David A. Kenny

University of Connecticut | UConn · Department of Psychology

Two rather surprising anomalies relating to statistical power occur in testing mediation. First, in a model with no direct effect for which the total effect and indirect effect are identical, the power for the test of the total effect can be dramatically smaller than the power for the test of the indirect effect. Second, when there is a direct effe...

Studying dyads, very often there is a theoretical construct that has an effect on both members, such as relationship harmony or shared environment. To model such influences, the common fate model (CFM) is often the most appropriate approach. In this article, we address conceptual and statistical issues in the use of the standard CFM and present a s...

We present a new model for the general study of how the truth and biases affect human judgment. In the truth and bias model, judgments about the world are pulled by 2 primary forces, the truth force and the bias force, and these 2 forces are interrelated. The truth and bias model differentiates force and value, where the force is the strength of th...

A general theoretical model of interpersonal perception called PERSON (personality, error, residual, stereotype, opinion, and norm) is developed. This model reparameterizes a weighted-average model (WAM; Kenny, 1991; Kenny, Albright, Malloy, & Kashy, 1994) into six components. Two of those components refer to categorical information and 4 to behavi...

Many topics in organizational research involve examining the interpersonal perceptions and behaviors of group members. The resulting data can be analyzed using the social relations model (SRM). This model enables researchers to address several important questions regarding relational phenomena. In the model, variance can be partitioned into group,...

Face individuation involves sensitivity to physical characteristics that provide information about identity. We examined whether Black and White American faces differ in terms of individuating information, and whether Black and White perceivers differentially weight information when judging same-race and cross-race faces. Study 1 analyzed 20 struct...

The Social Relations Model (SRM), which has been an important tool for personality researchers, presumes the variabilities in the SRM components, perceiver, target, and relationship effects, are consistent across perceivers and targets. We introduce the extended SRM (eSRM) to examine individual differences in the variances of each component of the...

Methods to increase Campbell's (1957) internal and external validity as well as Cook and Campbell's (1979) construct and conclusion validity are reviewed. For internal validity or valid causal inference, designs and methods to draw causal conclusions from nonrandomized studies are considered. Greater collaboration between the causal inference and s...

Repeated investigations of the same phenomenon typically yield effect sizes that vary more than one would expect from sampling error alone. Such variation is even found in exact replication studies, suggesting that it is not only because of identifiable moderators but also to subtler random variation across studies. Such heterogeneity of effect siz...

The actor–partner interdependence model (APIM) has been widely successful and has become the default method for analyses of dyadic data in which both members of the pair have the same measurements. This article reviews the history of the development of the APIM, focusing on the original presentation of the model, a 1979 article by Kraemer and Jackl...

Background: The association between the couple relationship and the mothers' and fathers' psychological adjustment to the transition to parenthood has been examined in the literature. However, the direction of effects between these variables has not been extensively explored. This study aimed to assess the direction of effects between mothers' and...

Aim To examine whether mode of conception and gender are associated with parents’ psychological adjustment across the transition to twin parenthood. Background There is limited knowledge on the psychological adjustment of couples to twin parenthood during pregnancy and early postpartum , especially for fathers. The available research suggests that...

The actor-partner interdependence model (APIM) is widely used for analyzing dyadic data. Although dyadic research has become immensely popular, its statistical complexity might be a barrier. To remedy this, a free user-friendly web application, called APIM_SEM, has been developed. This app automatically performs the statistical analyses (i.e., stru...

In response to recommendations to redefine statistical significance to P ≤ 0.005, we propose that researchers should transparently report and justify all choices they make when designing a study, including the alpha level.

In response to recommendations to redefine statistical significance to P ≤ 0.005, we propose that researchers should transparently report and justify all choices they make when designing a study, including the alpha level.

In response to recommendations to redefine statistical significance to p ≤ .005, we propose that researchers should transparently report and justify all choices they make when designing a study, including the alpha level.

This volume brings together academic scholars and judges together to consider themes flowing from the often complex relationships between ‘law’ and ‘politics’, ‘adjudication’ and ‘policy-making’, the ‘judicial’ and ‘political’ branches of government. Part I addresses questions concerning the nature and extent of judicial power from a largely theore...

