Research utilizing experimentation is undertaken in a variety of contexts and settings. Decisions concerning the circumstances under which to conduct an experiment typically reflect a combination of considerations including the nature of the research question, the availability of research resources, and the researcher’s interest in balancing concerns about the validity and the generalizability of subsequent findings. Overwhelmingly, the most common setting for experimentation in communication is the laboratory.
Advantages Of Laboratory Experiments
Laboratory experiments, when effectively operationalized and carried out, afford strict experimental control by allowing for isolation of the research situation from the variety of extraneous influences that can impact both experimental treatment or intervention (i.e., independent variable) and the subsequent outcome (i.e., dependent variable). Accordingly, laboratory experiments are typically structured on the more rigorous “true experimental designs” and, consequently, yield the strongest evidence of causality (Wimmer & Dominick 2003).
An array of locales can be utilized in staging laboratory experiments ranging from general purpose accommodations such as conference rooms, classrooms, lecture halls, and theatres to facilities specifically designed for experimentation. It is the researcher’s ability to structure and manipulate the experimental environment (e.g., lighting, temperature, soundproofing, seating arrangement of research participants), not the specific locale, that is the defining characteristic of experiments in the laboratory setting.
Beyond situational and environment control, the laboratory setting allows substantial control over all aspects of the research process. Working in a laboratory setting, for example, significantly enhances the researcher’s ability to accurately identify eligible research participants, insure their random assignment to treatment conditions, and extensively observe their progression through the research activity. The level of specificity achievable in the operationalization of independent variables – or, in other words, the extent and certainty with which treatment manipulations can be accomplished – and the consistency of their reenactment are both extremely high in laboratory experiments. Equally important, the degree of precision possible in outcomes (i.e., dependent variables) assessment, which promotes measurement reliability, is a key aspect of experimentation in laboratory settings (Kerlinger 1986).
Disadvantages Of Laboratory Experiments
While many aspects of laboratory experiments offer crucial advantages in research endeavors testing causal relationships, experiments in laboratory settings can also involve potential disadvantages. Perhaps the most obvious shortcoming is operational and environmental artificiality. Laboratory experiments facilitate the precise and systematic observation of human reactions under controlled conditions; but sometimes the experimental situation and/or experimental procedure is rather sterile and unnatural. Some behavioral and perceptual outcomes observed under such circumstances can have little direct application to those occurring in natural surroundings. In extreme cases, laboratory experiments can involve such artificiality that little or no confidence in the external validity and generalizability of resulting observations is warranted. Consequently, several research questions simply do not lend themselves to examination through laboratory experiments (Miller & Salkind 2002).
Because laboratory experiments typically involve extensive interaction between the researcher and research participants, the potential is great that researcher biases can emerge as threats to internal and external validity. Two forms of subjective bias, that unconsciously and unintentionally affect laboratory experiment results, are experimenter bias and observer bias. Experimenter bias is introduced into the experimental process when the researcher subtly communicates expectations about outcomes to the research participants. Observer bias occurs during outcome measurement when the researcher overemphasizes expected behaviors and ignores unanticipated ones. Blinding methods are frequently incorporated into experimental procedures to avoid such distorting influences. The double-blind method, in which neither the researcher nor the research participants know who belongs to the control or experimental groups, is most common (Myers & Hansen 2005).
An Example Of A Laboratory Experiment
Laboratory experiments have proven instrumental in communication research enlightening both our understanding of basic communication phenomena and informing theory construction. The scholarship of Judee Burgoon, James Dillard, Robin Nabi, and David Roskos-Ewoldsen provides numerous examples. Perhaps the most sophisticated and diverse utilization of laboratory experiments in communication research emerges from the work of Dolf Zillmann.
