Finnish longitudinal study on parent-child communication, sex, age and direct aggression (Wallenius & Punamäki, 2008)

Poor Wallenius and Punamäki… They conducted a longitudinal study on violent video games, had their study published several weeks before Anderson et al. (2008) and nary a squeak from the English media on their psychology study. Some might say that this study bears more merit than Anderson et al.’s because they’ve taken some aspect of parenting and child development into consideration. I do wonder once more about how sensationalism trumps serious journalism, it makes me think that the world is in a rut.


This study investigated the roles of sex, age, and parent-child communication in moderating the association between digital game violence and direct aggression in a two-year longitudinal study. Finnish 12- and 15-year-old adolescents (N = 316) participated in the follow-up survey. As hypothesized, digital game violence was linked to direct aggression both longitudinally and synchronously, and the link was moderated by parent-child communication in interaction with sex and age. Results suggest that the moderating role of parent-child communication changes with increasing age. Poor parent-child communication may be one of the factors in an adolescent’s development that may strengthen the negative effects of digital game violence, but even good parent-child communication does not necessarily protect the adolescent in the long run. Digital game violence seems to be one of the risk factors of increased aggressive behavior.

I decided to kill a few trees to read the article on paper rather than the computer since the screen glare makes reading slower. It also helps me to highlight some passages and write notes on it. Something that’s not possible on adobe reader.

The authors investigated on several factors that they argued would moderate the relationship between violent video game use and direct aggression (e.g. physical and verbal). These factors are sex, age and parent-child communication.

Sex is one important factor in that males are consistently more physically aggressive than girls. That is all.

Age is rather important at certain developmental stages, especially adolescence where many psychological and physiological changes occur (i.e. puberty). According to their literature review, physical aggression peaks between the ages of 13 and 15, then decreases. Please note that age is a biological factor, other factors include personality, situational, cultural, and environmental factors. In the case of violent video games, I’d say it falls under situational factors. Situational because they are exposed to violence which can influence a person’s thinking leaning towards hostile feelings and aggressive thoughts. IMO, it would be interesting if violent video games might exacerbate aggression among adolescents and perhaps contribute to its stability from a developmental perspective.

Parent-child communication, which seems to be Wallenius’ specialty, is another situational factor. Let’s see… it might seem easy to understand what it means, but I have to be sure of its definitive definition. So, it sounds like the state of relationship between the child and parent. So the way I see it is that a good parent-child communication means that they have good family relations, communication is two-way, non-confrontational and authoritative parenting. Whereas poor parent-child communication means that they have bad family relations, communication could be one-way, confrontational and authoritarian parenting. So poor relations mean more likelihood of increased aggression for youths than those with good relations. Another point to consider is how the parent’s importance within a child’s development changes as the child ages. So it’s quite logical that parents are extremely important for a baby, a gray area for teenagers (kinda like a love-hate relationship), and adults, well, they’re grown up, so they can live on their own, unlike some man-child I know of. IMO, parents might be a critical factor on a child’s development of aggression and may become a protective factor from the effects of violent video games.


Participants: Since this is a longitudinal study, I will be referring to time frames for the participant sample. At Time 1 (spring 2004), there are 222 4th graders (M = 10 years old) and 256 7th graders (M = 13 years old). At Time 2 (two years later), there are 132 6th graders and 184 9th graders. The reason for Time 2 having fewer participants is called attrition. Many things can happen in two years, for example, moving out of town, death, no longer interested in study, etc. In addition, this is study is conducted in Finland.


Violence in video games: One question asked how often they played games that contained violence (i.e. killing, fighting, kicking, etc.) on a 4-point scale. Next is how often they played certain gaming genres, specifically “action, fighting and shooting”, again on a 4-point scale. I have no complaints, seems simple for participants to answer and it doesn’t matter if you have someone who never played a fighting game because the three genres contain violence and that’s what they’re looking for.

Parent-child communication: They used the Parent-Adolescent Communication scale which consist of 14 items answered on a 5-point scale. 7 items consist of statements on open/parent communication, so positive parent-child communication (e.g. mom is always a good listener), and the 7 other consist of statements on problematic/parent communication (e.g. mom tells me things that make me feel bad). The scale has been used before.

