• Marital Communication;
  • Spousal Caregiving;
  • Sequential Analyses


  1. Top of page
  2. Abstract

To read this article's abstract in both Spanish and Mandarin Chinese, please visit the article's full-text page on Wiley InterScience (

Dementia research has frequently documented high rates of caregiver depression and distress in spouses providing care for a partner suffering from dementia. However, the role of marital communication in understanding caregiver distress has not been examined sufficiently. Studies with healthy couples demonstrated an association between marital communication and the partners' psychological well-being, depressiveness, respectively (e.g., Heene, Buysee, & Van Oost, 2005). The current study investigates the relationship between caregiver depression and communication in 37 couples in which the wives care for their partners with dementia. Nonsequential and sequential analyses revealed significant correlations between caregiver depression and marital communication quality. Caregivers whose husbands used more positive communication reported less depression and distress. Additionally, caregiver depression was negatively correlated with rates of positive reciprocal communication indicating dependence between the couples' interaction patterns. This study is one of the first to illustrate the relevance of spousal communication in understanding caregiver distress and depression.

Frecuentemente, las investigaciones sobre demencia han documentado índices altos de depresión y angustia en esposas que cuidan a su pareja con demencia. Sin embargo, el papel que juega la comunicación matrimonial a la hora de comprender la angustia de las cuidadoras no se ha analizado suficientemente. Los estudios realizados con parejas sanas demostraron que existe una asociación entre la comunicación matrimonial, y el bienestar psicológico o el estado depresivo de los integrantes de la pareja respectivamente (p. ej.: Heene, Buysee, & Van Oost, 2005). En el presente estudio se investiga la relación entre la depresión de las cuidadoras y la comunicación en 37 parejas en las cuales las esposas cuidan a sus parejas con demencia. Los análisis no secuenciales y secuenciales revelaron correlaciones importantes entre la depresión de las cuidadoras y la calidad de la comunicación matrimonial. Las cuidadoras cuyos maridos usaron una comunicación más positiva manifestaron menos depresión y angustia. Además, la depresión de las cuidadoras estuvo correlacionada negativamente con índices de comunicación recíproca positiva, lo cual indica dependencia entre los patrones de interacción de las parejas. Este estudio es uno de los primeros en poner de manifiesto la relevancia de la comunicación conyugal a la hora de comprender la angustia y la depresión de las cuidadoras.

Palabras clave: comunicación matrimonial, cuidado conyugal, análisis secuenciales

  • image

Because of the alarming prevalence rates of dementing illnesses, research demonstrates growing interest in older couples affected with dementia of one partner. A multitude of studies have exhibited various negative consequences for the healthy, and often caregiving, partner of individuals suffering from dementia (e.g., Pinquart & Soerensen, 2003; Vitaliano, Zhang, & Scanlan, 2003). Research focusing on risk and resilience factors that influence psychological well-being of dementia caregiving spouses identified several aspects contributing to caregiver burden, such as neuropsychiatric symptoms (e.g., Perren, Schmid, & Wettstein, 2006), gender of the caregiving spouse indicating that female caregivers suffer from more depressive symptoms, stress, and somatic symptoms (e.g., Gilhooly, Sweeting, Whittick, & McKee, 1994; Thompson et al., 2004), and relationship quality (e.g., Knop, Bergman-Evans, & McCabe, 1998) (for details see Braun et al., 2009).

Previous research on healthy young, middle-aged, and older couples has clearly demonstrated the importance of marital communication for the partners' well-being and health (e.g., Hefner et al., 2006; Kiecolt-Glaser et al., 2005; Manne et al., 2004). Furthermore, several studies have revealed the influence of marital interaction on both relationship satisfaction (e.g., Schmitt, Kliegel, & Shapiro, 2007) and relationship quality (e.g., Stanley, Markman, & Whitton, 2002; Weiss & Heyman, 1997). First, there is evidence that negative marital communication is associated with concurrent relationship dissatisfaction (e.g., Bodenmann, Kaiser, Hahlweg, & Fehm-Wolfsdorf, 1998; Caughlin & Huston, 2002). Second, positive communication behavior has been demonstrated to predict future marital satisfaction (e.g., Gill, Christensen, & Fincham, 2005). Third, research studies identified a high communication quality as a resilience factor limiting the probability of divorce (for an overview see Gottman & Notarius, 2000). However, not only the amount of positive and negative communication influences partner well-being and satisfaction, but also the reciprocity of positive or negative communication. Nondistressed couples show more positive reciprocity indicating a higher likelihood of responding positively after a positive statement of the partner (Revenstorf, Vogel, Wegener, Hahlweg, & Schindler, 1980). Focusing on negative reciprocity, distressed couples use more reciprocal negative communication than nondistressed couples (Cordova, Jacobson, Gottman, Rushe, & Cox, 1993; Margolin & Wampold, 1981). Thus, compared with nondistressed relationships, individuals in distressed relationships show a greater likelihood of negative behavior by one partner being followed by a negative response by the other partner. Consequently, not only the analyses of nonsequential communication patterns without examining the way the one partner responds to the communication of the other partner (e.g., rates and frequencies of communication characteristics, comparison of communication patterns of the interacting partners), but also partners' reactions to each other, contain essential information concerning marital quality and relationship satisfaction. Hence, the evaluation of sequential communication dynamics is useful to gain an understanding about how dyadic and individual well-being is related to couple communication patterns.

In couples afflicted with dementia, due to growing impairments in the course of the illness, both dyadic communication (e.g., Medina & Weintraub, 2007) and relationship quality may be negatively impacted (DeVugt et al., 2003; Eloniemi-Sulkava et al., 2002; Narayan, Lewis, Tornatore, Hepburn, & Corcoran-Perry, 2001). Therefore, it is plausible to assume that relations between mental health and accordingly well-being, marital satisfaction, and communication in dementia caregiving dyads are concordant with the associations found in healthy couples. However, only a few studies have focused on communication of couples affected by dementia of one partner. In one of these studies, interviews with spousal caregivers were conducted to describe the dyadic interaction of couples with one spouse suffering from dementia (Polk, 2005). Results showed that caregivers are often confronted with uncertainty about dyadic interactions and about how to attribute reactions of their partners (e.g., when the patient is unable to verbally and unequivocally express pleasure or displeasure). Gallagher-Thompson et al. (1997) assessed the communication style of dementia spousal caregiving wives and their partners. The authors found a substantial negative correlation between caregivers' positive communication and depression. Another study by Gallagher-Thompson, Dal Canto, Jacob, and Thompson (2001) compared spousal interaction of couples with the husband suffering from dementia with healthy couples. They reported relevant group differences: First, healthy spouses were more interactive and expressed more support to each other compared with carer–care recipient dyads. Second, husbands with dementia were highest in interactions that build rapport (e.g., smiling). Third, the results demonstrated that dementia affects marital communication patterns, although some aspects of interaction seemed to remain intact (e.g., use of humor of the husbands with dementia).