Multilevel modeling (MLM) and structural equation modeling (SEM) are the dominant methods for the analysis of dyadic data. Both methods are extensively reviewed for the widely used actor–partner interdependence model and the dyadic growth curve model, as well as other less frequently adopted models, including the common fate model and the mutual in...

Traditional methods of analyzing data from psychological experiments are based on the assumption that there is a single random factor (normally participants) to which generalization is sought. However, many studies involve at least two random factors (e.g., participants and the targets to which they respond, such as words, pictures, or individuals)...

Mediation analysis requires a number of strong assumptions be met in order to make valid causal inferences. Failing to account for violations of these assumptions, such as not modeling measurement error or omitting a common cause of the effects in the model, can bias the parameter estimates of the mediated effect. When the independent variable is p...

Group researchers often observe dyadic behaviors or ask group members to make judgments of one another. The resulting round-robin structures can be analyzed using the Social Relations Model (SRM). This model allows researchers to address the extent to which group members share the same perceptions of a common target (e.g., how much each person in t...

Members enter groups with different characteristics, for example, gender and ethnicity, and the Group Actor–Partner Interdependence Model (GAPIM) systematically tests several different effects of group composition for a given characteristic. By finding submodels of these effects, the GAPIM allows for empirically testing many theoretically meaningfu...

Potential moderators of effects in the actor–partner interdependence model (APIM) include variables that vary within dyads, between dyads, or both between and within dyads (i.e., mixed moderators). Another factor in the moderation of the APIM is whether dyads are indistinguishable (e.g., same-sex friendship pairs) or distinguishable (e.g., heterose...

When one gathers dyadic data, one is very often faced with the burdensome task of restructuring the data. For instance, the use of multiple regression analysis or structural equation modeling (SEM) requires one type of data structure, whereas multilevel modeling or multilevel SEM usually requires a different data structure. However, data are often...

A mediator explains, at least in part, a causal relationship between two variables. The total relationship between the two variables is due to a indirect effect, that part explained by the mediator, and due to an direct effect, that part not explained be the mediator. Strong assumptions must be made to estimate and test mediation. The inclusion of...

Researchers designing experiments in which a sample of participants responds to a sample of stimuli are faced with difficult questions about optimal study design. The conventional procedures of statistical power analysis fail to provide appropriate answers to these questions because they are based on statistical models in which stimuli are not assu...

The single mediator model is commonly used by prevention researchers and other behavioral and social scientists to investigate how prevention interventions change outcome variables such as risky behaviors. The single mediator model seems quite simple at first glance since it only contains three variables, but there are many assumptions that must be...

The study of intergroup perceptions is a key defining topic in the study of groups. In this paper, we describe how the Social Relations Model can be modified to become the Social Relations Intergroup Model. With this model, the effects of in-group favoritism, out-group homogeneity, and reciprocity can be measured at the group, individual, and dyad...

This paper has been revised. Please write the first author for a copy.

In a direct replication, the typical goal is to reproduce a prior experimental result with a new but comparable sample of participants in a high-powered replication study. Often in psychology, the research to be replicated involves a sample of participants responding to a sample of stimuli. In replicating such studies, we argue that the same criter...

Given that the root mean square error of approximation (RMSEA) is currently one of the most popular measures of goodness-of-model fit within structural equation modeling (SEM), it is important to know how well the RMSEA performs in models with small degrees of freedom (df). Unfortunately, most previous work on the RMSEA and its confidence interval...

Background: Few studies focus on the different dyadic relations among family members to study physical activity (PA) levels. Aim: The aim was to investigate predictors and sources of variance of PA levels in nuclear families using multi-level modelling. Subjects and methods: The sample consisted of 2661 Portuguese four-member nuclear families...

Recent theorizing has begun to suggest that spirituality may be transmitted through the observation and imitation of spiritual exemplars. However, there has been little systematic research on how and with what criteria individuals identify such models. The current study evaluates how religious constructs are judged among members of faith communitie...

