QJER logo 2
[ Contents Vol 4, 1988 ] [ QJER Home ]

Aggression in young children

R. M. Norton


What is Aggression?

Researchers consider that the two aspects which must be present in an aggressive act are intention and the wish to inflict hurt. Dollard (1939, in Eron, et al, 1971, p.4) regards aggression as an act whose goal response is injury to another object. Eron, Walder and Lefkowitz (1971, p.4) consider it is an "act which injures or irritates another person. " Mussen, Conger and Kagan (1969, p.329) say aggression is "intended to cause injury or anxiety to others". They include hostile motivation which is shown through a wide variety of behaviours.

In Dawe's (1934, in Hartup 1977, p.244) study of two hundred quarrels, aggressive events were defined as intentional physical and verbal responses that are directed toward an object or another person and that have the capacity to damage or injure. It does not require that the aggressive act has hostile intent or actually injures. How it is perceived by the person is not taken into consideration.

Some writers consider that children under the age of five are not aggressive because they have not developed fully their concepts of causality. They believe that their actions are automatic and their egocentricity prevents them from comprehending the effect it will have on another person. It is a normal part of development for under threes to snatch, push, bite and hit. This could be genuine anger and frustration, boredom, desire for attention, imitation, disruptive behaviour rewarded, tiredness, over-excitement, jealousy, high activity level, or social clumsiness. The label aggressive is misused and overused. Adults can see solutions the child does not see. Parents, however, regard aggression in toddlers seriously. One-third of parents who take their children to mental health clinics do so because of aggression (Patterson, 1975, Roach, 1958, in Fagot and Hagan, p.344).

In the developmental approach, the acquiring of language and growing means of communicating between the ages of three and five assist the child in finding alternatives to aggression (Hartup, 1977, p.243). Olweus (1979, in Cummings, 1985, p.495), however, found that the tendency to use aggression is fixed by the age of three. Using a longitudinal study, he found that more aggressive children at this age were likely to commit more aggressive offences as adults. Sears (1953, in Hilgard, p.148) claims that true hostile aggression, i.e. making others experience pain, develops slowly.

Burton White (1975, in Patterson, 1980) uses the term "assertive acts". However, assertiveness is usually considered to be insistence on one's own rights - a defensive act. Aggression requires an attack onto another person - an offensive act.

This study has focussed on the measuring of aggression by observation of prescribed behaviours that have been listed by other researchers as aggressive.


(a) Aggressive Behaviours

Studies that have observed and measured aggressive acts in very young children list biting, hitting, pushing, yelling, pulling hair, throwing objects at someone, breaking an object. Steur, Applefield and Smith (1971, in Liebert, et al, p.430) used four categories: hitting or pushing; kicking; assaultive contact; throwing objects.

Observers are required to avoid inferences about particular intentions, motivations and feelings of the subject. Thus a child who hits another repeatedly and gains a playful response of wrestling and laughter, has instigated it by aggression. Was the intent to really hurt?

Poking out tongues is a non-verbal response which can have more intent to hurt and more hostility than a hit on the head with a saucepan (Appendix 2, Record 6). Lyons and Serban (1986) and Condry and Ross (1985) found that observers differed in their perception of what was aggressive depending on whether they thought the subjects were girls or boys.

(b) Outcome Behaviours

Many psychologists and educators consider that aggression is learnt because of reinforcement of the action (Buss, 1961, in Eron, et al, p.80). Aggression has instrumental value, i.e. the individual gets what he wants, eliminates annoyance or fends off attack. In a study by Fagot and Hagan (1985) it was found that large amounts of aggression were ignored but more so for girls. Ignoring aggression usually ended the episode. Bandura (1973, in Fagot and Hagan, p.349) found that if the children were rewarded for copying the aggression, the girls' aggression was as high as the boys. It is suggested that this occurs because in our culture girls learn to inhibit aggressiveness at a very young age. It does not offer an intrinsic motivation for them.


