Social Relations Model Definition
The social relations model is a theoretical and statistical approach to studying how people perceive others. Although investigations of person perception have a long history in social psychology, early methodological approaches relied on research participants reporting their perceptions of fictitious others who were described in brief stories. The social relations model allows researchers to move beyond such vignette studies and address a variety of questions related to interpersonal perception while studying real people engaged in real social interactions.
Social Relations Model Background and History
Perceptions of other people are fundamental components of social interactions and, therefore, have a prominent place in social psychology. A person must perceive other people’s traits accurately so that he or she can predict how they will behave. If you correctly perceive that someone is friendly, then you can probably expect that person to help you. What is more, people should also value knowing what other people think of them. For example, knowing that someone doesn’t like you might be useful so that you can avoid interactions with that person. Such beliefs about how others perceive one’s self are termed metaperceptions. Because person perception is so basic to social interaction, researchers have conducted many studies to learn how people form perceptions (Is John seen as friendly?), the attributions people make following perceptions (Why is John seen as friendly?), and the relative accuracy or inaccuracy of perceptions (Is John really friendly?).
Many of the early person perception studies, however, relied on vignettes, or stories, about imaginary other people. So, for example, a research participant might be given a paragraph that purportedly describes another student. After reading the paragraph, the participant would be asked to report his or her perceptions of the student in the story. Using such an approach makes person perception akin to object perception. That is, the target person becomes static and noninteractive, no different than perceptions about a chair or book. The vignette method has a clear advantage in that the researcher can control and manipulate the information that participants receive about the target person. Yet relying on written descriptions of another person removes much of the richness of real social interactions. Some researchers have improved the vignette approach by using videotapes of a person’s behavior, which allows for a more vivid portrayal of the target person. Regardless of whether the vignette is presented as written or videotaped, participants in these studies know that the perceptual process is a one-way street. That is, although participants can make perceptions of the person in the vignette, the fictitious character cannot make a perception of the participant. Therefore, it would be unrealistic to ask participants to report on metaperceptions in such circumstances.
To further enhance the vignette method, some researchers have used assistants (called confederates) to pretend as if they are participants and engage the real participant in a seemingly authentic social interaction. Thus, participants believe that they are having an active, spontaneous interaction with another person. Under these conditions, participants should be able to report both perceptions and metaperceptions. Yet even a confederate approach to person perception is limited because the research assistant is generally required to play a prespecified role and use scripted responses. Thus, regardless of the specific approach described earlier, the researcher is not able to study the real give-and-take of an unscripted social interaction.
Given the problems associated with vignette and confederate approaches to studying person perception, one might conclude that an easy solution is to study real interactions between real people. Unfortunately, the major strength of such an approach (the give-and-take of real interactions) leads to challenges when interpreting data. Specifically, the perceptions in a real interaction depend on a variety of factors, including the particular person making the perception, the particular person being perceived, and the unique relationship between those two people. For example, John’s perception of Mary’s friendliness could be due to the way John perceives most people, the way Mary is perceived by most people, or something about the specific relationship between John and Mary. This situation is often referred to as a problem of nonindependence between the perceiver and the target (e.g., John’s perception of Mary might be dependent on John, Mary, or both people). Thus, until the development of the social relations model, it would have been difficult for most researchers to account for these different components of perceptions that are generated by real social interactions.
Details of the Social Relations Model
David Kenny developed the social relations model to give researchers a means to account for the nonindependence of perceivers and targets that emerges from real social interactions. The social relations model provides a theoretical way to conceptualize interpersonal perceptions, a methodological guide for designing studies that use real interactions, and a statistical approach to analyze the data from these studies. According to the model, interpersonal perceptions are a function of five components: a constant, a perceiver effect, a target effect, a relationship effect, and error. These components, described further later, can be summed to yield the overall perception or metaperception. So John’s perception of Mary’s friendliness (P) could be described with the following equation:
P = constant + John’s perceiver effect + Mary’s target effect + relationship effect + error
The constant is the average score on the perception across all perceivers and all targets. Perhaps, on average, people are seen as somewhat friendly. The perceiver effect is how a participant views others in general. For example, John might view everyone as very friendly, including Mary. The target effect is the degree to which a person elicits a certain perception. So Mary might be somewhat reserved and, in turn, be rated as less friendly by most of her interaction partners. The relationship effect represents the variance caused by the unique combination of a specific perceiver and a specific target. John may view Mary as especially friendly beyond his perceiver effect and beyond her target effect. Although social relations model analyses often lump together the relationship effect and error, it is possible to separate these two components. If a researcher has John rate Mary on friendliness using multiple measures or on multiple occasions, one can determine how much of the rating is due to the John-Mary relationship and how much is random across measure or time.
Conducting a study to use the social relations model requires a particular methodological approach. Specifically, multiple perceivers must rate multiple targets. So John would need to report not only Mary’s friendliness but Bill’s and Sally’s too. Having multiple raters and targets can be accomplished in several ways, but is most easily done using a round-robin design in which each person rates, and is rated by, each other person in the group. A block design, in which people rate some members of the group but not others, is a common alternative to the round-robin study. Block designs are often used when a researcher is specifically interested in intergroup perceptions. For example, do men perceive women in the same way that women perceive men? Specialized software programs (SOREMO and BLOCKO) are used to analyze data from the different social relations model designs.
Using the social relations model, a researcher can investigate a variety of topics, including what Kenny calls the nine basic questions of interpersonal perception. Three of these questions can be answered by evaluating variability in the perceiver, target, and relationship effects. If perceptions are largely a function of the perceiver effect, then one has evidence for assimilation. That is, perceivers tend to see all of their targets in a similar way. For example, John might see Mary, Bill, and Sally as friendly, regardless of actual differences among them. Conversely, when perceptions are driven by the target effect, a researcher has evidence for consensus. In this case, perceivers tend to agree on which targets are high or low on a trait. John, Bill, and Sally might concur that Mary is somewhat unfriendly. Finally, strong relationship effects make the case for uniqueness: A given person’s perception of another person is idiosyncratic.
The remaining basic questions are addressed by evaluating the degree to which the social relations model effects are related to each other, self-perceptions, or metaperceptions. For example, one might wonder whether people see others as others see them, called reciprocity. Evidence for reciprocity would be documented by an overlap between perceiver effects (how people see others) and target effects (how people are seen by others). Another intriguing possibility is assessing meta-accuracy which is the degree to which people know how others see them. This would be evaluated by related perceiver effects in metaperceptions (how people think they are seen by others) with target effects in perceptions (how people are actually seen by others). The last four questions include assessing whether people can accurately perceive another person’s traits (target accuracy), whether people assume others see them as they see others (assumed reciprocity), whether people see others as they see themselves (assumed similarity), and whether people see themselves as others see them (self-other agreement).
- DePaulo, B. M., Kenny, D. A., Hoover, C. W., Webb, W., & Oliver, P. V. (1987). Accuracy of person perception: Do people know what kinds of impressions they convey? Journal of Personality and Social Psychology, 52, 303-315.
- Kenny, D. A. (1994). Interpersonal perception: A social relations analysis. New York: Guilford Press.
- Kenny, D. A., Albright, L., Malloy, T. E., & Kashy, D. A. (1994). Consensus in interpersonal perception: acquaintance and the Big Five. Psychological Bulletin, 116, 245-258.