Dyadic designs have been used in health research to investigate intra- and inter-personal mechanisms of health and well-being in various types of dyads, including parent–child dyads, siblings, friends, and romantic partners. Although a growing number of researchers are designing studies that capture the interdependent complexities of relationships,...

Abstract Project EXPLORE - a large-scale, behavioral intervention tested among men who have sex with men (MSM) at-risk for HIV infection - was generally deemed as ineffective in reducing HIV incidence. Using novel and more precise data analytic techniques we reanalyzed Project EXPLORE by including both direct and indirect paths of intervention effe...

The present research used a latent variable trait-state model to evaluate the longitudinal consistency of self-esteem during the transition from adolescence to adulthood. Analyses were based on ten administrations of the Rosenberg Self-Esteem scale (Rosenberg, 1965) spanning the ages of approximately 13 to 32 for a sample of 451 participants. Resul...

The rate of casual sexual encounters is increasing on college campuses. To decrease sexual risk behavior, information used to judge sexual risk in others needs to be identified. Women rated male targets on willingness to have unprotected sex with the target and likelihood that the target has a sexually transmitted infection. Physical attractiveness...

The social relations model (SRM) is a conceptual, methodological, and analytical approach that is widely used to examine dyadic behaviors and interpersonal perception within groups. This article introduces a general and flexible approach to estimating the parameters of the SRM that is based on Bayesian methods using Markov chain Monte Carlo techniq...

We extend the actor–partner interdependence model (APIM), a model originally proposed for the analysis of dyadic data, to the study of groups. We call this extended model the group actor–partner interdependence model or GAPIM. For individual outcomes (e.g., satisfaction with the group), we propose a group composition model with four effects; for gr...

Throughout social and cognitive psychology, participants are routinely asked to respond in some way to experimental stimuli that are thought to represent categories of theoretical interest. For instance, in measures of implicit attitudes, participants are primed with pictures of specific African American and White stimulus persons sampled in some w...

Heterosexual couples (N = 57) discussed features about each other they wanted to change. During a review of their recorded discussions, for each 30 s of interaction, perceivers provided judgments of their partner's regard, and partners reported their actual regard for the perceiver. The authors simultaneously assessed the extent to which perceivers...

The assessment of mediation in dyadic data is an important issue if researchers are to test process models. Using an extended version of the actor–partner interdependence model the estimation and testing of mediation is complex, especially when dyad members are distinguishable (e.g., heterosexual couples). We show how the complexity of the model ca...

In this brief commentary, the author discusses the PERSOC framework and the five empirical papers in this issue. Copyright © 2011 John Wiley & Sons, Ltd.

DataToText is a project developed where the user communicates the relevant information for an analysis and DataToText computer routine produces text output that describes in words, tables, and figures the results from the analyses. Two extended examples are given, one an example of a moderator analysis and the other an example of a dyadic data anal...

The social relations model (SRM) is an intriguing tool both to conceptualize and to analyze dyadic processes. We begin with explaining why interpersonal phenomena in everyday life are more complex than often considered. We then show how the SRM accounts for these complexities by decomposing interpersonal perceptions and behaviors into three indepen...

A classic question in social and organizational psychology is whether low-status persons are more accurate in the perception of their high-status partners than the latter are in their perception of their subordinates. In a series of studies, Snodgrass (1985, 1992) tested this idea. She found that subordinates were more accurate at judging how their...

The social relations model (SRM) is a useful tool for measuring relationship effects, defined as the unique perceptions or behaviors of 2 people. The sources of variance in SRM studies are persons (actors and partners), groups, and items; the relationship effect is defined as the actor–partner interaction. By removing variance because of persons an...

Family researchers have used the actor-partner interdependence model (APIM) to study romantic couples, parent-child dyads, and siblings. We discuss a new method to detect, measure, and test different theoretically important patterns in the APIM: equal actor and partner effect (couple pattern); same size, but different signs of actor and partner eff...

Doctor-patient communication is an interpersonal process and essential to relationship-centered care. However, in many studies, doctors and patients are studied as if living in separate worlds. This study assessed whether: 1) doctors' perception of their communication skills is congruent with their patients' perception; and 2) patients of a specifi...