Children under six were observed for thirty minutes in a day-care and a preschool with some mixed age and some same age groups. All aggressive behaviours, the antecedents, and the outcome behaviours were recorded. Three types of coding scale were attempted (Appendix 1) but it was found that without substantial practice, running records were more effective and were able to provide information about all three aspects of the event - cause and type of aggression and response of recipient. The observer was as close as possible to the children to enable reporting of vocabulary and facial expression. Because these children are used to a range of adults being present, they were not noticeably inhibited and made little comment. Interaction with the observer was discouraged by avoidance of eye contact and facial expression. There was a deliberate attempt in using these settings, to avoid interfering with the ongoing stream of behaviour (Medinnus, 1966).

The day-care children could attend from 7.30 a.m. to 5.00 p.m., five days a week. Groups observed were: 2-year olds; 3-year-olds; 3- to 5-year-olds; and 2- to 5-year-olds. The preschool children attend on a sessional basis. 3 year-olds attend for three hours, five days per fortnight; 4-year-olds for four hours; and S-year-olds for six hours, five days a week. Most groups were observed during free-play both inside and outside.

After observation, the following categories were chosen and used in the recordings:

  1. Facial - to do with verbal and non-verbal communication, i.e. yell, blah!, poke tongue, pull faces, spit.
  2. Physical discrete - clearly one person's act, i.e. punch, hit, kick, tug hair, bite.
  3. Physical combined - wrestle, fight together, push.
  4. Threaten - pretend violence, i.e. raise hand or object as if to hit; a deliberate miss.
  5. Related to an object - tugging, destroying.
Responses were categorised into seven groupings:
  1. Ignore
  2. Positive social response, e.g. smile, talk to, cuddle, join in
  3. Retaliate, i.e. fight, hit
  4. Verbal aggression
  5. Move away
  6. Cry or look sad
  7. Seek help, i.e. move near to an adult or ask for help
It was attempted to judge the purpose of the action. This requires an evaluation by the observer which is frowned on in anecdotal records. However, if intent to injure is considered to be an element required for an action to be aggressive, the observer must consider the emotional state of the aggressor and the cause or reasons for the action. In social learning theory, the antecedent is an important determinant of behaviour. The precise nature of the antecedent in a particular environment, as well as the reaction to the behaviour, determines whether it becomes a learnt way of acting.

Eight categories were devised for purpose:

  1. To gain recipient's attention, play, curiosity.
  2. Physical comfort - touching.
  3. Acquisitive, self-oriented, power, ownership of object, space, or person.
  4. Frustration - at game disruption, being blocked, at own lack of skill.
  5. Sadness.
  6. Rejection, fear of being disliked.
  7. Protecting friend, idea, maintaining rule.
  8. Wish to irritate, annoy, tease.


Table 1 shows the frequency of aggressive acts for each age group during inside and outside play for the two different settings. As it was not possible to keep the size of the groups observed the same, the average number of aggressive acts per child can be compared. Two different day-care centres were used for observations, so some age groups show two periods of thirty minutes' observing. One centre had all the age groups together for outside play.

There was very little aggression observed in the 2-year-olds. The size of these groups was small and two staff members were present. Several children were absent because of sickness in one centre so the children had more adult contact than usual. It was expected that the number of aggressive acts would diminish with age as children became more able to use language, but this was not the case.

Unlike the study of aggression in day-care children by Haskins (1985), the level of aggression is higher for the preschool children. It must be noted, however, that this table does not show the intensity, duration or outcome of aggression.

Table 1: Aggressive Acts by Age and Setting

Type of
AgePlaceNumber of
Number of
Per Child
Day care2In50

Table 2 shows the types of behaviours observed for each age group. The physical discrete category was the most common in all cases. This involved a deliberate strike in the form of a hit or kick to another person. The second most frequent category was facial which includes verbal. The term facial was used as it was felt that verbal aggression was accompanied by distinctive facial gestures whereas assertion was not.

Physical combined was chosen as a separate category as some events began so quickly, it was not an aggressive act by an individual. This category increased with age. The results support Hartup's (1977, p.245) hypothesis that threats to self-esteem would lead more frequently to hostile attempts to injure the agent of frustration than to object-oriented aggression, especially for older children.

