#10. The Special Value of Age-Mixed Play I: How Age Mixing Promotes Learning

When children of different ages play together, the older ones elevate the level of play of the younger ones and learn to nurture, teach, and lead.

Dear friends,

Normal play for children is age mixed. Play just among children very close in age is an artifact of modern times. It began with age-segregated schooling, and then became more common as we have, over time, increasingly segregated children by age even in out-of-school settings.

Anthropologists report that children in hunter-gatherer bands essentially always play in broadly age-mixed groups (Gray, 2012). There aren’t enough children in any given band for them to form playgroups in which everyone is close in age, even if they wanted to, and there is no evidence that they want to. There is reason to think that throughout the 99% of our biological history when we were all hunter-gatherers—the period in which the human drive to play was shaped by natural selection—essentially all play was age-mixed. The play drive evolved to serve its developmental magic best when the players differ from one another in age.

Even in 20^th century America, until near the end of the century, much if not most play was age mixed. Neighborhood play, which occupied so much of the life of children as recently as the 1950s, ‘60s, and ‘70s, was usually age mixed. The kids in any given neighborhood varied widely in age, and kids played with those who were there. Best friends were usually close in age, but play included others over a wide range of ages.

It's not just necessity that brings kids of different ages together to play; it’s also choice. Older kids are attracted to younger ones, and younger ones adore the older ones. Some years ago, my then graduate student Jay Feldman and I (OK, mostly Jay) conducted several studies of age mixing at a democratic school where children of all ages, from 4 on through late teenage years, are free all day to play and interact with one another. The school was large enough (135 students) that if they had wanted to play just with kids close to their own age, they could have. But, in a quantitative study, we found that more than half of the naturally occurring social interactions among students spanned age gaps greater than two years and a quarter of them spanned gaps greater than four years. Age mixing was especially common in play; it was less common in serious conversations (Gray & Feldman, 1997).

This is the first of two (or maybe even three) letters I plan on the special value of age-mixed play. This letter focuses on the idea that children learn more when they play in age-mixed groups than when they play with others who are similar in age.

The Zone of Proximal Development

In the 1930s, the Russian psychologist Lev Vygotsky coined the term zone of proximal development to refer to the set of activities that a child cannot do alone or just with others of the same ability but can do in collaboration with others who are more advanced. He suggested that such collaboration is the most natural and common way that children develop new skills and understanding. Extending Vygotsky’s idea, the Harvard psychologist Jerome Bruner and his colleagues introduced the term scaffolding as a metaphor for the ways by which skilled participants enable novices to engage in a shared activity. The scaffolds consist of the reminders, hints, boosts, instructions, affordances, and other forms of help that lift the child up to a higher level of activity.

Educators typically apply Vygotsky’s and Bruner’s concepts to interactions between children and adult teachers, but I think they apply even better to age-mixed interactions among children. When children play in age-mixed pairs or groups, the older, more skilled participants naturally, often unconsciously, provide scaffolds that raise the level of the younger participants’ play. Here are some examples that have been documented in the research literature on age-mixed play.

How Older Children Make Social Play Possible for Toddlers

Some of you may remember reading, in a developmental psychology course, about Mildred Parton’s classic theory of stages in the development of play. According to Parton, children of 2 or 3 years in age are incapable of collaborative social play. When placed together, they engage in what Parton termed parallel play; they play side by side, paying some attention to one another, but not merging their play into a socially combined activity. As anthropologist Melvin Konner (1975) pointed out decades ago, nowhere in the world would you see such parallel play except in the artificial modern Western environment of an age-segregated daycare center or developmental psychology lab. The normal playgroup, throughout the world and especially in traditional societies, would always include some older children who, in their play, would scaffold the little ones into truly social play. Several research studies have revealed how effective such scaffolding can be.

In one study (Howes & Farver, 1987), with an experimental design, 2-year-olds and 5-year-olds were observed playing in pairs. The children in each pair were members of the same age-mixed daycare center, so they knew one another. The play situation was one in which the children were shown an interesting new toy and were invited to play together with it. The researchers made comparisons across three types of pairings: 2-year-olds with other 2-year-olds, 5-year-olds with other 5-year-olds, and 2-year-olds with 5-year-olds. Not surprisingly, the pairs of 5-year-old played in much more complex ways, and much more socially, than did the pairs of 2-year-olds. More interesting was the finding that the 5-year-olds played at the same advanced level when paired with 2-year-olds that they did when paired with other 5-year-olds, and they used a variety of verbal and nonverbal scaffolding techniques to draw the 2-year-olds up toward their level of play. Because the 5-year-olds could structure the roles for the 2-year-olds in shared fantasy play and could help them play out those roles—such as by providing them with the appropriate props and instructing them in what to do—the 2-year-olds could engage in truly social, cooperative pretend play with 5-year-olds, which was beyond their ability when paired with 2-year-olds.