The authors examined the consistency of person perception in two domains: agreement (i.e., do two raters of the same person agree?) and similarity (i.e., does a perceiver view two persons as similar to one another?). In each domain, they compared self-judgments with judgments not involving the self (i.e., self-other agreement vs. consensus, in the...

An implicit assumption concerning many latent variables studied within health psychology, (e.g., personality, emotions, health behaviours, illness repre-sentations, symptoms), is that they are continuous/ dimensional constructs. However, it is an empirical question whether or not a latent construct is truly di-mensional (i.e., distributed as a cont...

Recent research focuses on developing robots that are meant to be partners of humans instead of pure machines. This makes enhanced communication necessary. Especially in scenarios embedding physical interaction between the two partners dominance is an urgent matter. To overcome one-sided dominance as in passive following or trajectory replay in fav...

Psychotherapy research can be thought to involve multiple levels of analysis: client, therapist, and group. The authors argue that such research would benefit from the application of multilevel modeling. They illustrate the use of such a procedure for research studies of individual psychotherapy, group psychotherapy, and rolling group psychotherapy...

The social relations model (SRM; Kenny, 1994) explicitly proposes that leadership simultaneously operates at three levels of analysis: group, dyad, and individual (perceiver and target). With this model, researchers can empirically determine the amount of variance at each level as well as those factors that explain variance at these different level...

Hall, Lord, and Foster (2009) have commented on whether variance partitioning in the social relations model would be the same in long-term groups and when groups have formal leaders. This reply follows their lead and speculates on how the variance partitioning would change. It considers the design and analysis issues in natural workgroups as well a...

Dyadic research is becoming more common in the social and behavioral sciences. The most common dyadic design is one in which two persons are measured on the same set of variables. Very often, the first analysis of dyadic data is to determine the extent to which the responses of the two persons are correlated-that is, whether there is nonindependenc...

Decisional conflict is defined as personal uncertainty about which course of action to take when choice among competing options involves risk, regret, or challenge to personal life values. It is influenced by inadequate knowledge, unclear values, inadequate support, and the perception that an ineffective decision has been made. Until recently, it h...

Selecting sex partners of the same HIV status or serosorting is a sexual risk reduction strategy used by many men who have sex with men. However, the effectiveness of serosorting for protection against HIV is potentially limited. We sought to examine how men perceive the protective benefits of factors related to serosorting including beliefs about...

Different theoretical approaches assume, either implicitly or explicitly, that leadership operates at one of three levels of analysis: group, dyad, or individual (perceiver and target). The social relations model [Kenny, D. A. (1994). Interpersonal perception: A social relations analysis New York: Guilford] allows for all of these levels to operate...

This commentary reviews the author's history with mediation. In the second part, the author attempts to explain why there is such interest in the topic. In the third part, the author comments on the five articles in this feature topic.

The study of gender differences is a pervasive topic in relationship science. However, there are several neglected issues in this area that require special care and attention. First, there is not just one gender effect but rather three gender effects: gender of the respondent, gender of the partner, and the gender of respondent by gender of the par...

There is considerable interest today in shared decision-making (SDM), defined as a decision-making process jointly shared by patients and their health care provider. However, the data show that SDM has not been broadly adopted yet. Consequently, the main goal of this proposal is to bring together the resources and the expertise needed to develop an...

Perceivers are both accurate and biased in their understanding of others. Past research has distinguished between three types of accuracy: generalized accuracy, a perceiver's accuracy about how a target interacts with others in general; perceiver accuracy, a perceiver's view of others corresponding with how the perceiver is treated by others in gen...

This article addresses conceptual and methodological levels of analysis issues in research on work group and organizational settings. Using organizational climate data, it provides a detailed example of the use of a recent data analysis model (Kenny & La Voie, 1985) which separates individual and group effects. The reanalysis of an earlier study of...

Structural equation modeling (SEM) can be adapted in a relatively straightforward fashion to analyze data from interchangeable dyads (i.e., dyads in which the 2 members cannot be differentiated). The authors describe a general strategy for SEM model estimation, comparison, and fit assessment that can be used with either dyad-level or pairwise (doub...