Direct aggression: That’s what’s written in the article. But it’s basically direct physical aggression. 10 items on a 5-point scale is used. Somewhat better than Anderson’s longitudinal Japanese sample where one sample group had only used one item to measure aggression. Something that would irritate me.


Consent for participation is given by the participants and their parents. Basically, questionnaires are given in the classroom with a researcher who would answer any questions. Two years later (Time 2), the same procedure as in Time 1. There are some procedural differences between Time 1 and Time 2, but they are quite trivial.


The abstract isn’t enough to describe the results, so I’ll give noteworthy details. But remember, it’s always better to read the article directly than my post.

From their preliminary analysis, boys reported more violent video game play and direct aggression than girls at both Time 1 and Time 2. No surprise. Older children reported more violent video game play, direct aggression than younger children and poorer parent-child communication than younger children. But if we look at Time 1 data to Time 2 data, it’s no surprise why older children reported such difference to younger children since violent video game use increased and parent-child communication become poorer for the whole sample. Although, nothing was revealed about direct aggression or maybe I’m missing something.

To examine violent video games’ role in direct aggression, they conducted regressional analyses and used a statistical technique called destructive testing approach, used by Anderson. It basically starts with a target predictor variable (in this case violent video games play) to a criterion variable (direct aggression), then we add relevant variables (age, sex and parent-child communication) to see if the target predictor variable is still statistically significant to the criterion variable. So if the predictor variable fails to achieve significance after one addition, then it’s a weak predictor, if not after several or all additional variables then it becomes a strong valid predictor or link.

First up is the longitudinal link, our target predictor variable is violent video game play at Time 1 and our criterion variable is direct aggression at Time 2. Before I start, I should relearn what explained variance means. Our target predictor variable started up statistically significant which accounted 11.6% of the variance for direct aggression. Adding sex, it dropped down to 2%. Next is age, it dropped down to 1.4%, then parent-child communication (Time 1 and Time 2), it finally failed significance when direct aggression at Time 1 was added, which dropped to 0.2%. From my understanding, after adding sex so violent video game play would account for 2% of children’s aggression score. So it seems sex is a bigger predictor for direct aggression than violent video games, although violent VG is still a statistically signicant and minor predictor, but still warrants attention.

Looking at Time 2 violent video game play and direct aggression. Following the same pattern, it started at 12.2%, then dropped to 2.4%, then 1.8%, then 1.5% and finally collapsed at 1%. Again, IMO sex matters in aggression.

There’s an interesting remark from the authors, the effect size of violent video games on direct aggression between Time 1 and Time 2 are approximately the same. I can’t really use it as an argument that increasing graphical realism has no effect on violence because they’re dealing with children and later on adolescents, the games they play change according to their development and graphical development is not equally distributed, side-scrolling shooters place more emphasis on fanciness whereas FPS are heading towards “grittiness”, “realism” and shades of grey or brown (Ben “Yathzee” Croshaw, 2008). Thinking about it, kids would lean more on Mario than Halo, but in addition we need to know kids’ preference for graphical realism. Then we have the parents, whose role in kids’ gaming choices is unknown in this study, but it’s quite possible they have a significant influence. The more I think about it, looking at Kutner and Olson’s Grand Theft Childhood data, the violent games kids play are those that are popular and not those are ultra-violent like Postal or Manhunt series. I don’t remember Resident Evil being popular among kids. Hmmm… I wonder if kids stay away from horror-themed media? Finally, I don’t think there are any significant graphical difference between 2004 and 2006, but then again I’m not a game journalist or developer and I’m not a console player, more of a PC player.

The authors then conducted hierarchical multiple regression to check out how sex, age and parent-child communication moderated the link between violent video game play and direct aggression. As the abstract said, there is some mixed results. The main effects are that poor parent-child communication is correlated with higher direct aggression, both longitudinally and in each time points.

Let’s start longitudinally, among girls who reported good parent-child communication and played a lot of violent VG at Time 1, they reported a decrease in aggression at Time 2. Whereas those with poor parent-child reported an increase at Time 2. They also found that among younger girls, violent video game play and parent-child communication at Time 1 doesn’t have any significant effect, whereas among older girls it followed the general pattern mentioned earlier. However, the authors noted that younger girls’ level of violent video game play is low in comparison to boys, so this may have affected the results. So we can’t say that exposing girls to violent video games would be beneficial since girls don’t play as much violent video games as boys do. In any case, the authors argued that based on previous studies there are no age differences between violent video game play and direct aggression.