Another study (Maynard, 2010), with a quite different method, involved filming and coding the behavior of pairs of siblings playing naturally together in 36 different Mayan households in a Mexican village. The study focused specifically on pairs in which the younger child was 2 years old and the older one was anywhere from 3 to 11. The children played at such everyday activities as making pretend tortillas, caring for baby dolls, selling products at a make-believe store, and soccer. According to the researcher, every play episode was also a teaching and learning episode, as the older child always (consciously or not) helped the younger one play in more advanced ways than would have been possible otherwise. Even the 3-year-olds helped by providing models of more advanced actions, which the 2-year-olds observed and imitated. Not surprisingly, the older the play partner, the more skilled that partner was in increasing the complexity and sociability of the 2-year-old’s play.

By the age of 8, these children were quite sophisticated guides for their younger siblings, giving them verbal explanations of how to play specific roles, providing them with appropriate props, physically helping them with difficult maneuvers, and modifying their own activities in ways that allowed the 2-year-olds to respond appropriately.

An example is a scene in which 8-year-old Tonik and 2-year-old Katal played together at giving a baby doll a bath. Katal wanted to do the washing herself and Tonik enabled her to do so by demonstrating the process, providing her with a glass of water to pour over the doll, and giving step-by-step verbal instructions on the appropriate way to wash a baby. It is important to note that such examples are real play for both the older and younger participants. They are both having fun, both engaged by choice for no extrinsic ends, and both are exercising new skills. These are not pedantic interactions, as might have occurred if an adult were “playing” with a 2-year-old at washing a baby doll.

Although my focus in this letter is on the learning that occurs by the younger play partner, note that the older one is practicing skills that are at least as valuable, such as how to teach, lead, nurture, and in other ways help a younger person or novice. The older one is also thinking consciously about, and putting into words, ideas that might not have been so conscious before.

How Young Children Gain Literacy and Numeracy Skills in Age-Mixed Play

Several research studies have shown the value of age-mixed classes, especially if the children have ample time to play together. In one such study (Christie & Stone 2002), for example, the play activities of kindergarten children in a mixed-age classroom that also included first and second graders were compared with the play activities of kindergarteners in the same classroom the next year, when it was kindergarten only. The play area both years included items that might foster reading and writing, including children’s books, cookbooks, newspapers, store coupons, empty food containers and paper and pencils.

The researchers found that the complexity of play and the amount of activity involving reading and writing were much greater when the classroom was age-mixed than when it was kindergarten only. This is not surprising, as the sophistication of play was no doubt raised by the older children in the age-mixed setting. More interesting were the comparisons just of the kindergarteners’ behavior in the two conditions. In the age-mixed condition, the kindergarteners played most often in groups that included at least one and usually more than one first or second grader, and, consequently, they were drawn into more complex play and more play involving reading and writing than occurred in the kindergarten-only condition. Per pupil, the kindergarteners engaged in nearly four times more reading and six times more writing in the age-mixed condition than in the same-age condition. Most of this literacy behavior occurred in the context of shared fantasy play. For examples, children would read recipes while playing at cooking, read bedtime stories in playfully putting a baby to bed, and write labels on presents in preparation for a pretend birthday party.

In a conceptually similar study (Emfinger, 2009), of numeracy rather than literacy, conducted at an age-mixed summer enrichment program, the researcher focused on episodes of free play among children ranging in age from 4 to 10. She found many instances in which the older children exposed younger ones to numerical concepts that would have been beyond the younger children’s abilities to understand or use alone.  For example, in one scene an older child explained how to give exactly 7 drops of medicine, no more and no less, to a sick doll. In another, an older child explained to a younger child, in a game of store, how much it would cost to purchase two items in the case where one cost $10 and the other $5, and how much change to give for a $20 bill. It’s obvious that such concepts are far more meaningful to children in the context of their own, self-directed pretend play than in the involuntary and more abstract realm of typical classroom instruction.

Such findings are consistent with informal observations I and others have made at the previously mentioned democratic school. At any given time of day at the school, it is possible to find older and younger children collaborating at activities that involve one or more of the three Rs. In card games, board games, and computer games that involve keeping score, older children teach younger ones how to compute scores, a process that usually involves addition and sometimes subtraction or more complex calculations. In games involving written words, older children with literacy skills read the words aloud to the younger children or tell younger children how to spell words that they need or wish to type or write. In the process, the latter soon learn to recognize the most frequently used words. In such play the older ones are not thinking about “educating” the younger ones, they are just doing what is necessary to play the game. Many children at the school learn to read and write and become skilled with numbers with no formal instruction at all, largely through their age-mixed play with older children.