Table 2: Method of Expression of Aggression by Age

BehaviourAge 2Age 3Age 4/5
Average Number
Acts Per Child
Average Number
Acts Per Child
Average Number
Acts Per Child
1. Facial00.060.23
2. Phys. dis0.060.170.28
3. Phys. comb00.090.12
4. Threaten000.05
5. Object00.060.14
No. Children1835100
Observation Time2 hours1 hour2 hours

Although the 4/5 age group had the highest frequency of aggressive acts, it can be seen by Table 3 that the likelihood of physical retaliation was quite small. Most of the behaviour was ignored or rewarded by a positive social response including play.

Table 3: Response to Aggression by Age

Age 2Age 3Age 4/5
1. Ignore0112
2. Positive social response0112
3. Retaliate453
4. Verbal aggression0110
5. Moved away195
6. Cry/sad006
7. Seek help311
N =1840100

The reason or purpose was allocated only to the initial act of aggression in the behaviour episode. It was not always possible to assign it because the reason was not always clear. Different interpretations would be made by different observers. It was actually included in the results because the observer was surprised that its main aim was so frequently play as opposed to frustration from blocking.

Table 4: Reason/Purpose of Aggression

of times
1. Attention, play, curiosity15
2. Physical comfort2
3. Acquisitiveness7
4. Blocking7
5. Sadness2
6. Rejection1
7. Protecting friend2
8. Wishing to annoy, tease3


Observations in the current study raise the following points:

(a) Recording numbers of acts of aggression without considering the purpose of it, may lead to false assumptions.
Much of the aggressive behaviour in the 4-5-year-old group was instigated as a part of play. The initial act was meant to hurt - but not too much. If it was ignored (perhaps because the other child was too engrossed in something else), it was likely to be repeated more strongly. It was often followed by smiles, laughter, cuddling, secret sharing, racing off to the next activity. It was not "real" aggression, but had the appearance of it. It could be misinterpreted by an adult, by the recipient, or a bystander (Record 2).

Children seem to learn very early that such acts could be in fun or meant. The means of differentiation are very subtle. Record 4 describes this for a 2-year-old. A 4-year-old said it depended on whether their hand was open (by demonstration) or shut. A 7-year-old said she would say "I'm going to dob" and if they were playing they would say so. Two-year-olds were observed pretending to hit and attack.

(b) Children learn to use verbal skills.
Young children do not have the finesse required by society in use of assertiveness skills (Record 1). However, these were observed in one 3 year-old (Record 6) and a 4-year-old (Record 5). As language skills develop, they are more able to express their feelings and not use physical means.

(c) Teachers provide a structure to reduce aggression.
Teachers stressed non-aggression by use of distraction - alternative toy/activity, shame, threat; rules, command to be friendly, removal, withdrawal of privileges, relating to self. Group rules were clearly stated. One 4-year-old was asked why he hit at home but not at kindy and said that it was not allowed at kindy!

The teachers of each group were able to identify one or two children who were more likely to aggress; at which times of day, day of the week and type of weather, e.g. late afternoon on a windy Monday was considered to be a prime time. Tiredness and sickness, changes at home, were expected to cause less tolerance to frustration. Lack of space was mentioned frequently as a factor. Some children needed more room to move. Transition times, i.e. moving from outside play to the bathroom and to lunch was difficult for some children. The teachers expected to anticipate difficulties and avoid them - by being present, having duplicates of most toys, and offering lots of activities specifically related to the age.

The environment and routine were designed to suit the needs of the age groups. There was plenty of space and choice of activity. The adults were warm, caring and available. These factors contributed to the low frequency of aggression. However, it was noticed that the tone of voice used by teachers to children they had labelled as aggressive, was different from that used with other children. Further study would be necessary to see if they were less warm, less flexible and more punitive of children they found difficult. This would be hard to measure because of the subtle differences in tone, touching behaviour and smiling but it could contribute to the different long-term self perception and feelings of rejection created in these children.