Unfortunately, several methodological issues limit the explanatory power of these studies. The majority of study designs focus on the caregiving partner and basically neglect the perspective of the individual with dementia, resulting in incomplete information about afflicted couples (Braun et al., 2009). From marital research, it is well known that directly assessing both partners of a dyad is necessary to gain valid information of not only their individual view, but also their relational dyadic perspectives. As an example, Snyder, Heyman, and Haynes (2005) outlined several recommendations for measuring couple distress and emphasized the relevance of assessing both self-reports of the two partners and observational methods of their dyadic interaction. Another methodological issue leads to gaps in empirical knowledge about spousal interaction in couples affected by dementia. Most studies on marital interaction (with healthy couples as well as with couples affected by dementia, spousal dementia caregiving, respectively) assess data via nonsequential analyses (e.g., overall communication patterns of wife vs. communication patterns of husband). Doubtlessly, these study designs reveal useful information, but they cannot assess the level of information gained by sequential analyses. As mentioned by Bakeman and Gottman (1986), sequential methods are necessary when the research question is about the way dyadic behavior works in an ongoing interaction. Hence, the authors recommended sequential techniques added to observational nonsequential methods to get information about how behavior is sequenced in time (Bakeman & Gottman, 1986). These analyses explain whether a certain behavior (such as positive communication of one partner) increases the probability of another subsequent behavior (such as positive communication of the other partner). Thereby, possible dependencies between the interacting partners can be identified (Wampold & Margolin, 1982).

In sum, due to the fact that knowledge about interaction of couples in which one partner suffers from dementia is sparse, the question of how communication and psychological well-being in caregiving and care-receiving spouses are related cannot yet be answered sufficiently. Moreover, to the best of our knowledge, past research provides no information about how afflicted couples' sequential communication patterns are related with their well-being and mental health.


  1. Top of page
  2. Abstract

This explorative study aims to examine the following four questions: How do couples with one spouse suffering from dementia communicate? Do the communication patterns of the caring partner differ significantly from the patient's communication? How is marital communication associated with psychological well-being of afflicted partners? Are communication patterns of the caregiving and the care-receiving spouse related to each other and to each other's well-being?


  1. Top of page
  2. Abstract

Sample and Design

Thirty-seven couples (N=37) with the husband suffering from dementia participated in the present study conducted by the Department of Psychology of the University of Zurich. Participants were recruited via five clinical institutions in Switzerland and Germany and advertisements in local newspapers. This study focuses on baseline data from a longitudinal research project about dyadic communication development in caregiver–care recipient dyads. Because of the difficult enrollment of eligible couples and the already mentioned gender differences in caregiver stress, we exclusively assessed caring females and male patients. In order to control for the potential confounding effects of gender, a bigger sample size would have been necessary. Couples who participated in the two follow-ups received 100 Swiss francs, 75 euros, respectively, as compensation. Exclusion criteria included alcohol dementia, symptoms that would hinder study participation (e.g., pain, hearing deficits), and suspicion of dementia or cognitive decline of the caring wife. All participants were Caucasians.

Every patient was professionally diagnosed with dementia, in most cases Alzheimer's disease (65%). The husbands' average Mini Mental State Examination (MMSE) score was 17.11 (SD=8.0), suggestive of dementia of moderate severity. On average, they were 72 years old (SD=7.3) and suffered from dementia for 37 months (SD=21.1). The care receivers reported a mean Center for Epidemiological Studies Depression Scale (CES-D) score of 10.22 (SD=8.4) (see Table 1 for patient characteristics). Couples were on average married for 41 years (SD=11.4). The wives were the primary caregivers. As shown in Table 1, on average, the caring wives were 67 years old (SD=7.7), married for 41 years (SD=11.4), and spent 72 hours per week (SD=62.6) on caregiving tasks. The mean CES-D score was 16.82 (SD=8.3) and more than the half of caregiving wives (n=20) exceeded the clinical cut-off value of 16.

Table 1.  Sample Characteristics for Caregivers and Care Receivers
  1. Note. CES-D=Center for Epidemiologic Studies Depression Scale (Radloff, 1977); NPI-Q=Brief Version of the Neuropsychiatric Inventory Questionnaire (Kaufer et al., 2000); MMSE=Mini Mental State Examination (Folstein, Folstein, & McHugh, 1975); RAS=Shortened Version of the Relationship Assessment Scale (original version Hendrick, 1988).

Age (years)67.357.7539237
Duration of marriage (years)40.9611.4126836
Weekly hours spent caregiving72.3862.6016826
NPI-Q distress9.107.402429
Care receiver
Age (years)72.417.3589234
Duration of illness (months)37.4121.118534
Duration of care (months)23.3326.0012027
NPI-Q severity7.445.702034


For each couple, we obtained quantitative data from questionnaires and qualitative data by observing the spousal communication. The instruments are described in the following sections.


The caring wives completed several questionnaires: First, to assess depressive symptoms as a measure of psychological well-being, mental health, respectively, we applied the widely used German version of the CES-D (Radloff, 1977; Hautzinger, 1988). This instrument consists of 20 four-point Likert-scale items regarding the frequency of occurrence of depressive symptoms in the past week. Higher CES-D total scores represent more depressive symptoms. Scores of 16 or higher are commonly taken as indicative of enhanced risk of clinical depression. Second, in order to evaluate relationship satisfaction, a shortened version of Hendrick's (1988) Relationship Assessment Scale was included. The original scale consists of seven items. In the present study, a modified four-item version was used. Third, the brief Neuropsychiatric Inventory (NPI-Q; Kaufer et al., 2000) was conducted with the caregiving wives to assess psychopathology of their husbands. In the NPI-Q, the caregiver ranks the severity of the patient's neuropsychiatric symptoms exhibited on a scale of 1–3, with 3 being the most severe, leading to a total severity score for the patient. Furthermore, the caregiver reports the level of subjective distress caused by each symptom or behavior on a scale of 1–5 (with 5 indicating the most severe distress level). Therefore, the NPI-Q yields a total distress score consisting of the cumulative distress ratings for each neuropsychiatric symptom. Higher total scores indicate greater frequency or severity of the husband's symptoms as well as more distress of the caregiving wife.

