Couple Observational Coding Systems

Patricia K. Kerig, Donald H. Baucom, Patricia K. Kerig, Donald H. Baucom

• Publisher: Routledge
• Year: 2004
• Print ISBN: 9780805843576
• eBook ISBN: 9781135629809
• Topic: Psychology
• Subtopic: Developmental Psychology
• Index: Psychology

A companion volume to Family Observational Coding Systems, this book moves from the triad to the dyad and provides a showcase for significant developments in the coding of intimate couple interactions. The hope is that this book will contribute to the broadening and deepening of the field by disseminating information both about the coding systems that have been developed, as well as the conceptual and methodological issues involved in couple observational research.


The first three chapters present overviews of conceptual and methodological issues in the study of couple processes. The remaining chapters describe contributions to the field by 16 teams of researchers. Each chapter provides information about the conceptual underpinnings and structure of the coding system developed by the authors and evidence for its psychometric properties.

Couple Observational Coding Systems will be of interest to researchers studying couple interactions as well as clinicians who work with couples.

I

Conceptual and Methodological Issues

1

Coding Couples’ Interactions: Introduction and Overview

Donald H. Baucom and Patricia K. Kerig

University of North Carolina—Chapel Hill

People have been observing couples interact since the dawn of time: “Did you see the way he ignored her at the party?”… “Wasn’t that elderly couple sweet? I hope we’re that affectionate with each other when we get older.”… “I wonder if their marriage is in trouble. No matter what one says, the other disagrees.”… “They’re going to have a hard time coping with the medical problems, but if anyone can do it, they can. They are so supportive of each other.” Over the past several decades, couple researchers have joined the brigade of “people watchers,” focusing on the interactions that occur in these most important intimate relationships.

This emphasis on couple interactions is based not only on our inherent interest in watching people. Instead, the focus on dyadic interactions derives from a broader behavioral commitment to the direct observation of human behavior. If we are going to understand intimate relationships, then we need to observe directly how partners behave toward each other. And as scientists, we must derive systematic ways to rate, describe, and categorize these ongoing flows of complex interaction. Direct observation is not necessarily a superior source of data about couples; the relative utility of various sources of data must be established empirically. How couples respond to questionnaires or their physiological reactions during interactions can be valuable sources of information about relationship functioning. Couple interaction data is one potentially valuable source of couple information, and we believe that the findings described in this volume strongly support what we all might assume: How individuals interact with their partners tells us a great deal about them as individuals and as a unit.

The challenge for couple researchers committed to developing an interaction coding system is to take an ongoing stream of dyadic behaviors and devise a way to parse it into meaningful units that can be reliably coded, yet capture important aspects of this very rich interaction. We have been fortunate to obtain contributions from the majority of couple interaction researchers who have helped to shape the field since the 1970s. At present, there is no single source for researchers and clinicians to read to gain an understanding of the different ways to evaluate couples as they interact; hopefully this volume will help to fill that void.

ORGANIZATION AND CONTENT OF THE CHAPTERS

Before focusing on specific coding systems, it is important to understand the state of the field of couple interaction research: the issues it confronts, the successes and limitations of the field to date, methodological issues that must be understood in evaluating couple coding systems, and a variety of strategies that can be employed to analyze the data that are derived from the coding systems. Weiss and Heyman provide the reader with a frank and thoughtful perspective on the current state of the field. Although describing themselves tongue-in-cheek as the village idiots of the couple coding village, we believe the reader will recognize the wise sages who challenge us not to rest on our laurels and to integrate our impressive technologies with theories of relationship functioning that will guide future research. Anyone who has delved into coding couples’ interactions likely has experienced the following: “This stuff is complicated. I have this huge amount of detailed data on couples, but I’m not quite certain what to do with it.” There is a great deal of complicated methodological and statistical information to understand to make good use of interactional data. Floyd and Rogers do an excellent job of explaining in understandable language the variety of methodological issues to consider in creating, evaluating, and employing a couple coding system. Whereas the vast majority of this volume is about the coding systems themselves, once “raw data” from interactions are boiled down into codes or ratings, an investigator must know how to analyze the data. There are a variety of strategies for such purposes, ranging from statements about the frequencies with which couple phenomena occur during the interaction to complex analyses that take sequences of behaviors and contingencies among behaviors into account. Sayers and McGrath provide a clear and thoughtful discussion of these data analytic strategies, along with essential references for more detailed discussions of technical, statistical issues for couple interaction researchers.