Examples of Scaffolding on the Outdoor Playground

In an extensive observational study, Jay Feldman and I (Gray & Feldman, 2004) focused specifically on age-mixed interactions among adolescents (age 12 or older) and children who were at least 4 years younger and not yet adolescents. Essentially every one of the 200 such interactions we recorded involved natural teaching by the older participant. Here are three examples in the realm of outdoor physical activities on the school’s campus:

•     In a game of four square, the older players allowed Ernie (age 4) to catch and throw the ball rather than hit it. Shawn (age 17) was particularly good at hitting the ball softly into Ernie’s square so he could catch it.

•     In an exuberant bout of “boffing” (fencing with soft padded swords), Sam (age 17) fended off seven 6- to 10-year-old attackers, who chased him around trying to hit him with their swords.  Sam adjusted his fencing movements to accord with the skills and style of each of the younger boys, thereby presenting each with an exhilarating challenge without overpowering any of them.

•     Ed (a tall, athletic 15-year-old) was playing basketball with a group of 8- to 10-year-olds. He rarely shot but spent much time dribbling while the horde of small boys who made up the opposing team tried to steal the ball from him. Then he would pass to his single teammate, Daryl (age 8) and encourage him to shoot.

In each of these cases, the teenagers adjusted their play to allow the younger players to engage in and enjoy the game, but the adjustments were clearly not sacrifices. In each example, the teenagers appeared to enjoy the game and learn at least as much from it as did the younger players. Shawn enjoyed exercising his ability to keep little Ernie in the game. Sam’s fencing and gymnastic skills were exercised to the maximum as he fended off the horde of young attackers. And Ed, by not shooting and by repeatedly dribbling through the crowd of short defenders and setting up his young teammate to shoot, was maximally exercising his dribbling, passing, and play-directing abilities. Shooting and scoring himself would have been too easy and would have spoiled the game for everyone.

Here’s a quite different example:

•     A group of children ranging from age 4 to 8 were playing “bumper cars” on the slide (one of those old-fashioned very tall, steep, “dangerous” slides, the kind that is fun).  One child would slide slowly down and then remain sitting at the bottom, and then the next one would slide down rapidly and attempt to bump the first one off the end of the slide.  After about 20 minutes of this and variations of it, Rebecca (age 14) and two of her somewhat younger friends came along and asked if they could join the slide play.  After joining, Rebecca made suggestions that modified the game. First, she introduced a game in which the object was to get everyone on the slide at once and all release at once, creating a huge pileup at the bottom. Then she modified that into a game in which all the players had to line up by height and slide down together in that order, so the biggest person would be at the bottom and the shortest person at the top of the pile.

Here, the sliding game became more complex after Rebecca joined the group than it had been before. If Piaget’s theory of cognitive development is correct, the youngest children in this scene would not and could not have thought of ordering themselves by height (because they lack the concept of seriation), but under Rebecca’s guidance they quickly grasped the idea and eagerly sorted themselves by height to engage in this new adventure. Rebecca’s way of playing also made the game safer, by ensuring that the littlest kids would be at the top rather than the bottom of the pile.

Final Thoughts

And now, what are your memories of age-mixed play? Are they congruent with what I have described here? If you have children, do they have as much opportunity for such play as you did? I invite you to share your stories in the comments section below.

If you like this this series of letters, please subscribe if you haven’t already, and please let others know about it.

With respect and best wishes,

Peter

References

Emfinger, K. (2009). Numerical conceptions reflected during multiage child-initiated pretend play. Journal of Instructional Psychology, 36, 326-334.

Gray, P. (2012). The value of a play-filled childhood in development of the hunter-gatherer individual. In Narvaez, D., Panksepp, J., Schore, A., & Gleason, T. (Eds.), Evolution, early experience and human development: from research to practice and policy, pp 252-370. New York: Oxford University Press.

Gray, P., & Feldman, J. (1997). Patterns of age mixing and gender mixing among children and adolescents at an ungraded democratic school. Merrill-Palmer Quarterly, 43, 67-86.

Gray, P., & Feldman, J. (2004). Playing in the zone of proximal development: qualities of self-directed age mixing between adolescents and young children at a democratic school. American Journal of Education, 110, 108-145.

Howes, C., & Farver, J. (1987), Social pretend play in 2-year-olds: effects of age of partner. Early Childhood Research Quarterly 2, 305-314.

Konner, M. (1975). Relations among infants and juveniles in comparative perspective. In The origins of behavior, vol. 4: friendship and peer relations, ed. Michael Lewis and Leonard A. Rosenblum (1975), 99-129.Maynard, 2002.