In order to measure characteristics of the partner with dementia, the MMSE (German version: Kessler, Markowitsch, & Denzler, 1990) was used as a brief screening instrument to determine the level of cognitive function. The MMSE includes 30 questions and assesses the severity of cognitive impairment. A maximum of 30 points is possible, whereas lower scores reveal higher cognitive impairment and greater dementia severity. Because of lack of sensitivity, the MMSE is not suitable for making a differential diagnosis of dementia (e.g., Grober, Hall, Lipton, & Teresi, 2008), but is useful to measure the presence of cognitive impairment of dementia patients at a given time and to screen for dementing illnesses (e.g., Maki et al., 2000). To assess depressive symptoms of the husbands with dementia as a measure of psychological well-being, we also included the CES-D.

Despite their cognitive impairment, with the assistance of the interviewer, the majority of participants were able to answer the questions of the scale. Furthermore, we assessed duration of care, duration of marriage, and duration of illness as control variables. The descriptive statistics of the described tools are listed in Table 1.

Observational Coding System

The spousal interaction was evaluated by the Rapid Marital Interaction Coding System (RMICS; Heyman & Vivian, 1993), an observational coding system adapted from the Marital Interaction Coding System-IV (Heyman, Weiss, & Eddy, 1995). The basic coding unit of the RMICS is the speaker turn. In order to deal with long monologues, a new unit starts every 30 seconds that a person continues to speak. The RMICS consists of 11 hierarchically ordered communication codes ranging from negative (e.g., hostility), positive (e.g., humor), to neutral communication categories (e.g., problem discussion). The codes are ordered hierarchically, based on both communication theory and substantial research that demonstrates that negative, followed by positive, followed by neutral codes are of decreasing importance in understanding marital conflict (see Weiss & Heyman, 1997). If the partner emits more than one code during a speaker turn, s/he receives the code highest on the hierarchy (for more details about the instrument see Heyman, 2007). The RMICS has been shown to have excellent reliability and validity in a number of couple interaction studies with healthy adults (Heyman, 2004).


The sessions with couples took place in clinical institutions in Switzerland and Germany. Both patient and caregiver were separately interviewed face to face by a psychologist or trained graduate student. The interviewers were present while the participants completed the self-reports in case assistance was needed or wanted. Furthermore, a 10-minute interaction was videotaped. As suggested by past studies, participants were told that their task is to plan a future event together (Gallagher-Thompson et al., 2001). During the dyadic communication situation, the couple was alone in a silent room to avoid possible influences (e.g., effect of experimenter bias).

Statistical Analyses

In order to analyze self-reported data, we used basic statistical methods (e.g., variable frequency analyses). Focusing on communication data, two main methodological procedures were used to evaluate individual and dyadic communication: (1) analyses of both frequencies and base rates, (2) exploratory sequential analyses. Control variables were also included to assure the validity of the results (i.e., duration of illness, duration of care, duration of marriage). In the following section, these procedures are described.

First, we counted the frequency of every communication code (separately for caregiving wives and care-receiving husbands) and the sum score of all codes (total communication codes of caregiver vs. care receiver). Frequency analyses enable the identification of communication differences within a dyad, that is, differences between caregiver and care receiver communication characteristics. Furthermore, we were interested whether an imbalance of total communication between caregiver and care receiver exists and whether this possible imbalance is associated with indicators of psychological well-being. In the first step, we computed the frequency difference between the total communication scores of the spouses. In the second step, the resulting variable was correlated with self-reported data. Additionally, in order to obtain the base rates, we divided the frequencies of the single codes by the total number of codes (separated by caregiver and care receiver communication codes). The advantage of base rates over frequencies is their independence of communication duration and of total number of codes, which differ between couples. Thus, comparability between the dyads is possible (Bakeman & Gottman, 1986).

Second, we determined sequential patterns of the couples' interaction by conducting sequential analyses. This method aims to detect dynamic aspects of the communication process. As mentioned earlier, we assessed positive (i.e., dependence of a given positive behavior of one partner and a positive reaction of the other partner) and negative reciprocity (i.e., dependence of a given negative behavior of one partner and a negative response of the other partner) of couple communication because they are known to be important sequential patterns in marital interaction (Bakeman, 1997). Because the main focus of the sequential communication analyses was on dyadic interaction of couples affected by dementia, communication codes indicating a floor switch between the two partners were exclusively considered. Therefore, sequential data have not been evaluated for interaction flows of one partner. For data preparation, RMICS frequency data were transformed into a 6 × 6 transitional matrix for each couple. Our main focus was on the direct partner's reaction promptly after the target behavior (i.e., lag 1). Rows of the matrix represented the positive, negative, or neutral communication stimulus by either caregiver or care receiver. Columns represented the partner's response to this stimulus (positive, negative, or neutral communication). Thus, the cells included the summed number of the stimulus–response combinations of the spouses. According to other authors using RMICS data (Manne et al., 2004), we conducted the sequential analyses with the transformed κ statistic (Wampold, 1989). This method is advantageous because of its independence of length of sequence, and hence allows comparisons between couples. However, no computation of κ is possible if neither of the particular behaviors appears.

Third, t tests, correlational analyses, and partial correlations were calculated to clarify if caregiver and care receiver communication differ and how communication and self-reported data are related.


  1. Top of page
  2. Abstract

General Communication Characteristics

The inter-rater reliability of the RMICS coders (two trained graduate students) was Cohen's k=.63. In line with past studies using the RMICS (e.g., Crowell et al., 2002), the 11 communication codes were aggregated into the three main communication groups: positive, negative, and neutral communication. This aggregation procedure is advantageous when confronted with differing and partially low rates of occurrences of specific codes. Table 2 includes means, standard deviations, minimum, and maximum values of the three aggregated RMICS codes of the couples' communication (readers interested in the detailed frequency data of all original 11 codes may request them from the authors). Six of 37 wives showed a zero frequency of positive codes and 24 used no negative communication. Four husbands showed no positive communication and 27 had a zero-occurrence of negative communication codes.