The second section of this volume is devoted to the coding systems themselves, with a separate chapter describing each of the 16 measures. To assist the reader in comparing various coding systems, each chapter employs the same subdivisions. First, the authors introduce the coding system with a brief summary description and then discuss the theoretical foundations guiding the research. Next, the authors describe the development of the coding system and the strategic decisions that they made along the way. Following this, the authors present details of the coding system, including the task and setting for which the coding system is appropriate, the dimensions and categories that are coded, and the coding process, including how coders are trained and what training materials are available. The authors then provide information about the psychometric properties of the measure, including reliability and validity as well as limitations to, or evidence of, the generalizability of the coding system across different tasks and samples. We also asked the authors to discuss ways in which their coding systems might be used clinically. Finally, the authors give an overview of the range of studies that have been conducted using the coding system.

Dimensions of Coding Systems

Deciding on the sequence of chapters for the coding systems was complicated, reflecting the multifaceted nature of coding systems themselves. As Floyd, Baucom, Godfrey, and Palmer (1998) pointed out in their review of issues to consider in creating an observational coding system, any couple observational coding system involves a large number of decisions by the investigator creating the coding system. These decisions shape the coding system and what information can be obtained from it. For example, the constructor must decide what aspects of couple interaction are important to him or her (e.g., specific behaviors such as interruptions, patterns of interaction such as mutually avoiding addressing areas of concern, supporting each other during difficult personal times, etc.). Second, the coding system must be applied to some interaction, and the constructor, researcher, or clinician must once decide on the type of interaction or instructions for interaction, if instructions are provided to the couple. Thus, couples might be asked merely to talk to each other, to try to resolve some relationship problem, to support each other as individuals, to share feelings openly with each other, or interact with each other as naturally as possible in a laboratory apartment over a number of hours. After deciding on the aspects of a couple’s interaction to code and the instructions or “task” presented to the couple, the constructor must decide whether to create a coding system that looks at the interaction in an extremely detailed, microanalytic manner (e.g., coding every few seconds) or in a more global, macroanalytic manner (e.g., rate the entire interaction on some dimension). In addition, someone has to rate or evaluate the couple’s interaction. In most of the coding systems described in this volume, outside trained raters are employed—an outsider’s perspective; however, at times the partners themselves are asked to rate their behaviors and interactions—an insider’s perspective. Clearly, insiders’ versus outsiders’ perspectives provide potentially different information about the interaction. As a result of the many decisions made during the development of a coding system, any coding system exists in multidimensional space, just as couples’ interactions themselves are multidimensional. As a result, grouping the wide variety of coding systems described in this volume into broad categories is somewhat arbitrary because two coding systems might be quite similar in that they are both microanalytic, yet they might be very different in the content of what they are assessing in a detailed manner. In this volume, we have grouped the coding systems into broad categories based on the specific domains, or dimensions that they assess.

Problem Solving and Communication

We begin with a set of coding systems that were developed primarily to assess a broad range of couple behaviors that occur while partners are problem solving or discussing conflictual issues. Historically, this is where couple observational coding systems began within a social learning perspective. In the 1960s and 1970s, couples’ communication was almost synonymous with problem solving or conflict resolution. Behavioral couple therapy (then called behavioral marital therapy) and couple observational research developed “interactively,” with basic observational research shaping treatment, and treatment findings setting the way for additional basic research on couple interactions. At present, there are over 20 controlled treatment outcome investigations of behavioral (or cognitive—behavioral) couple therapy (Baucom, Hahlweg, & Kuschel, in press). Almost all of these treatment studies included communication training, which fundamentally meant strategies for resolving problems or conflict. Three major microanalytic coding systems evolved to assess couples’ abilities to problem solve: the Marital Interaction Coding System (MICS; Hops, Wills, Patterson, & Weiss, 1972), the Couple Interaction Coding System (CISS; Gottman, 1979), and the Kategoriensystem für Partnerschaftliche Interaktion (KPI; Hahlweg, Reisner, et al., 1984). In this volume, Hahlweg provides a description of the KPI, along with an impressive set of validational studies which demonstrate that coding systems initially developed for a specific purpose often have much broader applicability.

