His or Her Parents? Perceived Parental Approval of Romantic Relationships Among College Students and Their Partners

Using social capital theory, this study of 194 dating couples examined the connection between parents’ approval of the dating relationship (reported by each couple member forhis/her own parents and the partner’s parents) and participants’ relationship distress. The Actor-Partner-Interdependence Model within Structural Equation Modeling served as the data-analytic tool. Results showed that, in support of the theory, relationship approval from strong ties (one’s own parents) and from weak ties (one’s partner’s parents) manifested themselves differently in romantic relationships. Specifically, both men’s and women’sperception of relationship approval from their own parents (strong ties) and from their partner’s parents (weak ties) negatively predicted couple members’ own relationshipdistress. Moreover, path coefficients between men’s and women’s strong ties and their own relationship distress were roughly twice as large as those between men’s and women’sweak ties and their relationship distress. Findings were less clear for the association between perceptions of relationship approval from one’s own and one’s partner’s parents and the dating partners’ relationship distress. The findings are discussed in light of prior research and theory on social capital.

1997 ;Sprecher, 1988).One particular area of interest in network support has been the role of parental relationship approval and its influence on romantic relationships (e.g., Blair & Homberg, 2008;Driscoll, Davis, & Lipetz, 1972;Etcheverry, Le, & Charania, 2008;Felmlee, 2001;Sprecher & Felmlee, 1992, 2000).This research has generally shown positive associations between parental approval of their grown children's romantic relationships and the quality and stability of those relationships (e.g., Sprecher, 1988;Sprecher & Felmlee, 2000).(An exception to this pattern is Driscoll and colleagues' study on the Romeo and Juliet effect.)Social capital theory maintains that the networks that link people (e.g., relationships between individuals and their parents, friends, or colleagues) are important conduits to social resources (Lin, 1999;Portes, 1998), with the usefulness of these resources depending on the situation and strength of the ties.This article applies social capital theory, especially the concept of strong and weak ties, to romantic relationships and examines the ways in which different strengths of ties may influence romantic relationship distress.We conceptualize parental approval as a form of social capital, relationships with one's own parents as strong ties, and relationships with the partner's parents as weak ties.The first goal of this study is to explore whether approval from strong (own parents) and weak (partner's parents) ties function differently in relation to each partner's romantic relationship quality (assessed as relationship distress, as described below).A second, related goal is to examine how the same parents' approval, viewed from two different perspectives (his and hers), predicts relationship quality.For example, with the approval of the female participant's parents being assessed both via her perception of her parents and the male partner's perception of her parents, we can ask how these two views of the same parents' approval predict each partner's relationship quality (with the same reasoning applying to the male participant's parents).

Parental Approval as Social Capital in Romantic Relationships
Social capital is "the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition" (Bourdieu, 1985, p. 248).This concept was introduced by sociologists in an attempt to include intangible social ties as part of resources or capital.
Unlike other forms of capital (e.g., financial, human), social capital arises entirely from relationships between actors, whether it is people or organizations.Social capital can take different forms, but all derive from social structure and the actions of persons within the structure (Coleman, 1988).Social capital can be a source of social control, social support, and tangible benefits through networks inside and outside the family (Portes, 1998).
There is little research conceptualizing parental approval in terms of social capital.
Having a certain network does not necessarily mean that the network itself is one's social capital; rather the network can facilitate a member's access to further, external resources.
Social capital can be defined as "the ability to secure resources by virtue of membership in social networks or larger social structures" (Portes & Landolt, 2000, p. 532).Romantic relationships potentially can evolve into marriage, and marriages are a means of maintaining or gaining social status (Burgess & Cottrell, 1939;Glenn, Ross, & Tully, 1974;Martin, 1970).Parents' approval of their grown children's romantic relationship may thus convey the parents' willingness to accept the partner into the family.Membership in the family may also include access to the family's social resources.In this regard, parental approval can be viewed as a form of social capital.This last example also illustrates the distinction between bonding and bridging social capital (Putnam, 2000), which in some ways, parallels the framework of strong and weak ties.Within a familiar, strong-tie group such as one's family of origin, bonding can reinforce group norms and patterns of behavior.Support and encouragement from close family members may be a strength, but redundancy in members' perspectives may limit awareness of outside views.The bridging type of social capital occurs when one or more members of a cohesive group form (weak) ties with one or more members of another group, thus linking the two groups.Bridging/weak-tie social capital has the advantage of bringing new information, perspectives, and values to the attention of all involved (Granovetter, 1973), but social influence across boundaries of the two original groups may be more limited.When each of the two sets of parents welcomes their respective grown child's romantic partner into the family, this can be seen as a form of bridging social capital.
Several studies, although not necessarily within the framework of social capital, have examined the effects of social-network and parental approval on romantic relationship quality and stability (e.g., Felmlee, 2001;Sprecher & Felmlee, 2000).These studies support the notion that parental approval of a romantic relationship is important to the relationship.