Table 2.  Descriptive Statistics of Communication Categories and Total Number of Codes as Well as Paired t Test and Effect Sizes for the Frequency Differences Between Caregiver and Care Receiver (N=37 Couples)
RMICS codeCaregiverCare receivertda
M (%)SDMinimumMaximumM (%)SDMinimumMaximum
  • Note. RMICS=Rapid Marital Interaction Coding System (Heyman & Vivian, 1993).

  • a

    Effect sizes of differences were computed with Cohen's d for matched pairs.

  • *


Positive codes3.8 (7.6)3.10113.9 (8.2)4.3016−0.28.05
Negative codes0.9 (1.8)1.7060.5 (1)1.0041.80.30
Neutral codes47.1 (90.6)12.4228146 (90.8)14.218831.70.28
Total codes51.8 (100)12.6308650.4 (100)13.919862.64*.43

Base Rate Analyses

In the following, RMICS codes' frequencies, differences between caregiver and care receiver codes, and correlations between base rate communication data and self-reported data are presented.

As shown in Table 2, neutral communication codes were overrepresented in the dyadic interactions with a mean base rate of almost 91% for both spouses. The caring wives showed higher communication frequencies in neutral and negative communication, but less positive communication than the husbands with dementia. Paired t tests revealed a small but significant difference in amount of total codes, t(36)=2.64, p=.01, d=.43 between the partners. Thus, wives averagely showed more communication than the husbands.

No substantial changes in base rates of communication codes and total amount of communication can be observed with decreasing cognitive function. As shown in Table 3, dementia severity was only strongly negatively related with caregiver's amount of negative communication (r=.50, p=.00). In other words, the higher the level of patient's cognitive function, the more negative communication of the caring wife was observed. This finding, however, needs to be considered with caution due to the small number of caring wives using negative communication in this study (n=13). In order to detect a possible effect of the imbalance in total communication between spouses, the difference value between total amount of communication between wife and husband was correlated with subjective indicators of psychological well-being (see Table 3), but revealed no significant correlations. A correlation of marginal significance, however, was obtained for dementia severity (r=−.32, p=.05), suggesting that the higher the severity level of dementia, the higher the gap in communication between partners.

Table 3.  Bivariate Correlations Between Communication Measures (RMICS Codes, Total Communication, Dyadic Total Communication Difference) and Participant Characteristics
 RMICS codes caregiverRMICS codes care receiverRMICS total codes differencea
  1. Note. Pos=positive codes; Neg=negative codes; Neutr=neutral codes; Total=total of RMICS communication codes; RMICS=Rapid Marital Interaction Coding System (Heyman & Vivian, 1993); CES-D=Center for Epidemiologic Studies Depression Scale (Radloff, 1977); NPI-Q=Brief Version of the Neuropsychiatric Inventory Questionnaire (Kaufer et al., 2000); RAS=shortened version of the Relationship Assessment Scale (original version Hendrick, 1988); MMSE=Mini Mental State Examination (Folstein, Folstein, & McHugh, 1975); r=Pearson's correlation coefficient; rs=Spearman's rank-correlation coefficient. a Caregiver total codes minus care receiver total codes. *p<.05 , **p<.01.

NPI-Q distress−.25−.12.32−.15−.34*.13.27−.10−.19
Care receiver
NPI-Q severity−.22−.17.29−.29−.34*.11.25−.23−.17

Next, we were interested in possible relations between specific communication characteristics and mental health. The question was whether the rates of the communication codes and the total number of codes could account for variance in self-reported psychological well-being. Normal distribution fitting was checked by the Kolmogorov–Smirnov test. Correlations were estimated by Pearson's or Spearman's rank correlation coefficients. As seen in Table 3, caregiver depression and distress are significantly negatively associated with care receiver's positive communication. Furthermore, there was a substantial negative correlation between caregiver positive communication and caregiver depression, indicating that wives suffering from more depressive symptoms use less positive communication. In order to control for a possible effect of relationship satisfaction that correlates highly with caregiver depression (r=−.45, p=.01), we conducted partial correlations (data not shown). Results revealed, however, no significant relationship between satisfaction and positive communication. Actually, a suppressor effect of relationship satisfaction could be found for the association between depression and positive codes of caregiving wife (rpartial=−.36, p=.03). Moreover, there was no significant correlation between care receiver's depression and communication (data not shown are available upon request from the authors).

Sequential Analyses

Because of the limited number of dyads in which both spouses used negative codes (n=6, 16% of participants), sequential analyses of communication reciprocity were restricted to positive reciprocity. We considered this number of subjects as inappropriate to obtain either valid results or significant associations with self-reported data. However, identifying positive reciprocity was possible for a subsample of 29 couples. We used the transformed κ statistic as a measure of pattern in sequential data that ranges from +1 to −1 (the transformed κ differs from the κ used to assess interrater agreement, for details see Bakeman, McArthur, & Quera, 1996). Values close to 1 indicate that the occurrence of the stimulus behavior (e.g., husband's positive communication) increases the probability that a certain response behavior will follow (e.g., wife's positive communication). Hence, the maximum value of 1 means that the consequent behavior follows the antecedent behavior perfectly. Values near zero indicate weak or no associations between stimulus behavior of the one partner and response of the other partner. Values close to −1 imply that the occurrence of the stimulus behavior decreases the probability that a certain response behavior will follow. Consequently, a κ index of −1 means that the consequent behavior is suppressed by the antecedent behavior (for details about the transformed κ statistic see Wampold, 1989; see also Bakeman et al., 1996). In order to assess positive reciprocity, we used two unidirectional κ (i.e., positive behavior of wife followed by positive behavior of husband and vice versa), and one bidirectional κ (mean of unidirectional κ). Table 4 displays the descriptive statistics for the reciprocity variables.

Table 4.  Descriptive Statistics of the Couples' Positive Reciprocity Assessed by Unidirectional and Bidirectional κ Values (N=29 couples)
  • Note. κ statistics (Wampold, 1989) computed from positive reciprocity measures.

  • a

    Positive reciprocity=dependence between positive communication of the caregiver as the stimulus behavior and positive communication of the care receiver as the response behavior.

  • b

    Positive reciprocity=dependence between positive communication of the care receiver as the stimulus behavior and positive communication of the caregiver as the response behavior.

Positive reciprocity
Unidirectional κ: Pos caregiver[RIGHTWARDS ARROW]Pos care receivera.33.4−.191
Unidirectional κ: Pos care receiver[RIGHTWARDS ARROW]Pos caregiverb.64.4−.191
Bidirectional κ.48.3−.191

The positive mean values of κ variables indicate that couples showed positive reciprocity in the analyzed video sequence. The displayed ranges indicate the existence of differences among couples in their observed reciprocal communication behavior (see Table 4). In some couples, there was a constant co-occurrence of their positive codes (indicated by a κ value of 1), whereas other couples did not show a dependence of their positive behaviors (values near 0 or slightly negative).