These microanalytic coding systems have the virtue of providing detailed information about couples’ interactions and have resulted in many valuable findings. On some occasions, however, investigators do not need or want this level of detail, and the time and labor required for microanalytic coding is considerable. As a result, a new generation of less detailed coding systems that focus on partners’ communications during problem solving or conflict resolution conversations has been developed. Even among these less detailed coding systems, the level of specificity varies considerably. As an example of a coding system that retains an intermediate level of detail, Heyman describes the Rapid Marital Interaction Coding System (RMICS), the successor to the MICS. The RMICS provides codes for a number of positive, negative, and neutral behaviors, and raters provide a code each time that the speaker changes. The MICSEASE described by Griffin, Greene, and Decker—Haas also was inspired by the MICS and includes the opportunity for the partners to code their own affect experienced during the interaction employing a video recall procedure. Kline and colleagues describe a more macroanalytic coding system, the Interaction Dimensions Coding System (IDCS). The IDCS employed basic research findings from investigations using microanalytic coding systems such as the MICS and CISS to provide a global rating system in which coders provide ratings on a number of dimensions (e.g., withdrawal) after viewing the entire interaction. Whereas the IDCS moves toward an increasingly macroanalytic approach by providing overall ratings based on the entire interaction, Floyd took a different macroanalytic approach in the development of the Communication Skills Test (CST). Noting that in many investigations, detailed behavioral codes have been grouped into broader positive and negative ratings, Floyd created a system in which coders rate each person’s talk turn from very positive to very negative, rather than coding specific categories of positive and negative communication.

The aforementioned coding systems were developed to assess a broad range of types of communication during couples’ interactions. In recent years, more specific aspects of couples’ interaction have been investigated and corresponding coding systems have been developed. For example, Sevier, Simpson, and Christensen describe the Couples Interaction Rating System (CIRS), which focuses on one specific area of communication that has received a great deal of theoretical and empirical attention: demand—withdraw patterns. This pattern has been studied extensively by Christensen and his colleagues, focusing on the tendency for one partner to criticize and demand change while the other withdraws in a variety of ways. Several investigations have demonstrated that this pattern is related to lower levels of relationship satisfaction. In turn, Malik and Lindahl present the System for Coding Interactions in Dyads (SCID), which was designed to focus on maladaptive dynamics of power and control within couple relationships, including such behaviors as verbal aggression, coerciveness, and control. In addition, it assesses aspects of communication that may be related to domestic violence and other indicators of relationship power dynamics such as negative escalation and conflict management style.

Affect and Intimacy

One of the best known couple coding systems is the Specific Affect Coding System (SPAFF), which assesses emotion at a nonverbal level. The SPAFF was designed to teach coders about cues that reflect specific affects, while using coders’ judgments as socially competent cultural judges to code specific affects that are recognizable from a cultural instead of a physical features perspective. This differentiation among different affective states is important, given that Gottman and his colleagues have demonstrated that various emotions have differential predictability regarding the long-term adjustment of marriages. In this volume, Shapiro, Gottman, and Driver describe the most recent version of the SPAFF.

Recognizing that negative interaction is unavoidable, Tabares, Driver, and Gottman describe the Repair Attempts Observational Coding System, which assesses the de-escalation of negative affect during marital conflict. Repair attempts are behaviors that are aimed at preventing or reducing negativity during conflict. Tabares et al. describe 17 different repair attempt codes that they have gleaned from intensive study of couples’ interactions. In addition, they describe 11 responses to these repair attempts from the other individual. Not only is this a valuable coding system, but it provides great insight for both researchers and clinicians about how partners attempt to halt and alt...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Contributors
• Preface
• Part I Conceptual and Methodological Issues
• Part II Problem-Solving and Communication
• Part III Affect and Intimacy
• Part IV Information Processing
• Part V Social Support
• References
• Author Index
• Subject Index