(d) Children behave differently in different situations.
Loeber and Dishion (1984) studied children who were identified as aggressive in the home and/or at school. There was a great difference between groups and only a small number were aggressive in both places. Several parents spoken to during the course of the observations reported frequent outbursts of aggression between siblings at home. Parents vary in their use of rewards and punishment; instigations of aggression; and the standards of behaviour as a role model. These are determined by personal and socio-cultural factors (Eron et al, 1971).

According to social learning theory, children learn aggression by selective imitation. This is supported by research evidence that individuals who observe more violence in the home and on television and, in particular, males who have the role model of violent heroes, will behave more aggressively (Eron, et al, 1971). Cummings (1985, p.497) found that angry behaviour was a highly arousing stimulus for aggression in others.

The learning is not purely imitative. It is reinforced if the aggressive behaviour is successful for the individual in gaining someone's attention, achieving the desired reaction from the recipient, or receiving the goods that were being fought over. Ringness (1975, p.81) sees it as a "reciprocal relationships between behaviour and the environment in that the environment may stimulate the behaviour but cognitive processes interpret the environmental stimuli, determine choice of behaviours and even help determine whether consequences of that behaviour are reinforced or not". Hence, parents can easily train their children to fight each other by providing particular responses. However, ignoring it permissively may lead to its increase (Patterson, 1979, in Cummings, 1985, p.495).

Children identify with other role models and imitate their behaviour. Gradually the individual's self-reinforcement is sufficient to continue the behaviour without reinforcement from others.

Countless studies have investigated the relationship between observing television violence and children's aggressive behaviour. Although the many independent variables made claims difficult, Bandura (1973, in Mussen, et al, 1969, p.337) showed that watching one violent scene made children carry out violence in a free play situation where other alternatives were available, more than a control group. Steur, Applefield and Smith (1971, in Liebert, et al, 1977, p.429) used a large group of children, paired on the amount of aggressive behaviour they normally exhibited in a play situation, to show that one child given dosages of violent television became more violent than the other. Similarly children could learn to increase their co-operative behaviour by observing such programs.

Parents can restrict the amount of violence children watch, but as noted by Chaffe (1972, in Liebert, et al, p.435) these children will then be the recipients of the other children's aggression. It is not just people more disposed to violence that increased their aggression. "For relatively average children from average home environments, continual exposure to violence is positively related to acceptance of aggression as a mode of behaviour." (Dominick and Greenberg, 1972, in Liebert, et al, p.433)

In the samples of behaviour observed, there was no distinct evidence of copying violence from television programs. Several S-year-olds were involved in a prolonged episode of fantasy play involving a wolf, but no guns or equivalent were used. None of the research studies has included verbal aggression as frequently demonstrated on soap operas, to see its effects on inter-family aggression.

Other studies have observed the effect of child-rearing practices on the amount of violence. Ashley Montagu (1978) compared the lifestyle and training of children in seven cultures where the adults are non-aggressive. The similar feature in these cultures was that aggression was observed in the children but strong moral codes taught that it was not acceptable. Fear of aggression, fear of anger, fear of strangers (and then fright) were methods used. The writer points out that the geographical and climatic conditions, the methods of food-gathering, the familial relationships and the religious customs supported and enabled the training to be appropriate. Studies that have investigated the relationship between sex roles and aggression have contributed to the concept that aggressive behaviour is learnt. Reasons offered for this are: difference in hormone level (Reinish and Sanders, 1986); girls aggression is ignored more frequently (Fagot and Hagan, 1985); girls are less affected by television violence because they think it is less realistic than boys do (Lefkowitz et al, 1977, in Huesmann, et al, 1984, p.748).

Blurton-Jones (1976, in Perlmutter, et al, 1982) considers rough-house play is similar to that in other primates and is essential for development of normal social and sexual patterns. Smith (1977, in Perlmutter, et al) says that one of its basic functions is to form and maintain friendships. The observations made included a large proportion of this form of "play" that could not be distinguished from aggression on the basis of descriptors alone. To differentiate, it was necessary to observe the reactions of the receiver and to judge the intent of the instigator.

The observations showed that descriptors for behaviours like hitting, kicking and yelling are not accurate as they do not describe the intent. Observing and categorising aggressive behaviour is complex. Deliberate hurtful acts occur rarely but aggressive behaviour is used in play. What is aggressive for one age group is not necessarily so later.