Strong and Weak Ties
The value and usefulness of social capital usually depends on the strength of the ties that bind involved parties together.The strengths of ties are decided by the "combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and their reciprocal services which characterize the tie" (Granovetter, 1973(Granovetter, , p. 1361)).It seems that different strengths of ties are effective in different situations.For example, many researchers have found that strong ties are important in certain areas, such as immigrant entrepreneurship and ethnic businesses where a high level of trust is crucial (Light, 1984;Light & Bonacich, 1988;Portes & Stepick, 1993;Waldinger, 1996;Zhou, 1992).In a similar vein, the lack of strong ties is related to unemployment and welfare dependency (Wacquant & Wilson, 1989;Wilson, 1996).On the other hand, weak ties have been found to be more useful in finding a job than strong ties (Granovetter, 1973); as noted above, the apparent reason is that weak ties can provide new information that individuals need to find a job whereas information from strong ties can be redundant (Portes, 1998).
Previous research has mostly failed to consider that approval from at least two sets of parents is operating in romantic relationships.When an individual enters into a romantic relationship, the partner's parents become a part of the focal individual's own network, thus adding to the individual's social capital.However, many studies have examined only one partner and his or her perception of own parents' approval (e.g., Blair & Holmberg, 2008;Lehmiller & Agnew, 2007;Sprecher, 1988).Although findings from such studies have been useful in building the literature on parental approval, these studies ignore the dyadic nature of romantic relationships.Given that individuals' relationships with their own parents usually differ from those with their partner's parents, examining individuals' perceptions of approval from both sets of parents at the same time will provide a more complete picture.The current study, therefore, examines how the two sets of ties to parents differing in strengths operate in romantic relationships.We conceptualize the tie with one's own parents as strong and the one with the partner's parents as weak.This conceptualization draws from previous studies in which individuals' own networks (e.g., immediate family, friends, and relatives) constituted strong ties and their indirect network (e.g., friend's family and relatives), weak ties (e.g., Granovetter, 1973;Lin, Vaughn, & Ensel, 1981).

The Current Study
To evaluate our first research questionhow do strong and weak ties work in regard to parental approval in romantic relationships?we proposed and tested a Structural Equation Model (SEM) illuminating associations between individuals' perceptions of relationship approval from their own and their partner's parents, and participants' romantic relationship distress.In line with the earlier discussion of bonding vs. bridging social capital, we hypothesize (H1) that approval from one's own parents (i.e., strong-tie, bonding social capital) will more strongly predict participants' relationship distress (negatively) than will approval from one's partner's parents.
Our second research question asks how individuals' and their partners' perspectives of approval from the same parents (e.g., the female participant's) are related to relationship distress.Most previous parental-approval studies have included only one partner of a couple (e.g., Blair & Holmberg, 2008;Etcheverry et al., 2008;Felmlee, 2001;Sprecher, 1988).However, in studies examining interpersonal aspects of relationships, including both partners is important (Karney & Bradbury, 1995).Moreover, including both partners in the study has not always yielded findings that illuminated the dyadic nature of the relationship because of the use of inappropriate analytic strategies (e.g., analyzing male and female data separately).One of the important characteristics of dyadic data is the interdependence of partners' scores, thus requiring appropriate dyadic analytic tools to account for this interdependence (Kenny, Kashy, & Cook, 2006).Accordingly, we use the Actor-Partner Interdependence Model (APIM: Kenny, 1996;Kenny et al., 2006) for this purpose.APIM is "a model of dyadic relationships that integrates a conceptual view of interdependence in two-person relationships with the appropriate statistical techniques for measuring and testing it" (Cook & Kenny, 2005, p. 102).Actor effects refer to statistical associations between the same reporter's independent and dependent variables, whereas partner effects refer to associations across dyad members (e.g., the partner's independent variables predicting one's own dependent variables).The APIM estimates effects for both dyadic partners simultaneously while controlling for their nonindependence, and is appropriate when the model has both individual-and dyad-level variables (Kenny et al., 2006).Furman and Simon (2006) offered, as a seemingly general rationale for expecting actor effects to be stronger than partner effects, that "the links of one's views with one's own behavior are more direct than those with the other's behavior" (p.591).Our second hypothesis (H2) therefore is that each couple member's relationship distress will be more strongly predicted by his or her own estimations of parental approval than by the partner's estimations.