Family assessment instruments attempt to measure family functioning at a particular level of the family system: individual, dyad, or family as a whole. This article introduces the concept of level validity, that is, whether an assessment measures family functioning at the level that it was intended to measure. The authors argue that whenever higher...

Impression formation has been conceptualized as a process whereby perceivers form person models of others, describing what the person is like and why (Park, DeKay, & Kraus, 1994). If they exist, person models could help explain why there appears to be little consensus in personality judgments; perceivers forming different models of a person would a...

ABSTRACT Research on accuracy and consensus in interpersonal perception has become a major topic in the social sciences. Key methodological issues in this research are research design, statistical measures of agreement, unit of analysis, and data aggregation. I discuss the twelve articles devoted to these topics in this special issue in terms of mo...

Our research examined whether relationship factors account for the positive association between genetic relatedness and willingness to help. College students’ (31 male and 46 female) willingness to help family members was measured using hypothetical dilemmas involving life-or-death and everyday-favor situations. Relationship factors were measured u...

We examine the advantages and disadvantages of 2 types of analyses used in interpersonal perception studies: componential and noncomponential. Componential analysis of interpersonal perception data (Kenny, 1994) partitions a judgment into components and then estimates the variances of and the correlations between these components. A noncomponential...

Statistical models summarize by a set of equations the covariation between a set of variables. Models can either be causal or predictive. Models usually contain a deterministic component or fit and a random component or error. Assumptions are made about the distribution of the error component. Models can be either exploratory or confirmatory.

The minimal cross-lagged panel design is two variables measured at two times. Multiple regression analyses, in which the time two measures are regressed on the time one measures, have become the standard method of analyzing the design. Currently, methods to rule out third-variable causation are little used. Several analysts have argued that two wav...

Three moderators of agreement in person perception, behavioral consistency, observability and social desirability, were studied. The major hypothesis is that the moderators can be estimated using the standing of targets on traits; that is, that as targets vary on a given trait, they vary on how they are seen as on the moderators. Using Korean (N =...

Relationship researchers regularly gather data from both members of the dyad, and these two scores are likely to be correlated. This nonindependence of observations can bias p values in significance testing if person is the unit in the statistical analysis. A method for determining how much bias results from dyadic interdependence is presented. Cor...

This article discusses the conceptual meaning of partner effects, which occur when one person is affected by the behavior or characteristics of his or her partner. We show that partner effects can be used to validate the presence of a relationship and can elaborate the particular nature of that relationship. We discuss possible moderation of partne...

Multilevel models are proposed to study relational or dyadic data from multiple persons in families or other groups. The variable under study is assumed to refer to a dyadic relation between individuals in the groups. The proposed models are elaborations of the Social Relations Model. The different roles of father, mother, and child are emphasized...

The actor–partner interdependence model (APIM) is a model of dyadic relationships that integrates a conceptual view of interdependence with the appropriate statistical techniques for measuring and testing it. In this article we present the APIM as a general, longitudinal model for measuring bidirectional effects in interpersonal relationships. We a...

This article presents two methodological techniques that are particularly well suited to the study of family communication when data from multiple family members are gathered. Many continue to analyze family data using less than optimal data analytic approaches, thus limiting the ability of these results to reveal new insights into the family commu...

An experiment varying the racial (Black, White) and opinion composition in small-group discussions was conducted with college students (N = 357) at three universities to test for effects on the perceived novelty of group members' contributions to discussion and on participants' integrative complexity. Results showed that racial and opinion minoriti...

This study evaluated several statistical models for estimating treatment effects in a randomized, longitudinal experiment comparing assertive community treatment (ACT) versus brokered case management (BCM). In addition, mediator and moderator analyses were conducted. The ACT clients had improved outcomes in terms of housing and psychiatric symptoms...

A family assessment serves two primary purposes: the guidance of clinical interventions and the evaluation of clinical outcomes. To support these activities, a family assessment should provide a level of descriptive detail with respect to family functioning that is commensurate with a family systems perspective. Unfortunately, most widely used self...