شباهت شخصیت و رضایت از زندگی در زوج ها
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|37631||2013||7 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Research in Personality, Volume 47, Issue 4, August 2013, Pages 369–375
The present study examined the association between personality similarity and life satisfaction in a large, nationally representative sample of 1608 romantic couples. Similarity effects were computed for the Big Five personality traits as well as for personality profiles with global and differentiated indices of similarity. Results showed substantial actor and partner effects, indicating that both partners’ personality traits were related to both partners’ life satisfaction. Personality similarity, however, was not related to either partner’s life satisfaction. We emphasize the importance of thoroughly controlling for each partner’s personality and for applying appropriate analytical methods for dyadic data when assessing the effect of personality similarity in couples.
Personal relationships in general, and romantic relationships in particular, are essential for people’s well-being. Romantic relationships represent a unique social unit in which partners spend a considerable amount of time together and share closeness and intimacy with each other (Argyle, 1999). A recent longitudinal study on development of life satisfaction in couples revealed that both members of a couple influence each other and mutually affect the other partner’s well-being (Hoppmann, Gerstorf, Willis, & Schaie, 2011). In this regard, a dyadic approach that focuses on couples as the unit of analysis is of crucial interest for personality psychology. Correspondingly, within- and between-person as well as similarity effects of personality on well-being have received considerable attention in recent research and both one’s own as well as the partner’s personality characteristics have been found to be important factors in intimate relationships (e.g., Barelds, 2005, Dyrenforth et al., 2010, Gonzaga et al., 2007 and Luo and Klohnen, 2005). Evidence so far speaks for consistent effects of one’s own personality on well-being (actor effects), such that being extraverted, agreeable, conscientious, emotionally stable and open to experience is positively related to a person’s well-being (e.g., Steel, Schmidt, & Shultz, 2008). But it is not only one’s own personality that affects well-being, it is also the partner’s personality (partner effects). Being in an intimate relationship with someone who is extraverted, agreeable, conscientious, and emotionally stable is associated with higher well-being as well (Barelds, 2005, Dyrenforth et al., 2010, Headey et al., 2010 and Robins et al., 2000). However, as far as personality similarity in romantic couples is concerned, the everlasting question about who is a person’s perfect match has not yet been answered satisfactorily. Is having a partner whose personality is similar to one’s own associated with higher levels of satisfaction? Or do we prefer someone rather dissimilar? The present study tries to shed light on this issue and examines how personality similarity in couples relates to each partner’s life satisfaction. Research to date leans more towards “like attracts like” rather than “opposites attract” and assumes that similarity, as opposed to complementarity (Antill, 1983), is an important factor in romantic relationships. Studies on mate selection for instance support the notion that similarity between partners is essential in forming intimate relationships. People tend to choose partners who are similar to themselves on a number of variables, including age, education, values, physical attractiveness, and intelligence (Epstein and Guttman, 1985 and Vandenberg, 1972). Similarity is theorized to be beneficial for intimate relationships by coordinating partners’ thoughts and behaviors, increasing understanding of each other’s intentions and motivations, and reinforcing their appraisals, leading to relationship satisfaction and longevity (Anderson, Keltner, & John, 2003). Nevertheless, spousal similarity in personality traits is generally quite low (Watson et al., 2004) although people tend to prefer partners with similar personalities (Botwin, Buss, & Shackelford, 1997). The existing body of research on the association between personality similarity and well-being in couples does not provide a clear picture. Some studies show positive effects for relationship or life satisfaction respectively, such that the more similar couples’ personalities, the higher each partner’s satisfaction in the respective domain (Arrindell and Luteijn, 2000, Gaunt, 2006, Gonzaga et al., 2010, Gonzaga et al., 2007, Luo and Klohnen, 2005 and Robins et al., 2000). However, a recent study using nationally representative panel data from Great Britain, Australia, and Germany revealed no or only small associations between personality similarity and relationship or life satisfaction and neither of these small effects was consistent across the three samples (Dyrenforth et al., 2010). Thus, even with big sample sizes having the power to detect very small effects, there was little evidence that personality similarity predicted relationship or life satisfaction in those three large samples. Furthermore, in a representative sample of Dutch couples, personality similarity did not affect marital quality (Barelds, 2005) and in a sample of distressed and treatment-seeking couples, similarity on any of the Big Five traits was not a predictor of marital satisfaction (Gattis, Berns, Simpson, & Christensen, 2004). Given the inconsistent empirical evidence, research is needed in order to deepen our understanding of how personality similarity in romantic relationships relates to each partner’s well-being. For this purpose, three important factors that have been neglected in several previous studies have to be considered. First, when assessing the effect of personality similarity in couples, it is essential to control for each partner’s individual level of personality to get the unique similarity effect beyond each partner’s individual contribution (Griffin, Murray, & Gonzalez, 1999; Kenny & Acitelli, 1994). Not controlling for initial levels of both partners’ personality leads to an overestimation of the association between similarity and well-being. Some of the studies reporting significant similarity effects did not or did not thoroughly control for individual levels of partners’ personality (Gonzaga et al., 2007 and Robins et al., 2000). When main effects were included in the analysis, however, similarity was no longer a unique predictor of well-being (Barelds, 2005, Dyrenforth et al., 2010 and Gattis et al., 2004). Second, to gain generalizable results and a more accurate sense of the association between personality similarity and well-being, it is necessary to examine a large representative sample of couples. Several studies that found similarity effects on well-being analyzed relatively small (Arrindell and Luteijn, 2000 and