The final question was which relations, if any, exist between the amount of positive reciprocity and couples' self-reported data (see Table 5). Results show highly significant negative correlations between the three measures of positive reciprocity and caregiver depression indicating large negative effect sizes (Bortz & Doering, 1995; Cohen, 1992). To rule out a confounding effect of relationship satisfaction that correlates significantly with caregiver depression and positive reciprocity, we computed partial correlations. However, they revealed no significant changes in the correlations between depression and positive reciprocity. Moreover, we found both neuropsychiatric symptoms and caregiver distress to be substantially correlated with the positive reciprocity measures (correlations range from −.20 to −.28). All correlations between communication and self-reports were controlled for marriage duration, duration of dementia, and duration of care. The partial correlations showed no significant or substantial changes in the original correlations. In order to identify possible group differences between couples without reciprocity measures and couples with reciprocity measures (eight couples did not show positive communication codes), t tests were conducted. There was no significant difference between the groups, but an almost significant t-value for caregiver depression (t=2.01, p=.06; d=.50), indicating that wives without dyadic positive reciprocity in their couple interaction report higher levels of depression.

Table 5.  Bivariate Pearson's Correlations Between Positive Reciprocity and Self-Report Measures
 CES-D caregiverNPI-Q distressRASNPI-Q severityMMSECES-D care receiver
  1. Note. CES-D=Center for Epidemiologic Studies Depression Scale (Radloff, 1977); NPI-Q=Brief Version of the Neuropsychiatric Inventory Questionnaire (Kaufer et al., 2000); MMSE=Mini Mental State Examination (Folstein, Folstein, & McHugh, 1975). a Positive reciprocity=dependence between positive communication of the caregiver as the stimulus behavior and positive communication of the care receiver as the response behavior. b Positive reciprocity=dependence between positive communication of the care receiver as the stimulus behavior and positive communication of the caregiver as the response behavior. *p<.05 , **p<.01.

κ of positive reciprocity
Pos caregiver[RIGHTWARDS ARROW]Pos care receivera−.51**−.26.28−.25.12.08
Pos care receiver[RIGHTWARDS ARROW]Pos caregiverb−.58**−.20.41*−.24−.01.15

Finally, it must be clarified whether positive communication and positive reciprocity contribute independently to caregiver depression. No significant bivariate correlation between base rate of positive codes and positive reciprocity (by bidirectional κ) was found (r=.19, p=.32). However, in order to obtain if the significant association between caregiver depression and positive reciprocity can be explained by the base rates of couples' positive communication and vice versa, partial correlations were conducted (see Table 6). Results show that both couple's positive communication and positive reciprocity contribute independently to the variance of depressive symptoms reported by the caregiving wives.

Table 6.  Zero-Order and Partial Correlations of Caregiver's CES-D and Couple's Positive Communication
 CES-D caregiver
  1. Note. CES-D=Center for Epidemiologic Studies Depression Scale (Radloff, 1977). a Couple's base rate of positive codes (bidirectional)=sum of positive codes base rate of caregiver and care receiver. rpartial=correlation between the two variables after eliminating the effect of the third variable (positive reciprocity, couples base rates of positive communication). *p<.05 , **p<.01.

Couple's base rate of positive codes (bidirectional)ar−.40*
Positive reciprocityr−.62**


  1. Top of page
  2. Abstract

The high average depression scores reported by caregivers in this sample clearly confirm the often demonstrated negative psychological outcomes of spousal dementia caring (e.g., Pinquart & Soerensen, 2003). This finding emphasizes, first, the relevance of evaluating afflicted individuals caring for their partners with dementia and second, the need to identify correlates or influencing factors of caregiver psychological health and well-being. Indeed, our results identified dyadic communication as being significantly related to caregiver depressive symptoms. The importance of this relation will be discussed in the next sections.

The first aim of the study was to basically describe the way afflicted couples communicate. In general, our analyses revealed a high frequency of neutral communication as well as low frequencies of positive and negative communication. This uneven distribution of marital interaction characteristics was often reported in past studies (e.g., Heyman, 2004). The second research question concerned possible differences between communication of the caregiving wives and the care-receiving husbands. Findings demonstrated slightly different patterns: Caregiving wives used more total, neutral, and negative, but less positive communication than their husbands. These differences only partially overlap with the communication patterns reported by Gallagher-Thompson et al. (1997), which indicated less positive and negative communication as well as more neutral and more total communication of the caring wives compared with their partners with dementia. It might be possible that our findings are associated with the differing levels of depressive symptoms of the partners: The caring wives suffered from significantly higher levels of depression than their husbands. This is consistent with previous studies with noncaring couples demonstrating that distressed individuals show less positive and more negative communication patterns during marital interaction than nondistressed individuals (for an overview see Heyman, 2001). Furthermore, past research revealed evidence for the positive correlation between depression and negative communication (e.g., Gallagher-Thompson et al., 1997; Heene, Buysee, & Van Oost, 2003; Heene, Buysee, & Van Oost, 2005; Heene, Buysee, & Van Oost, 2007).

The third aim of this study was to identify associations between depression and marital communication. Results revealed a strong positive relation between caring wives' mental health and their husbands' level of positive communication. This raises the question of whether positive communication of the care-receiving partner has buffering, protective effects on spousal caregiver well-being. As previous research demonstrated, neuropsychiatric symptoms and behavioral problems of the patient are particularly burdensome for the spousal caregiver (e.g., Perren et al., 2006; Perren, Schmid, Herrmann, & Wettstein, 2007). The relevance of reduced dyadic communication quality and low occurrence of positive communication might be partially responsible for this negative impact of neuropsychiatric symptoms. The substantial, but nonsignificant negative correlations between patients' positive communication and level of neuropsychiatric symptoms and dementia severity, as found in this study, might support this assumption. However, due to the small sample size and the explorative character of the study, we cannot draw valid conclusions about the direction of the described associations.

Last but not least, the fourth aim of this study was the examination of potential dependencies between both partners' communication patterns. The conducted sequential analyses outlined the importance of dyadic reciprocal positive communication. The finding that wives whose marital interaction consisted of more positive reciprocity reported a higher psychological well-being than wives with less or no positive reciprocal communication underlines the elsewhere demonstrated relevance of assessing dyadic data in afflicted couples in research studies (e.g., Braun et al., 2009; Clare, 2002).