Participants
The current study included 194 college students and their romantic partners from two samples (Sample 1, N = 206 and Sample 2, N = 192), each at a different university.
One university was in the Mountain West and the other in the Southwest.The average age of male partners was 22.1 years (SD = 3.56, range = 18 to 44) and the average age of female partners was 20.6 years (SD = 2.50, range = 18 to 38).The average length of relationship was 15.52 months (SD = 15.87,range = 1 to 109).Approximately 83% of participants were Caucasian.Forty-seven percent of participants were of Mormon faith and 32.1% had other Christian affiliations.Sample 1 is mostly responsible for the high percentage of participants with Mormon faith -88.3% of Sample 1 was Mormon compared to only 0.5% in Sample 2. These results suggested that Samples 1 and 2 might be qualitatively different, thus possibly requiring separate models for each group.
A series of t-tests were conducted to explore the possibility.Results revealed that the two samples were statistically different in many ways: Sample 1 was significantly more religious, t (368) = 6.27, p < .001;perceived the partner's negative behaviors as more frequent, t (368) = -2.04,p < .05;experienced less disillusionment, t (369) = -2.04,p < .05;and perceived less approval for the relationship from the partner's father, t (343) = -2.41,p < .05than Sample 2. Therefore, a multiple-group SEM analysis (described below) was carried out to determine whether separate models would be required for the two samples.

Procedure
Procedures varied slightly for Samples 1 and 2. Upon approval from the Institutional Review Board of the relevant university, Sample 1 was recruited through an undergraduate course.Students in selected courses were able to receive extra credit for either participating in the study with their partner or finding a couple to participate in the study.Participants were also entered in a $20.00 cash drawing for participating.During several data-collection sessions, participating partners were instructed to sit across the room from each other and complete a survey (addressing numerous aspects of their relationship and demographic background) to reduce the potential for response bias.
The procedures for Sample 2 were similar but varied in some aspects.Upon approval from the Institutional Review Board, participants were recruited from several undergraduate courses and through a campus-wide email distribution service.Only participants who were recruited through undergraduate courses (through the course instructors' approval) received extra credit for participating in the study with their partner, as was the case for Sample 1. Participants attended data-collection sessions that occurred over the course of one year.Dating partners completed paper and pencil surveys simultaneously while sitting apart from each other in a classroom.All participants were entered in a drawing to win 1 of 5 $20.00 cash prizes.