Gonzaga et al., 2007) or very specific and thus possibly biased samples such as newlyweds (Luo & Klohnen, 2005), cohort study members (Robins et al., 2000) or couples who met via online dating platforms (Gonzaga et al., 2010). However, the few studies examining large representative samples failed to find similarity effects on well-being (Barelds, 2005 and Dyrenforth et al., 2010). Third, in light of the dyadic nature of couple data it is necessary to make use of proper analytic techniques and similarity measures. Thus far, many studies have treated dyadic data as if they were individual data, for instance by conducting analyses separately for husbands and wives or using simple correlational methods that fail to capture the interdependent nature of couple data. This shortcoming in previous studies has been pointed out as a major problem by relationship researchers and can be overcome by applying appropriate analytical methods that take the interdependence of dyadic data into account and are able to test for the unique effect of each independent variable on well-being (see Kenny, Kashy, & Cook, 2006). Furthermore, studies on personality similarity in couples have used a wide array of similarity measures. The most common measures reported include difference scores and profile correlations. Difference scores are straightforward and intuitively understandable in the way we usually think about differences in daily life (Griffin et al., 1999). These scores are typically computed at the trait level, taking the absolute value of difference between two partners’ scores on a given trait. They thus indicate how dissimilar two members of a couple are with respect to a specific trait. Yet researchers criticize this approach mainly due to the lowered reliability inherent in difference scores. If the two component variables are positively correlated, as is often the case given that the scores are usually measured with the same instrument, the reliability of the difference between those two components becomes less reliable (Edwards, 1994 and Edwards, 2001). Furthermore, difference scores can be confounded with each partner’s individual score. A simple solution to avoid this problem is to include both partners’ individual scores in the same analysis (Griffin et al., 1999). Profile correlations, on the other hand, are more difficult to interpret. Generally, personality similarity computed at the profile level represents the degree to which both couple members’ overall personality profiles are similar to each other; that is, how well two partners match on a set of personality traits (see Cronbach & Gleser, 1953). To interpret the effects of profile similarity adequately, it is important to know that a profile consists of three elements (Cronbach & Gleser, 1953; see also Furr, 2010). First, every profile has a shape that represents the pattern of scores in a profile. It reflects which traits have relatively high scores and which ones have relatively low scores within the same profile. Elevation represents the overall mean across all traits within a profile. And scatter refers to the variability or variance among the scores of a profile. It thus reflects how much the trait scores deviate within the same profile. Because a profile comprises these three elements, similarity between different profiles can be measured in various ways. A commonly used profile correlation measure is the Intraclass Correlation Coefficient (ICC). It reflects a global index because it captures all three characteristics of a profile at once (e.g., Dyrenforth et al., 2010). However, assuming ICC as a global index of personality similarity might confound findings because shape, elevation, and scatter are conceptually different from one another and should not be mixed within the same analysis. Instead, Furr (2010) suggests a differentiated analysis and argues for the necessity to separately examine all elements of a profile. Hence, shape similarity is calculated by correlating the scores in one partner’s profile with the scores of the other partner’s profile using Pearson correlations. Second, elevation similarity is measured using difference scores (i.e., absolute value of difference between the overall mean across traits: mean similarity). Third, scatter similarity is also computed using difference scores (i.e., absolute value of difference between the variances across all traits within the profiles: variance similarity). In light of the range of possible measures of similarity, another goal of the present study is to rule out that the use of different similarity indices results in different findings. By adopting a differentiated approach we can determine whether results vary depending on how profile similarity is measured. Some researchers who applied difference scores and profile correlations in the same study reported that profile-based similarity was more strongly associated with satisfaction (Gaunt, 2006 and Luo and Klohnen, 2005), whereas others did not find different effects with different similarity measures (Dyrenforth et al., 2010). No study to date has analyzed similarity by means of a differentiated approach; that is, examining each element of a profile separately (as proposed by Furr, 2010). In brief, the aim of the present study is to clarify the association between personality similarity and life satisfaction in couples. Three reasons speak for the relevance of this research question. First, we think that some previous findings mostly represent an overestimation of the effect of similarity because many studies did not take each partner’s personality into account (e.g., Robins et al., 2000). Thus, we will control for each partner’s individual level of personality to examine the effect of personality similarity on life satisfaction beyond the effects of one’s own and the partner’s personality. In line with previous research, we expect actor and partner effects for the association between personality and life satisfaction in couples (positive actor and partner effects for Extraversion, Agreeableness, Conscientiousness, Emotional Stability as well as positive actor effects for Openness; e.g., Dyrenforth et al., 2010). Second, with the exception of only a few studies (Barelds, 2005 and Dyrenforth et al., 2010), previous research mainly used small or specific samples (e.g., Arrindell and Luteijn, 2000 and Luo and Klohnen, 2005). Thus, in order to provide generalizable results we use a nationally representative sample of couples living in Switzerland. We expect to replicate findings of Dyrenforth et al., 2010 who analyzed comparable data sets of other countries. Third, we apply proper methods for dyadic data and extend prior research on similarity and well-being by adopting a differentiated approach: associations between personality similarity and life satisfaction in couples will be analyzed both at the trait and at the profile level. The latter includes global profile similarity (ICC), as well as similarity of shape, elevation (mean), and scatter (variance).