From a theoretical perspective, these results are consistent with the assumptions of equity theory and other social exchange theories (Walster, Walster, & Berscheid, 1978). These theories assume that the balance between what is given and what is received in an exchange essentially influences well-being of individuals (Gergen & Gergen, 1986). Furthermore, it is expected that individuals who are confronted with inequity in their relationships will feel distress. Accordingly, studies showed that equitable couples seem to be happier with their relationship, less depressed, and more confident that the relationship will last than inequitable couples (Schafer & Keith, 1980; Walster, Traupman, & Walster, 1978). These theoretical principles may also provide a theoretical framework to examine spousal dementia caregiving (Baikie, 2002; Braun et al., 2009). It can be assumed that positive reciprocity in dyadic communication patterns is interpreted as a still existing mutual give and take between caregiving and care-receiving partner. Accordingly, this positive dyadic exchange might be a protective factor reducing or buffering the negative consequences of spousal dementia caregiving, such as depression. Caregivers whose dyadic communication includes no reciprocity or low levels of reciprocity in their positive communication might feel confronted with the negative changes in their marital interaction caused by the disease and the patients' typical verbal impairments.

However, as mentioned before, this study was conducted as an explorative description of communication patterns in couples with one spouse suffering from dementia and to identify possible associations between marital interaction and depression. Nevertheless, due to the fact that dyadic communication is known to be an essential aspect of spousal relationships, it might be plausible to assume that the existence of not only positive, but also reciprocal interaction patterns positively affects caregiver well-being and psychological health.


Before discussing the implications of this investigation, we have to point out its shortcomings. The main limitations of this explorative study concern size and composition of the sample, as well as problematic validity of data of individuals with dementia raising the question of limited generalizability. Because of the fact that we exclusively focussed on female caregivers, it remains unclear whether these findings possess validity for the converse constellation, husbands providing care for their wives with dementia. Previous research documented more negative effects of caregiving and higher burden levels for female than male caregivers (e.g., Thompson et al., 2004). However, possible effects of gender on the strength of the relationship between spousal communication and psychological well-being have to be investigated in future studies with gender-balanced samples. Another problematic aspect that might limit this study's findings considers the insufficient control of socioeconomic and ethnic variables. Although all participants were Caucasian, it might be possible that factors such as education and income influence caregiver well-being. Additionally, due to the cross-sectional study design and the use of correlational methods, causal relationships were not depictable. Indeed, evaluations with regression or multivariate models are not recommended with small sample sizes and have to be tested in future studies.

In order to include dyadic variables (e.g., communication patterns, depressive symptoms), self-reported and observational data from both partners were assessed in this investigation. Because of the questionable accuracy of statements of individuals with dementia, this information might be biased due to concentration difficulties or misunderstanding during assessment of the patients. Consequently, dementia patients are often seen as unreliable respondents (Cotrell & Schulz, 1993). Nevertheless, a better insight into relationships of afflicted couples is unapproachable when one half of the dyad is neglected. Some authors critically outlined the small number of studies including the perspectives of individuals suffering from dementia and demonstrated the usefulness of including patients in dementia research, assessing the dyadic perspective, respectively (e.g., Clare, 2002; for a review see Braun et al., 2009). Therefore, this study's approach, the integration of self-reported and observational data of individuals with dementia, should be seen as an advantageous possibility to develop a profound knowledge of the situation of afflicted couples.

Implications for Future Research and Clinical Practice

Future longitudinal studies including subjective and observational as well as individual and dyadic data of both caregiver and care receiver are needed to address the direction of the associations between marital interaction and caregiver mental health. Even though past studies with healthy couples revealed evidence for direct effects of marital communication on well-being and health (for a review see Burman & Margolin, 1992), it might also be possible that caregiver stress (e.g., depressive symptoms, distress level due to patient's neuropsychiatric symptoms) negatively affects verbal abilities of the care receiver.

Despite these methodological issues, the demonstrated correlations between dyadic interaction and well-being in couples afflicted with dementia provide essential information for therapeutic interventions. If longitudinal investigations will show that positive communication within affected dyads actually reduces caregiver disposition to depression, interventions focusing on dyadic interaction could be an effective way to increase caregiver well-being and to facilitate caregiver adaptive capacities concerning changes in their partners' communication due to dementia symptoms (e.g., aphasic symptoms). Previous studies demonstrated that therapeutic communication trainings focusing on positive communication techniques (e.g., smiling, affective touch) for medical professionals increase patient and nursing staff quality of life (for a review see Levy-Storms, 2008). Consequently, trainings for carers focusing on caregiver communication skills to maintain verbal abilities of the spouse with dementia might enhance psychological well-being of both caregiving and care-receiving partner. Engel (2007) developed a training for caring family members in order to improve caregiver communication skills adapted to the individual capacities of the dementia patient and demonstrated positive intervention effects. Furthermore, communication trainings for individuals with dementia focusing on abilities and resources of the partner with dementia (in particular the use of positive communication) could strengthen caregiver well-being and psychological health, as well. As an example, Tribet, Boucharlat, and Myslinski (2008) showed that animal-assisted therapy for individuals suffering from severe dementia was related with an increase of the patients' social interactions and nonverbal communication behaviors (see Walsh 2009a, 2009b). These findings imply that dementia sufferers can ameliorate their competencies even in the severe stage of the illness. Thus, the connections between marital communication and caregiver depression demonstrated in the present study indicate that facilitating patient communicative abilities might reduce caregiver distress, as well. Consequently, we assume that communication interventions specialized on needs and impairments of the caring and the care-receiving partner are beneficial for both dyad members.


  1. Top of page
  2. Abstract

As demonstrated here, the evaluation of marital interaction in couples with dementia includes essential information to understand the burdensome situation of spousal caregivers and their partners suffering from dementia. Moreover, not only individual communication characteristics and decreasing verbal ability of the care receiver as shown in previous studies (e.g., Savundranayagam, Hummert, & Montgomery, 2005), but also dyadic communication patterns and in particular communication dependencies between the interacting partners are associated with caregiver psychological well-being.