One study in particular, entitled Effects of an opposite-gender companion’s affect to horror on distress, delight, and attraction (Zillmann et al. 1986), provides an informative illustration. The study was built on the observation that teenagers love to be terrorized – at least at the movies. Most of the audience for slasher or slice-and-dice horror films are adolescents and young adults. Further, the horror film audience is typically evenly split between male and female viewers. This is not coincidental, argued Zillmann and his colleagues, noting that horror films seems to serve an important function as “date movies” for young audiences. As one teenager, quoted in Seventeen, put it: “Sometimes you feel weird and self-conscious holding onto a guy’s hand on the first date but this way you can just grab him.” Another teenage girl observed, “Guys like to take you to horror movies, hoping you’ll be real afraid and need them to comfort you.” A third said, “You can get all rowdy with boys and jump into their lap.”
Casting these observations within the framework of a gender role socialization of affect model (Zillmann & Weaver 1996), Zillmann and his colleagues designed a post-test-only control group design experiment. Specifically, they implemented a 2 × 3 factorial design with research participant gender (male, female) and the affective reactions to a horror film that were enacted by an opposite-gender companion (mastery, indifference, distress) as independent variables. Male and female college students watched, with an opposite-gender companion, several of the final scenes from the horror film Friday the 13th, Part 3. The four companions (one with lower and one with higher physical appeal within each gender) participating in the study were actually research confederates; students who were trained by the experimenters to enact mastery, indifference, or distress at critical moments throughout the film segment. This behavioral enactment by the research confederates provided the crucial independent variable manipulation (i.e., experimental treatments) across which the research participants were randomly assigned.
Several outcomes were measured following the experimental treatments. These dependent variables included the research participants’ affective reactions to the movie; their perceptions of the companion’s physical appeal, personality traits, and desirability as a working partner; and the research participants’ tendency to acquiesce to apparently erroneous contentions on the part of the companion.
The findings from this laboratory experiment revealed that men enjoyed the film most when their female companion acted distressed and least when she displayed mastery. Women, in contrast, enjoyed the film most when viewed with a mastering male and least in the company of a distressed male companion. The intensity of distress experienced by research participants in response to the film followed the same pattern.
The enactment of mastery while co-viewing the horror film did not enhance the physical appeal of female companions. Perceptions reported by female research participants of the lower-appeal male companion were enhanced significantly, however. The lower-appeal companion also benefited from the display of mastery in that pronounced positive traits were ascribed to him by co-viewing female research participants. Male research participants co-viewing with female companions did not yield a significant pattern of perceptional shifts.
And, finally, the display of distress in response to the horror film segment reduced the desirability of both male and female companions as working partners. Furthermore, female research participants showed a clear tendency to acquiesce to assertions made by male companions who had shown mastery while viewing horror. Taken together, the results of this laboratory experiment support both the conventional-wisdom observations of teenage movie-goers and predictions derived from theory. They highlight a complex interaction between the mass media and their audiences and show that the social context in which mass media content is consumed can have a significant influence on perceptions of and reactions to both the media message and media users.
- Kerlinger, F. N. (1986). Foundations of behavioral research, 3rd edn. New York: Holt, Rinehart and Winston.
- Miller, D. C., & Salkind, N. J. (2002). Handbook of research design and social measurement, 6th edn. Thousand Oaks, CA: Sage.
- Myers, A., & Hansen, C. H. (2005). Experimental psychology, 6th edn. Belmont, CA: Wadsworth/ Thomson Learning.
- Wimmer, R. D., & Dominick, J. R. (2003). Mass media research: An introduction, 7th edn. Belmont, CA: Wadsworth/Thomson Learning.
- Zillmann, D., & Weaver, J. B., III (1996). Gender-socialization theory of reactions to horror. In J. B. Weaver, III & R. Tamborini (eds.), Horror films: Current research on audience preferences and reactions. Hillsdale, NJ: Lawrence Erlbaum, pp. 81–101
- Zillmann, D., Weaver, J. B., III, Mundorf, N., & Aust, C. F. (1986). Effects of an opposite-gender companion’s affect to horror on distress, delight, and attraction. Journal of Personality and Social Psychology, 51, 586 – 594.