Measures
Relationship approval from strong and weak ties.Four constructsfemale strong, female weak, male strong, male weakrepresenting perceived approval of the relationship had two single-item indicators each (referencing the mother and father).The items were adapted from Felmlee's (2001)  Relationship Distress.Three scales that addressed negative aspects of romantic relationshipsdisillusionment, uncertainty, and perception of partner's negative behaviorwere used to represent the larger distress construct.These scales were selected because of their ability to predict the overall quality and stability of romantic relationships; also, among the various distress measures used in this research, these three were the only ones administered in both samples.Hence, the use of the three measures preserved the largest possible overall sample size.
Past literature on disillusionment has shown it to be a predictor of relationship quality and a very strong predictor of marital stability (Huston, Caughlin, Houts, Smith & George, 2001).To measure disillusionment, we adapted the Marital Disillusionment Scale developed by Niehuis and Bartell (2006).The original 16-item measure examines disillusionment in a marriage as a decrease in the perception of positive feelings, cognitions, and behaviors, as well as an increase in perceptions of negative feelings, cognitions, and behaviors.Regret, in the sense that someone feels bad or sorry about something that happened at an earlier point in time and that now seems wrong or a mistake, as well as regret in the sense that someone feels sad at having lost something or someone, may be part of disillusionment (see Niehuis, Lee, Reifman, Swenson, & Hunsaker, in press, for an in-depth discussion of the concept).We reworded 11 of the original 16 items so that they addressed dating, as opposed to marital, relationships (e.g., changing the word "spouse" to "partner").Five items were excluded because they could not easily be translated into dating relationships (e.g., "Marriage used to be a scared bond; now/later it is/was just a legal document").Examples of items that were used include: "I am very disappointed in my partner;" "My partner seems to be an entirely different person now," "My partner used to be on her/his best behavior when with me, but now he/she doesn't bother trying to impress me," and "I used to think I was lucky to be with someone like my partner; now I'm not so sure that I am so lucky."Questions were answered on a 7-point Likert Scale, with 1 = Strongly disagree and 7 = Strongly agree.Higher scores on this scale indicated higher levels of disillusionment.Cronbach's alpha was .94.
Uncertainty has been found to be associated with the quality and stability of romantic relationships (Parks & Adelman, 1983;Planalp, Rutherford, & Honeycutt, 1988;Schwebel, Moss, & Fine, 1999;Siegert & Stamp, 1994).To measure relationship uncertainty, we used Parks and Adelman's (1983) Uncertainty scale.This eight-item scale assesses individuals' ability to predict their partner's behavior, to assess uncertainty in the relationship.This scale contains questions such as, "I am confident about my ability to accurately predict my dating partner's behavior," and "My dating partner often does or says things which surprise me."All answers were reported using a 5-point Likert scale indicating the frequency with which respondents felt able to predict their partners' behavior (1=Never, 5=Very often).Five items were recoded so that higher scores reflect higher levels of uncertainty.The Cronbach alpha was .80.Individuals' perception of partner's negativity has been reported to be related to relationship satisfaction and stability (Huston et al., 2001;Huston & Vangelisti, 1991).
Perceptions of partner's negative behavior were measured using an adaptation of Huston and Vangelisti's (1991) Socioemotional Behavior Interview.Participants were asked to provide the number of times over the past 24 hours that their partner behaved towards them in a negative manner on a total of seven items.Example items include, "How often did your partner seem bored or uninterested while you were talking?"A sum score across the responses to the seven items was calculated.Cronbach's alpha was .72.
Control Variables.Based on correlations among variables (see Appendix), we identified three control variables that could relate to parental approval and/or relationship distress: age, religiosity (which could also help to control for differences between the samples), and relationship seriousness.Age was measured by asking the participant, "How old are you?"Religiosity was measured by asking participants to indicate how religious they were on a 6-point Likert scale with 1 = Not at all religious and 6 = Very religious.
Relationship seriousness was assessed by averaging each pair of partners' responses to a question asking participants about the level of involvement with their partner (1 = Casually dating, 5 = Engaged to be married).The correlation between partners' reports was r = .84.

Data Analytic Strategy
Within our SEM/APIM approach, we created four constructs representing parental approval as social capital: approval of the relationship from own and partner's parents for the male and female couple members.These constructs reflect how strong and weak ties are represented in parental approval of the romantic relationship.Thus, in Figure 1

Results
SPSS 17.0 software (SPSS Inc, 2007) was used for the descriptive statistics, correlations, and t-tests, and AMOS 16.0 (Arbuckle, 2007) was used to conduct a confirmatory factor analysis (CFA) to verify the unidimensionality of measures, and for the SEM/APIM analyses.

