?
  30. Mother: You did get all wet that time.
  31. Mother: And that’s how the clothes get dry in the dryer, too.
  32. Mother: Now we have to find somewhere to sit and wait.
  33. Paul: (runs off down the Laundromat)
  34. Mother: I think you might be in the way down there.
  35. Paul: (chooses a magazine and brings it to his mother)

Answers for Paul’s Mom

1 D — 3 Q — 5 I — 8 I — 9 Q — 13 I — 15 I — 17 I — 19 D — 20 I — 23 I — 24 I — 25 D — 28 I — 30 S — 31 I — 32 I — 34 I
#5 She is not asking Paul to do it, she is doing it and saying what she is doing in English. Informational.
#9 I am assuming she doesn’t really know if he can reach it. Query
#30 When a statement acknowledges by restating the same thing, it is redundant. There is no reason to say it other than for social purposes. Social
#32 This is an example of cuing a child to do something by stating a fact not directing; the child still can initiate the act of finding somewhere to sit (or not). The demand is gone.

Ricky’s mom
Paul’s mom
Joan’s mom

Function Percentages

So, what do the numbers mean? Some of you may know the work of Joan Tough, lecturer at the University of Leeds, who started me on all of this, and Catherine Snow, Harvard Graduate School of Education, who has used something similar in her research. I am not an expert in the research, but one piece of that research base, Catherine Snow’s examination of interactions in child care centers in Bermuda, particularly intrigued me and fit with what I was finding on videotape. Her study controlled for the usual causes of variance, which are cultural norms and economic circumstance, since all of the children in her study were of African descent and the children that scored the best were the lowest income children enrolled in the government subsidized centers.

It was confirming to find in the research a function coding system similar to mine used in a controlled study of adult interactions with young children. Her data support my observations of the effect of teacher talk on children’s language development: the highly informational preschool environments resulted in greater school success.

You see the same in Joan Tough’s work and in The Early Catastrophe by Betty Hart & Todd Risley. It’s pretty obvious in the example of Paul how a home experience of thoughtfulness, inquiry, and responsive interactions would be an ideal preparation for school.

This Way or That Way

I and D are opposites. One can’t have high percentages of both I informational and D directive; talking leans to the non-demanding side or to the demanding side like a teeter-totter. Recognizing that the percentages of the five types varies with circumstance. However, the most effective teachers I know habitually speak to their children with I informational percentages above 60%.

Those teachers who habitually speak with D directive percentages above 40%, like Ricky’s mom, have trouble managing their classrooms. When demanding statements dominate adult discourse, the children are being controlled.

As you can see from the example of Ricky, his mom is not providing him with information about the Laundromat or the processes occurring while they’re together. She has no interactive discussions with him. Ricky’s Laundromat lesson is not about dryers: it’s about testing the limits. For him, talking is about control. The bigger, more powerful person controls other people.

Since talking is rarely about representing and thinking about events of the world, in itself it isn’t interesting to learn or to use, except to push back or boss others. Why would reading suddenly be interesting for Ricky in elementary school? As soon as anyone pushes him to read, he will probably push back, just as he did with Mom in the Laundromat.

I think it is obvious to most people who read the Ricky conversation, and proven in Catherine Snow’s research, that children who grow up in a Ricky culture face greater challenges in school.

Language Environment in the Early Years

I believe that negative carryover from home can be overcome by age seven, if you start early, but not in an authoritarian school. When children are directed, they have two choices: they can either comply (acquiescence) or not comply (rebellion). Neither option do I value, although it appears the public school system of the United States has made it explicitly clear that it values both acquiescence and rebellion.

It’s worth a moment’s pause to consider how the current academic readiness” discourse masks the real problem of authoritarian control. None of those people in power with private money or federal money, who favor a coercive public educational policy, would ever want to fund Pre-K or Kindergarten programs that emphasized educating children in an informative, non-directive, way, the way those powerful people want reserved for their own children.

Classroom Control

I can hear the traditional argument, “When you lift control, the children will go crazy!” Yes, that is very true, just like the YouTube videos of confined cows being let out to pasture. They may have to settle themselves.

What do percentages signify?

Back to conversation analysis. I’m reminding myself that we are here to learn to hold mutually enriching conversations with children. My assistance with your endeavor is to offer you, eventually, the Conversation Analysis Self-study Form PDF. A tool in that analysis is Function Coding, which is simply categorizing statements and computing percentages.

I recognize that the meaning of the percentages can only be usefully determined by the participating adult who records a conversation. However, based on my experience, I can offer some general guides about the Function Coding numbers I have seen and how the students thought of their own work.