However, there is a large amount of evidence to support the theory that "whatever genetic potentialities we may have for aggressive behaviour, early conditioning in co-operative behaviour and the discouragement of anything resembling aggressive behaviour serve to made an individual, and a society, essentially unaggressive and co-operative." (Montagu, 1978).



Coding Sheet Version 1


Coding Headings:






Coding Sheet Version 2

As above, but extra items were added.


It was found that because the behaviour happened so quickly it was impossible to categorise it instantaneously. It was better just to list it. New behaviours were added, e.g. flicking sand in a child's eyes.

The first version does not show intensity. It was attempted to show this on a rating scale of 3, so


meant the first child hit very lightly. The recipient returned it with a hard hit.

A lot of information about the event was lost by this style of categorisation and it requires immediate subjective judgments from the observer.


Shorthand Version of Running Record

In - 4/5 9.53 b H1 b on head with saucepan - I

rep.- I

H2- said "don't" cont with play

b THR with ladle


rep. laughed- smile, grabbed soup in the air


Inside play with 4/5 age group at 9.53 a.m.
Boy hit boy lightly on head with saucepan. He was ignored. He repeated it twice. He hit harder. The recipient said "don't" and continued to play. He threatened with a ladle and was ignored. He repeated it twice, laughed. The recipient smiled and grabbed soup in the air.


Event Running Record

1. Inside 4/5 years preschool
9.42 J body pushed D, ignored J, K laughing, teasing, pulling dress-up clothing off D. D resisting, took off remainder - sad face, put out, stomped away, J followed. D turned and pushed J. Wry look. D tongue poked at J. J. hit lightly on shoulder. D moved across room. J followed. Asked D to play with construction toy. D turned head away, walked backwards to another area; grabbed armful of wooden shapes; D singing; J followed; parallel play. J. building road and bridge identical to D without interaction. 9.44 D said "It's much more better eh."

2. Outside 3-5 years day-care
10.44 Two boys punching; aiming feet. Aden ran up, kicked one of the boys; grabbed him attempting to hold him down. Teacher said "They are just playing Aden." Aden helped the boy up. Ran after the other - chased each other. Faces serious, concerned. On being caught, cuddled and laughed. 10.45.

3. Inside 3-5 day-care
11.43 Girl G bit a puppet's hand, yelled at it. Used puppet to pull the hair of the girl S beside her. S said "OW!" kept on at puzzle. G repeated, pulled hair hard several times. G yelled at puppet "Stop it." G gave puppet to S. S pulled its nose. G took it back and said "S he's being good. " G used puppet to take out the puzzle. S moved away from the table.

4. Inside 2 years day-care
9.40 Girl J hit another girl A very lightly as she passed. A stopped, looked intently at J's face. J was scowling. A hit back lightly. J hit back a little harder. A hit back at the same intensity. Repeated twice. Separated by events.

5. Inside 4-5 day-care
10.20 Boy S approached group of three children playing with construction toys on a mat. "Well, well, well. Who broke my thing" he said with hands on his hips and pulling himself up in height.
"You did" said J.
"I didn't" said B. S walked around between the children looking at each one.
"I didn't do it" said B. "It was G (not there).
"He did not" said J.
Teacher from afar said, "J use your inside voice."
"What was it? Just tell me" said S, still moving imperiously.
"Not me" said J.
"But who stealed it? Who? Just tell me" said S.
S moved back to another area.

6. Inside Kindergarten
9.25 "Let's do a painting" said boy 1.
"I'm not going to do a painting, Michael".
Continued mild interest in cars and roads just built.
"What do you want to do after we paint?" said boy 1.
"I want to be in a circus" said boy 2.
"Why don't we paint a circus!" said boy 1.
Ignored. "I'm going to paint a circus" said boy 1. "Want to watch?"


Averill, J.R. (1982) Anger and Aggression, Springer Series in Social Psychology, New York: Springer- Verlag, 1982.

Berndt, T.J. and Bulleit, T.N. (1985) "Effects of siblings relationships on preschoolers behaviour at home and at school", Developmental Psychology, 21: 761-767.

