A dyad refers to two individuals treated as a unit, or a pair. Dyadic relationships are characterized by some kind of interdependence, such as romantic relationship, friendship, collaboration, or social interaction.
Dyads in Social Relations
In modern social relations research, scholars identify “dyad,” along with “individual” and “group,” as an essential unit of analysis (Collins 2002; Rubin et al. 2007). A dyad consists of two interdependent individuals. Both members stimulate and react to one another’s behaviors. Dyadic relationships can be long-term, in which both parties shared a sense of membership for an extended period of time. Examples include romantic relationships, friendships, parent and child, and supervisor and supervisee. Dyads can also be short-lived and context-dependent. For example, two strangers that engage in a conversation on a flight can also be considered as a dyad. Some dyads are not even naturally formed dyads, but artificially constructed for some purposes. For example, Lorenzo et al. (2010) recruited and paired unacquainted undergraduate students as dyads to study how physical attractiveness affects first impressions of personality.
The majority of psychological research on dyads focuses on close relationships, and romantic couple has received attention above all other forms of dyads. The emphasis of research has been on identifying correlates of relationship initiation, development, and outcome, examining how characteristics at both the individual and dyadic level affect dyadic interactions. Examples of individual-level measures are each couple member’s personality, attachment style, general well-being, health, aggression level, and relationship satisfaction. Dyadic measures include similarity in personality, relationship length, discrepancy in relationship satisfaction, and others. Combining these variables allow researchers to formulate a series of questions. Do husbands and wives have similar personalities? Does his personality influence her marital satisfaction? Does similarity in personality between couple members predict their individual marital satisfaction? Do couples become more similar in personality over time? Besides romantic couples, other forms of dyads including friendship dyads (e.g., Oswald et al. 2004) and parent-child relations (Kuczynski 2003) have been studied extensively as well.
The empirical study of dyads has been methodologically and statistically challenging. Researchers have traditionally used individual as the unit of analysis for dyads and ignored statistical dependence between dyad members. This is problematic because a design that examines effect at one level may involve phenomena at another level. Therefore, dyadic data require analysis techniques that are different from those of individuals to properly assess the effects of individual and dyadic variables concurrently. Analysis techniques such as multilevel modeling, intraclass correlation, structural equation modeling, and actor-partner model have been employed to account for the interpersonal dependence (Kenny et al. 2006).
A related concern is whether dyads in a sample are independent from one another. Sometimes dyads are formed from a friendship network where an individual can be connected to multiple friends. Researchers can either eliminate dyads so that no member of any dyad was also a member of another dyad or use adequate multilevel modeling to account for the interdependence.
Another consideration is whether dyad members are distinguishable (Griffin and Gonzalez 1995). Members of distinguishable dyads can be aligned by a theoretically meaningful variable such as gender in heterosexual dating couples. In contrast, the assignment of dyad members as Person A or Person B is arbitrary for dyads with indistinguishable members such as same-sex friends or homosexual dating couples. Techniques such as double-entry intraclass correlation are often used in analysis.
- Collins, W. A. (2002). Historical perspectives on contemporary research in social development. In P. K. Smith & C. H. Hart (Eds.), Blackwell handbooks of developmental psychology. Blackwell handbook of childhood social development, (pp. 3–23). Malden: Blackwell Publishing.Google Scholar
- Griffin, D., & Gonzalez, R. (1995). Correlational analysis of dyad-level data in the exchangeable case. Psychological Bulletin, 118(3), 430.Google Scholar
- Kenny, D., Kashy, D., & Cook, W. (2006). Dyadic data analysis. New York: Guilford Press.Google Scholar
- Kuczynski, L. (2003). Handbook of dynamics in parent-child relations. Thousand Oaks, CA: Sage.Google Scholar
- Rubin, K. H., Bukowski, W. M., & Parker, J. G. (2007). Peer interactions, relationships, and groups. In N. Eisenberg (Ed.), Handbook of child psychology. Hoboken, NJ: Wiley.Google Scholar