Preliminary Data Analysis
Three sets of preliminary data analyses were carried out to (a) establish that closeness to own parents was indeed higher (i.e., a stronger tie) than closeness to the know about your dating relationship?The means of the items for individuals' own mother and father, and their partner's mother and father were calculated to represent the strength of the ties.The mean score for individuals' own parents (i.e., mother and father) was 4.25, whereas that for their partner's parents was 2.88.A paired-t-test showed the difference to be significant, t (380) = 29.103,p < .001,supporting the assumption that the relationship with one's own parents is stronger than with one's partner's parents.In addition, we examined similarities between own mother's and father's approval, and between partner's mother's and father's approval, because each pair of items served as indicators of the respective strong-or weak-tie construct, and it is assumed co-indicators of the same construct are well-correlated.We found that 80.2% of individuals' ratings on their own mother's and father's approval, and 85.6% of their ratings on their partner's mother's and father's approval, matched exactly.
The second set of preliminary analyses tested the unidimensionality of the measures to be used for the APIM model.The correlations among indicators are presented in the Appendix.The results of the confirmatory factor analysis with six constructs and their indicators demonstrated a good fit, χ 2 = 89.18,df = 62, p < .05:χ 2 /df = 1.44,Comparative Fit Index (CFI) = .97,and Root Mean Square Error of Approximation (RMSEA) = .04.The models are considered to fit the data well if χ 2 /df ratio is less than 3.00, CFI is more than .90, and RMSEA is less than .10(Browne & Cudeck, 1993;Kline, 2005).
Correlations among constructs that were estimated through the CFA are reported in Table 1.All freely-estimated factor loadings were significant (female's perception of partner's negative behavior, p < .01;all the rest at p < .001)and all but two standardized factor loadings were greater than .40(.49 to .98),meaning the indicators reflected well the constructs to which they belonged.The remaining loadings, involving female and male participants' perceptions of their partner's negative behavior, were .25 and .39,respectively; these indicators were retained, in light of the model's good overall fit and significance of the factor loadings.Finally, because a series of t-tests revealed significant differences between Samples 1 and 2, the third set of preliminary analyses examined whether one overall model (the APIM model shown in Figure 1) would fit the data from Samples 1 and 2. Thus, a multiplegroup comparison using the delta chi-square test was conducted between the unconstrained (allowing parameters to vary across the two samples) and constrained (forcing parameters to be equal across the two samples) models.Coefficients were computed using Maximum Likelihood estimation.
First, the APIM model was run for both groups separately without any constraint, allowing parameters to vary across two groups (χ 2 = 309.24,df = 212, p < .001).Then, the same model was run again with equality constraints, forcing respective parameters to be equal across the two groups (χ 2 = 348.79,df = 235, p < .001).A delta chi-square test, based on Δ χ 2 (23) = 39.556,indicated some degree of harm to model fit due to constraining (with a criterion of p < .05).However, the difference in fit between the unconstrained and constrained models was not significant at more stringent levels (p < .01).In light of these somewhat ambiguous results, we opted for a single (combined-sample) model based on the principle of parsimony (i.e., it is simpler to characterize both samples with a single model than to have separate models for each group).

Main Data Analysis
Based on the findings of the preliminary data analyses, the model shown in Figure 1 was run with the pooled sample.The fit of the model was good, χ 2 = 181.33,df = 136, p < .01:χ 2 /df = 1.33,CFI = .96,and RMSEA = .04.As shown in Table 2, all freely estimated factor loadings were significant and showed a similar pattern with the CFA (note that the loadings of indicators on their respective factors may fluctuate slightly from a CFA to a full SEM that introduces directional paths between constructs).Female partners' relationship distress was significantly predicted by their own and their partner's perception of females' parents' (i.e., strong-tie) approval.The more parental approval female (β = -.37,p < .001)and male participants (β = -.19,p < .05)perceived from female's parents, the less relationship distress was reported by female couple members.
This result supports Hypothesis 1 on the importance of strong ties (i.e., females' parents apparently influencing females' distress).It also supports Hypothesis 2, in that the actor effect (female approval perception to female relationship distress) had a larger coefficient than did the partner effect (male approval perception to female distress).Female and male participants' perceptions of relationship approval from the males' parents did not predict females' relationship distress; this finding is also consistent with H1, as it shows the relative weakness of weak ties.On the other hand, males' relationship distress was predicted by their own and their partner's perception of the males' parents' approval (again, a strong-tie finding, consistent with H1).For approval from their own parents, the results for male participants showed a larger actor effect (male approval perception to male distress), β = -.38,p < .001,than partner effect (female approval perception to male distress), β = -.15,p < .05,consistent with H2.The more parental approval female and male partners perceived from males' parents, the less relationship distress experienced by males.
One result for male partners is especially interesting.Unlike the finding for female partners, males' relationship distress was also predicted by their own perception of approval from their female partner's parents, and in a direction opposite to that of other parental approval effects.The more approval male partners perceived from their partners' parents, the greater the relationship distress was for them (β = .26,p < .05).This result suggests that social capital from parents can also be detrimental to romantic relationships.
As expected, many results of control variables were significant.Relationship seriousness was positively related to approval of both sets of parents (regardless of approval reporter), and negatively related to relationship distress for both male and female couple members.Findings regarding religiosity revealed negative relations with male and female participants' relationship distress.The findings regarding age were different for male and female participants.Whereas men's age did not predict their own relationship distress, greater women's age predicted more relationship distress.