The relationship for boys are in contrast to the results found in girls: boys who play a lot of violent video games and reported good parent-child communication at Time 1 had reported an increase in direct aggression at Time 2, whereas those who reported poor parent-child communication reported a decrease.

Analysis on Time 2 data only, with direct aggression (Time 1) as a control variable. Among girls, parent-child communication has no moderating effect between the violent video game play and direct aggression. So girls who play a lot of violent video games reported higher direct aggression, regardless of parent-child quality. Interestingly, those with good parent-child communication have a lower average score than those with poor parent-child communication. For boys, it seems that those with good parent-child relationship and play a lot of violent games reported higher aggression than with poor parent-child communication.

The following are my mumblings and ramblings, may or may not contain traces of sound theories.

Now I must say this: I don’t how much of an increase or a decrease in their aggression scores. It is also noted that the average aggression score for good parent-child communication is lower than those with poor parent-child communication. So, it would mean that parent-child communication is a protective factor and a moderating factor in the relationship between violent video game play and direct aggression. It is noted that those who reported good parent-child communication at Time 1, there was an increase in direct aggression from Time 1 to Time 2. According to the authors, early good parent-child communication may have delayed children from playing more violent video games and the decrease in parent-child communication is due in part to children’s growing sense of autonomy in which changes the parents’ importance in children’s development, well they are growing teenagers, they can’t be wholly reliant on their parents forever.

As for those who reported poor parent-child communication, the authors suggested that it may have been due to emotional desensitization to violent video games, but I don’t get it how it relates. IMO, I’d say there is a gender difference in the importance of parent-child communication where girls are more responsive to the quality of relationships than boys do. For example, if you compare boys’ and girls’ social network and how they spend time with friends. So, girls would feel much more of a sting in relationships and would behave more aggressively if they have poor relationships than boys do. In addition, girls probably learn more on social skills from their parents than boys which may explain the increase in aggression scores for those with poor parent-child communication. And girls in good parent-child relationships would have learned to foster good and nice behaviours in order to maintain good social relations.

Another point to consider is what are the components that make a good or a poor parent-child relationship (as the authors had suggested as a limitation)? Would there be parents who would lower their children’s self-esteem, autonomy, self-efficacy in general, assertiveness or factors that they have affected their children’s aggression, especially for boys. Perhaps, these parents are psychologically controlling by not letting boys more autonomy which may have an effect on their aggression scores. But then again, having a poor parent-child communication entails some hostility, which in some sense, would made children probably more aggressive. Taking a different angle, perhaps the lack of authoritative parenting at Time 1, may have made those boys more unruly (aggressive), then later on they learned to control themselves (probably from peers and school to compensate the low of socialization (does such a concept exist?)) despite playing a lot of video games, or better yet that playing more violent video games may have brought them more friends in which they would have a greater socialization role in becoming less aggressively, albeit still more aggressive than boys with good parent-child communication.

There are limitations that the authors noted: all the measures are based on children’s report, they would have like to get some parents and teachers’ reports. They also noted that their measures in violent video game is not an easy objective measure (I would have said they might be dissatisfied). Their study does not necessarily reflect a causal relationship. Other parenting behaviours, such as parental monitoring, may have greater moderating role in the relationship between violent video game play and direct aggression. Another interesting limitation they wrote and that I’d like to explore is children’s interpretation of violent video games in relation to their attributional style (badly paraphrased).

So the take-home message is playing violent video games is a risk factor for aggression, albeit a low one IMO if you take sex, age and quality parent-child communication. In addition, parent-child communication can be a risk or protective factor depending on the quality and child’s developmental stage.

Wallenius, M., & Punamäki, R. (2008). Digital game violence and direct aggression in adolescence: A longitudinal study of the roles of sex, age, and parent-child communication. Journal of Applied Developmental Psychology, 29(4), 286-294.


4 thoughts on “Finnish longitudinal study on parent-child communication, sex, age and direct aggression (Wallenius & Punamäki, 2008)

  1. Pingback: Gaming parents, action videogames and civic engagement (Ferguson et al., 2010) « VG Researcher

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s