Humor in violent movies Introduction

Cynthia King’s article was selected for this assignment for two reasons: it shows us both the benefits of the scientific approach, as well as its limitations. One of the benefits is that her work illustrates the relevance for theory: the researcher emphasizes the relation between her study and several theories in the field of psychology and film studies. Notice that King uses contradictions between theorists as an inspiration for her research. For instance, on page 9, we see that there are reasons to assume that humor either enhances or reduces distress. A common complaint about empirical studies is that it produces trivial results, proving the obvious. It is better to avoid such criticism, as King does, by showing that the claims you examine are far from obvious.

Besides contradictions in theory, King finds inspiration for her own research in her critical reading of previous studies. In her article she looks at problems in relevant research and the questions that are left unanswered. Of course, in her own study she attempts to solve these weaknesses. This is exactly the purpose of the present assignment: finding out where we can improve on the research that is out there.

Before we move on to the discussion of our critical reading, let us first point toward three formal aspects of King’s report. The first is that some of the formulations suggest a bias on the part of the researcher. Three examples from King’s text: first, she reveals her concerns by suggesting that some of the “wisecracks” that we hear in movies may not always be “innocent”. Second, in her interpretations of the results she clearly shows that she is not neutral. What does she believe? Is it a bad thing that viewers experience less distress when watching fictive or non-fictive violence? Third, on what else than her own values does she base her idea of “socially responsible film strategies”? (p. 21). In your own study, try to avoid showing your personal norms and values in your own report. However, take a look at the place of this last example in the article: It is in the discussion section that she discusses the moral consequences of her findings. As a rule of thumb proceed on the principle that you have more liberty there to show your concerns. The function of the discussion section is, in part, to make clear to your readers what the possible implications of your results are.

A second aspect about the formulation of the text is that King mixes up terms like theory, hypothesis/prediction, and explanation. For students of the Humanities it is sometimes difficult to understand the ease with which the term “theory” is used in psychological literature. As an example, look at the disposition theory that states that your disposition toward someone determines how you will respond to that person. In comparison to some of the theories known to you in the Humanities, this seems a hypothesis rather than a theory. On page 10 King uses the term ‘explanation’ where she should use ‘prediction’: Does she explain why distress would be reduced by wisecracks? Take good care that you do not confuse such terms.

The final comment that we would like to make on the formal aspects of this article is that not all the results are reported. The author states that some findings were left out because they were “worth nothing”. It would have been better to give a full account of all the findings and let readers judge for themselves.

1. Conceptual validity

Estimating the conceptual validity of an experiment requires you to look at the dependent as well as the independent variables. On the side of the independent variables, we see that King does not worry enough about what humor is. She does not offer a thorough reflection on the concept.

Looking at the way King operationalizes some of her dependent variables it becomes clear that this is slightly problematic. For example, the predictability of a joke may vary from person to person: what is predictable for one may be surprising for someone else. The same holds for “serious”: the way this term is interpreted may vary from participant to participant. Nevertheless, King assumes that all participants will understand these terms in the same way. One of the disadvantages of working with closed questions is that we have to assume, as King does, that all participants will understand exactly what you mean by these terms, and that they all interpret your questions in the same way.

2. External validity

The participants in this study are all part of a small portion of the (world) population: American undergraduates. Obviously, it may be hard to generalize findings to other audiences. This is, however, a problem most studies suffer from, which makes our observation superfluous. What you should aim for in your criticism are the specific reasons why you believe the results are irrelevant for other populations. One point that could be made in this respect is that it seems likely that King’s sample is far more familiar with humor in violent films than most other audiences. It is unclear what this would mean in terms of effects. Maybe spectators’ with less experience would be shocked by humor in violent movies, whether used by heroes or villains.

But there is more. How did participants end up in King’s sample? Did they volunteer? Note that volunteers may differ from other subjects from the same subgroup. Because we do not know whether they were randomly chosen from a larger sample there is little we can say about the generalizability of the conclusions.

The design of the study does not allow conclusions about effects of frequent exposure to the type of movies King talks about. However, she does make claims about such effects (see what she says on page 21, 22, about “film buffs”).

Apart from these problems, we should note one important advantage of King’s approach: using existing films rather than experimenter-generated materials. This greatly enhances external validity. However, as you will see below, this comes at a price.

3. Internal validity

An essential independent variable in King’s conceptual model is the funniness of jokes. However, as anyone knows who ever listened to jokes, or told jokes themselves, not all jokes are equally funny. Or rather: some are considered funnier than others. King wants to know whether it makes a difference who (crook or hero) tells the joke; but in that case it should have been the same jokes. Now the causal relation between the moral valence of the characters and the ‘effects’ of their humorous remarks is problematic. Having a hero or crook tell the same joke probably requires researchers to make their own material – and this would reduce the ecological validity. Of course King did attempt to be precise in her manipulation. But as her examples reveal, it must have been difficult to manipulate the material without serious distortion. See page 14, where the hero says “or this,” but it must be a mystery to the participants what ‘this’ refers to, because the preceding sentence was deleted. King’s report does not suggest that she checked whether participants noticed the deletion, let alone whether this affected the results.

Another factor that may have been a distorting factor is participants’ genre preferences. Fans and non-fans may respond quite differently. It looks as if King did not ask her respondents whether they liked this type of movie. She might have used this information to enhance control over the causal relations she investigated.

Finally, participants should have been assigned to their condition (the version they saw) based on pure chance. Instead they entered a list for one of the groups themselves, and these groups were assigned to the film versions randomly. Chances are larger now that one particular type of participants was over-represented in one of the groups.