  1. Top of page
  2. Abstract
  • Baikie, E. (2002). The impact of dementia on marital relationships. Sexual and Relationship Therapy, 17 (3): 289299.
  • Bakeman, R., & Gottman, J.M. (1986). Observing interaction: An introduction to sequential analysis. Cambridge, UK: Cambridge University Press.
  • Bakeman, R., McArthur, D., & Quera, V. (1996). Detecting group differences in sequential association using samples permutations: Log odds, kappa, and phi compared. Behavior Research Methods, Instruments, & Computers, 28 (3): 446457.
  • Bakeman, R., Quera, V., McArthur, D., & Robinson, B.F. (1997). Detecting sequential patterns and determining their reliability with fallible observers. Psychological Methods, 2, 357370.
  • Bodenmann, G., Kaiser, A., Hahlweg, K., & Fehm-Wolfsdorf, G. (1998). Communication patterns during marital conflict: A cross-cultural replication. Personal Relationships, 5 (3): 343356.
  • Bortz, J., & Doering, N. (1995). Forschungsmethoden und evaluation. Berlin: Springer.
  • Braun, M., Scholz, U., Bailey, B., Perren, S., Hornung, R., & Martin, M. (2009). Dementia caregiving in spousal relationships: A dyadic perspective. Aging and Mental Health, 13 (3): 426436.
  • Burman, B., & Margolin, G. (1992). Analysis of the association between marital relationships and health problems: An interactional perspective. Psychological Bulletin, 112 (1): 3963.
  • Caughlin, J.P., & Huston, T.L. (2002). A contextual analysis of the association between demand/withdraw and marital satisfaction. Personal Relationships, 9 (1): 95119.
  • Clare, L. (2002). We'll fight it as long as we can: Coping with the onset of Alzheimer's disease. Aging and Mental Health, 6, 139148.
  • Cohen, J. (1992). A power primer. Psychological Bulletin, 112 (1): 155159.
  • Cordova, J.V., Jacobson, N.S., Gottman, J.M., Rushe, R., & Cox, G. (1993). Negative reciprocity and communication in couples with a violent husband. Journal of Abnormal Psychology, 102 (4): 559564.
  • Cotrell, V., & Schulz, R. (1993). The perspective of the patient with Alzheimer's disease: A neglected dimension of dementia research. The Gerontologist, 33 (2): 205211.
  • Crowell, J.A., Treboux, D., Gao, Y., Fyffe, C., Pan, H., & Waters, E. (2002). Assessing secure base behavior in adulthood: Development of a measure, links to adult attachment representations, and relations to couples' communication and reports of relationships. Developmental Psychology, 38 (5): 679693.
  • DeVugt, M.E., Stevens, F., Aalten, P., Lousberg, R., Jaspers, N., Winkens, I. et al. (2003). Behavioural disturbances in dementia patients and quality of the marital relationship. International Journal of Geriatric Psychiatry, 18, 149154.
  • Eloniemi-Sulkava, U., Notkola, I.L., Hamalainen, K., Rahkonen, T., Viramo, P., Hentinen, M. et al. (2002). Spouse caregivers' perceptions of influence of dementia on marriage. International Psychogeriatrics, 14 (1): 4758.
  • Engel, S. (2007). Belastungserleben bei Angehörigen Demenzkranker aufgrund von Kommunikationsstörungen. Berlin: LIT.
  • Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). “Mini mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12 (3): 189198.
  • Gallagher-Thompson, D., Dal Canto, P.G., Darnley, S., Basilio, L.A., Whelan, L., & Jacob, T. (1997). A feasibility study of videotaping to assess the relationship between distress in Alzheimer's disease caregivers and their interaction style. Aging and Mental Health, 1 (4): 346355.
  • Gallagher-Thompson, D., Dal Canto, P.G., Jacob, T., & Thompson, L.W. (2001). A comparison of marital interaction patterns between couples in which the husband does or does not have Alzheimer's disease. Journal of Gerontology Series B: Psychological Sciences and Social Sciences, 56, 140150.
  • Gergen, K.J., & Gergen, M.M. (1986). Social psychology (2nd ed.). New York: Springer.
  • Gilhooly, M.L., Sweeting, H.N., Whittick, J.E., & McKee, K. (1994). Family care of the dementing elderly. International Review of Psychiatry, 6, 2940.
  • Gill, D.S., Christensen, A., & Fincham, F.D. (2005). Predicting marital satisfaction from behavior: Do all roads really lead to Rome? Personal Relationships, 6 (3): 369387.
  • Gottman, J.M., & Notarius, C.I. (2000). Decade review: Observing marital interaction. Journal of Marriage and the Family, 62, 927947.
  • Grober, E., Hall, C., Lipton, R.B., & Teresi, J.A. (2008). Primary care screen for early dementia. Journal of the American Geriatrics Society, 56 (2): 206213.
  • Hautzinger, M. (1988). Die CES-D Skala. Ein Depressionsmessinstrument für Untersuchungen in der Allgemeinbevölkerung. Diagnostica, 38, 167173.
  • Heene, E., Buysee, A., & Van Oost, P. (2003). A categorical and dimensional perspective on depression within a nonclinical sample of couples. Family Process, 42, 133149.
  • Heene, E., Buysee, A., & Van Oost, P. (2005). Indirect pathways between depressive symptoms and marital distress: The role of conflict communication, attributions, and attachment style. Family Process, 44 (4): 413440.
  • Heene, E., Buysee, A., & Van Oost, P. (2007). An interpersonal perspective on depression: The role of marital adjustment, conflict communication, attributions, and attachment within a clinical sample. Family Process, 46 (4): 499514.
  • Hefner, K.L., Loving, T.J., Kiecolt-Glaser, J.K., Himawan, L.K., Glaser, R., & Malarkey, W.B. (2006). Older spouses' cortisol responses to marital conflict: Associations with demand/withdraw communication patterns. Journal of Behavioral Medicine, 29 (4): 317325.
  • Hendrick, S.S. (1988). A generic measure of relationship satisfaction. Journal of Marriage and the Family, 50, 9398.
  • Heyman, R.E. (2001). Observation of couple conflicts: Clinical assessment applications, stubborn truths, and shaky foundations. Psychological Assessment, 13 (1): 535.
  • Heyman, R.E. (2004). Rapid marital interaction coding system (RMICS). In P.K.Kerig & D.H.Baucom (Eds.), Couple observational coding systems (pp. 6793). London: Lawrence Erlbaum.
  • Heyman, R.E. (2007). RMICS coding center: Rapid marital interaction coding system. Retrieved August 25, 2008, from