Condry, J. and Ross, D. (1985) "Sex and Aggression: The influence of gender label on the perception of aggression in children", Child Development, 56: 224-233.

Cummings, E.M. (1985) "Influence of conflict between adults on the emotions and aggressions of young children", Developmental Psychology, 21. 495-507.

Deluty, R.H. (1985) "Consistency of assertive, aggressive and submissive behaviour for children", Journal of Personality and Social Psychology, 49: 1054- 1065.

Eron, L., Walder, L. and Lefkowitz, M. (1971) Learning of Aggression in Children, USA: Little, Brown and Co.

Fagot, B. and Hagan, R. (1985) "Aggression in toddlers: Response to the assertive acts of boys and girls", Sex Roles, 12: 341-351.

Hall, W. and Cairns, R. (1984) "Aggressive behaviour in children: An outcome of modelling or social reciprocity?" Developmental Psychology, 20:739-745.

Handal, P. and Hopper, S. (1985) "Relationship of sex, social class and rural urban locale to preschoolers AML scores", Psychology Reports, 57: 707-713.

Hartup, W. (1977) "Aggression in Childhood - Developmental Perspectives", in Contemporary Readings in Child Psychology, E. Hetherington and R. Parke (eds.), USA: McGraw-Hill.

Haskins, R. (1985) "Public school aggression among children with varying day-care experience", Child Development, 56: 689-703.

Hilgard and Atkinson (1967) Introduction to Psychology, 4th ed., USA: Harcourt.

Huesmann, L., Lagerspetz, K. and Eron, L. (1984) "Intervening Variables in the TV violence-aggression relation: Evidence from two countries", Developmental Psychology, 20(5): pp.74~775.

Liebert, R., Davidson, E. and Neale, J. (1977) "Aggression in childhood: The impact of TV", in Contemporary Readings in Child Psychology, E. Hetherington and R. Parke (eds.), USA: McGraw-Hill.

Loeber, R. and Dishion, T. (1984) "Boys who fight at home and school: family conditions influencing cross-setting consistency", Journal of Consultative Clinical Psychology, 52: 759-768.

Lyons, J. and Serbin, L. (1986) "Observer bias in scoring boys' and girls' aggression", Sex Roles, 14: 301-313.

McCoy, C. and Masters, J. (1985) "The development of children's strategies for the social control of emotion", Child Development, 56: 1214-1222.

Medinnus, G. (1976) Child Study and Observation Guide, USA: Wiley.

Montagu, A. (ed.) (1978) Learning Non-Aggression - The Experience of Non-Literate Societies, Oxford: Oxford University Press.

Mussen, P., Conger, J. and Kagan, J. (1969) Child Development and Personality, 3rd ed., New York: Harper.

Patterson, P. (1980) Pro-Social Behaviour of Two-Year-Old Children in Mixed and Same Age Settings. Unpublished paper, conducted by Lady Gowrie Child Centre, Brisbane.

Perlmutter, M. and Lamb, M. (1982) Child Psychology Today, USA: Random House.

Prochaster, J. (1985) "Children's views of the causes and cures of sibling rivalry", Child Welfare, 64: 427-433.

Reinish, V.M. and Sanders, S.A. (1986) "A test of sex differences in aggressive response to hypothetical situations", Journal of Personality and Social Psychology, 50: 1045-1049.

Ringness, T.A. (1975) The Affective Domain in Education, New York: Harper.

Stark, E. (1986) "Day care and aggressiveness", Psychology Today, 20:68.

Younger, A.J. (1985) "Age-related changes in children's perceptions of aggression and withdrawal from their peers", Developrnent Psychology, 21: 70-75.

Please cite as: Norton, R. M. (1988). Aggression in young children. Queensland Researcher, 4(3), 21-40. http://www.iier.org.au/qjer/qr4/norton.html

[ Contents Vol 4, 1988 ] [ QJER Home ]
Created 11 Mar 2008. Last revision: 11 Mar 2008.
URL: http://www.iier.org.au/qjer/qr4/norton.html