Contexts matter. The ratios found in conversations differ from the ratios found in general classroom talk and differ from the ratios found in group times. Here we are just looking at one-to-one conversations.

The conversations people are most happy with and seem to be the most mutually enriching have these characteristics:
The percentage of S social is between 20-40%, like Joan’s mom. The relatively high Social occurs when the adult is paraphrasing.
The percentage of Q queries is between 10-20%, happens through leading queries on the child’s topic.
When all of the above are happening, the leftover amount is Information. The I informational percentage hovers near 50%.

My general conclusion, after examining close to 400 coded conversations, is that in the most successful conversations:

I informational > 40%
D directive < 10%.

If the Informational fails to reach at least 40% in a conversation, I suggest there is a problem.

In general classroom time in lower elementary and preschools, I have recorded teachers who consistently talk in general classroom time with Informational above 90%, and every one of them is greatly loved by the children, the families, and the school. Of course, all of them had completed two Conversation Analyses with these tools. They knew what they were doing.

Task Seven

Understanding Check

Now that we have covered half of the analysis tools, I invite you to read this conversation between Lucy and her mom. This is a little self-test for you whether my markups for topic changes and function codes make sense.

The Mouse Trap Game

Lucy, 5 ½ years, and her mom talk about fairness. Mom gave Julian, 2 ¼ year-old younger sibling, the Mouse Trap game. He was happily playing with the materials in his own way. When Mom heard Lucy telling Julian he couldn’t do something, she stepped in.

  1. Mom: Something we have talked about is that people get upset when someone takes what they are using. I informational, topic is being upset
  2. Lucy: Then they get excited and they take it.
  3. Mom: Yeah. When Julian takes something from you, you yell and grab back. I informational
  4. Lucy: Yeah, because remember that time it was a long fight when I got the chick back.
  5. Mom: I remember that fight. S Social
  6. Lucy: …when we were going to school for the field trip.
  7. Mom: That was the first time you guys started arguing about a toy. I mean really the first time. It was a big deal. I informational
  8. Lucy: Well, I mean the longest fight, the longest we have ever had.
  9. Mom: Yeah, because you had brought the chick and then set it down to buckle yourself in.  I informational
  10. Mom: When you were done, Julian had the chick. I informational
  11. Lucy: Yeah, he had the little toy that said ‘Lucy’ on it.
  12. Mom. That was a big fight. Wait… and that was the day we talked about strong reactions like crying and yelling. I informational
  13. Lucy: Julian had two toys, so I was like, Julian that’s not fair. You have two toys! (she bangs her fist on the table)
  14. Mom: So, he had a toy for himself and then he took your toy. He had both toys.  S Social

  15. Mom: I want to talk about the Mouse Trap game today, because that was something Julian was using and you wanted to use it, too. I informational, topic is turns

  16. Lucy: And he took two turns that time and it wasn’t fair because he got two turns.
  17. Mom: And you wanted to make sure you were getting turns.  S social
  18. Mom: So, Mouse Trap must be different than your stuffed animal chick.  S social
  19. Lucy: Yea, because it’s for everyone.
  20. Mom: It’s a common toy?  S social
  21. Lucy: It’s not Julian’s. Everyone can play with it.
  22. Mom: But Julian did ask for it. I informational
  23. Mom: Do you think he could ask for it and want to play with it by himself? Q query
  24. Lucy: I don’t know.
  25. Mom: What if you wanted to play with it by yourself? Q query
  26. Lucy: I wouldn’t. I would want to play it with Julian every time I wanted to play with it.
  27. Mom: OK. We figured out Mouse Trap. I informational
  28. Mom: It’s a common toy and people can play with it together. S social
  29. Mom: Even if someone wants to play with it alone, they have to let the other person play if they want to, right? S social
  30. Lucy: Right
  31. Mom: Does that apply to all of those other games? Q query
  32. Lucy: All those other games.


Mom commented afterward that Lucy’s solution was better than her solution of letting him play with it undisturbed. After this, the invitations to join in all kinds of play became more common occurrences. More evidence that this was indeed a Mutually Enriching Conversation.

Note, too, how the numbers fit in the ranges I see in most Mutually Enriching Conversations:
S social — 20% to 40%
I informational — around 50%
Q queries — 10% to 20%


It’s like breaking a code. Now I understand what is going on in my conversations with children, and I really enjoy taking time to do that. I tell my friends how incredible it is to have conversations with children now that I am aware of my role in the discourse. The most important new understanding I gained was how to follow the child’s lead and allow them to mold it and bend it as they will. — Meredith Fourre

Next Responsiveness