  • Heyman, R.E., & Vivian, D. (1993). Rapid marital interaction coding system (RMICS): Training manual for coders. Retrieved August 1, 2008, from
  • Heyman, R.E., Weiss, R.L., & Eddy, J.M. (1995). Marital interaction coding system: Revision and empirical evaluation. Behavioural Research and Therapy, 33, 737746.
  • Kaufer, D.I., Cummings, J.L., Ketchel, P., Smith, V., MacMillan, A., Shelley, T. et al. (2000). Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. Journal of Neuropsychiatry and Clinical Neurosciences, 12 (2): 233239.
  • Kessler, J., Markowitsch, H.J., & Denzler, P. (2000). Mini-Mental-Status Test (MMST). Göttingen: Beltz Test GMBH.
  • Kiecolt-Glaser, J.K., Loving, T.J., Stowell, J.R., Malarkey, W.B., Lemeshow, S., Dickinson, S.L. et al. (2005). Hostile marital interaction, proinflammatory cytokine production, and wound healing. Archives of General Psychiatry, 62 (12): 13771384.
  • Knop, D.S., Bergman-Evans, B., & McCabe, B.W. (1998). In sickness and in health: An exploration of the perceived quality of the marital relationship, coping, and depression in caregivers of spouses with Alzheimer's disease. Journal of Psychosocial Nursing and Mental Health Services, 36 (1): 1621.
  • Levy-Storms, L. (2008). Therapeutic communication training in long-term care institutions: Recommendations for future research. Patient Education and Counseling, 73, 821.
  • Maki, N., Ikeda, M., Hokoishi, K., Nebu, A., Komori, K., Hirono, N. et al. (2000). The validity of the MMSE and SMQ as screening tests for dementia in the elderly general population—a study of one rural community in Japan. Dementia and Geriatric Cognitive Disorders, 11 (4): 193196.
  • Manne, S., Sherman, M., Ross, S., Ostroff, J., Heyman, R.E., & Fox, K. (2004). Couples' support-related communication, psychological distress, and relationship satisfaction among women with early stage breast cancer. Journal of Consulting and Clinical Psychology, 72 (4): 660670.
  • Margolin, G., & Wampold, B.E. (1981). Sequential analysis of conflict and accord in distressed and nondistressed marital partners. Journal of Consulting and Clinical Psychology, 49 (4): 554567.
  • Medina, J., & Weintraub, S. (2007). Depression in primary progressive aphasia. Journal of Geriatric Psychiatry and Neurology, 20 (3): 159160.
  • Narayan, S., Lewis, M., Tornatore, J., Hepburn, K., & Corcoran-Perry, S. (2001). Subjective responses to caregiving for a spouse with dementia. Journal of Gerontological Nursing, 27 (3): 1928.
  • Perren, S., Schmid, R., Herrmann, D., & Wettstein, A. (2007). The impact of attachment on dementia-related problem behavior and spousal caregivers' well-being. Attachment and Human Development, 9 (2): 163178.
  • Perren, S., Schmid, R., & Wettstein, A. (2006). Caregivers' adaptation to change: The impact of increasing impairment of persons suffering from dementia on their caregiver's subjective well-being. Aging and Mental Health, 10, 539548.
  • Pinquart, M., & Soerensen, S. (2003). Associations of stressors and uplifts of caregiving with caregiver burden and depressive mood: A meta-analysis. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58 (2): 112128.
  • Polk, D.M. (2005). Communication and family caregiving for Alzheimer's dementia: Linking attributions and problematic integration. Health Communication, 18 (3): 257273.
  • Radloff, L.S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385401.
  • Revenstorf, D., Vogel, B., Wegener, C., Hahlweg, K., & Schindler, L. (1980). Escalation phenomena in interaction sequences. Behaviour Analysis and Modification, 2, 97115.
  • Savundranayagam, M.Y., Hummert, M.L., & Montgomery, R.J.V. (2005). Investigating the effects of communication problems on caregiver burden. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60, 4855.
  • Schafer, R.B., & Keith, P.M. (1980). Equity and depression among married couples. Social Psychology Quarterly, 43 (4): 430435.
  • Schmitt, M., Kliegel, M., & Shapiro, A. (2007). Marital interaction in middle and old age: A predictor of marital satisfaction? International Journal of Aging and Human Development, 65 (4): 283300.
  • Snyder, D.K., Heyman, R.E., & Haynes, S.N. (2005). Evidence-based approaches to assessing couple distress. Psychological Assessment, 17 (3): 288307.
  • Stanley, S.M., Markman, H.J., & Whitton, S.W. (2002). Communication, conflict, and commitment: Insights on the foundations of relationships success from a national survey. Family Process, 41, 659675.
  • Thompson, R.L., Lewis, S.L., Murphy, M.R., Hale, J.M., Blackwell, P.H., Acton, G.J. et al. (2004). Are there sex differences in emotional and biological responses in spousal caregivers of patients with Alzheimer's disease? Biological Research for Nursing, 5 (4): 319330.
  • Tribet, J., Boucharlat, M., & Myslinski, M. (2008). Animal-assisted therapy for people suffering from severe dementia. L' Encéphale, 34 (2): 183186.
  • Vitaliano, P.P., Zhang, J., & Scanlan, J.M. (2003). Is caregiving hazardous to one's physical health? A meta-analysis. Psychological Bulletin, 129 (6): 946972.
  • Walsh, F. (2009a). Human-animal bonds I: The relational significance of companion animals. Family Process, 48 (4): 462480.
  • Walsh, F. (2009b). Human-animal bonds II: The role of pets in family systems and family therapy. Family Process, 48 (4): 481499.
  • Walster, E., Traupmann, J., & Walster, G.W. (1978). Equity and extramarital sexuality. Archives of Sexual Behavior, 7 (2): 127142.
  • Walster, E., Walster, G.W., & Berscheid, E. (1978). Equity: Theory and research. Boston: Allyn and Bacon.
  • Wampold, B.E. (1989). Kappa as a measure of pattern in sequential data. Quality and Quantity, 23, 171187.
  • Wampold, B.E., & Margolin, G. (1982). Nonparametric strategies to test the independence of behavioral states in sequential data. Psychological Bulletin, 92 (3): 755765.
  • Weiss, R.L., & Heyman, R.E. (1997). A clinical overview of couples interactions. In W.K.Halford & H.J.Markman (Eds.), Clinical handbook of marriage and couples interventions (pp. 1341). New York: Wiley & Sons.