Discussion
This study examined how relationship approval from both partners' sets of parents predicted romantic-relationship distress, using social capital theory.The results show that, in support of Hypothesis 1, relationship approval from strong ties (one's own parents) more strongly predicted relationship distress (negatively) than did approval from weak ties (one's partner's parents).This result supports previous findings in the social capital literature.
Many studies have found that strong ties are especially beneficial in situations where trust is essential (Light, 1984;Light & Bonacich, 1988;Portes & Stepick, 1993;Waldinger, 1996;Zhou, 1992).The nature of romantic relationships requires and encourages trust among related parties.Thus, it is not surprising to find the same pattern in the context of romantic relationships.Findings also appeared to provide strong support for Hypothesis 2, that actor effects (i.e., associations between independent and dependent variables reported by the same person) would be more potent than partner effects (i.e., associations between variables reported by different persons).
A less clear picture emerged for the association between individuals' perception of relationship approval from their own parents and their perception of relationship approval from their partner's parents with their dating partners' relationship distress.Although the results seem to suggest that a person's relationship distress is not well predicted by approval from the partner's parents (whether perceived by the participant or the partner), one exception emerged for men in a way that suggests social capital may sometimes have negative consequences for individuals (Portes, 1998).Specifically, greater male perception of their female partners' parents' approval significantly predicted greater relationship distress in men.This finding could reflect societal norms and the role the female partners' family plays in romantic relationships.Leslie, Huston, and Johnson (1986) proposed that parents may have a greater investment in their daughters' romantic relationships than in their sons'.Parents may feel that they need to protect their daughters more than their sons.
They may also try to ensure that their daughters do not marry somebody who might prevent their child from maintaining kinship ties.Thus, women's parents' approval of the dating relationship (as perceived by the male dating partner) may be interpreted by the male not only as approval, but perhaps also as interference or as a push toward greater commitment.
Research by Milardo, Johnson, and Huston (1983) has shown that as a romantic relationship progresses into more committed stages, partners' involvement with their social networks tends to decrease and the network responds to this change with interference especially during the middle stages of increasing commitment (Johnson & Milardo, 1984).
We can speculate that interference may be stronger when the involvement and investment are high.Thus, it is possible that high levels of approval from female partners' parents accompany high levels of interference and a greater push toward commitment, and these adversely affect the relationship quality of the male partners, who are not familiar with the higher involvement between female partners and their parents.example, feeling more or less distressed with one's relationship may affect how someone perceives the relationship approval of one's own and the partner's parents.
Despite these limitations, this study has significantly contributed to our understanding of the role social capital may play with regard to dating partners' relationship distress.First, our analyses demonstrated the importance of using dyadic data and examining actor-partner effects in relationship research.By using these strategies, we were able to isolate differential effects of parental approval in many ways: strong vs. weak ties, and actor vs. partner effects.Our efforts to apply social capital theory to romantic relationships and to obtain and analyze couple data appropriately, allowed us to demonstrate that the concept of strong vs. weak ties has a place in romantic relationship research and that the theory should be explored further.
Second, our study focuses on the non-instrumental consequences of social capital.
Theoretically, the consequences of social capital can be instrumental, such as power and wealth, or expressive such as physical and mental health (Lin, 1999).To date, most studies on social capital have focused on instrumental consequences (e.g., Granovetter, 1973;Portes & Stepick, 1993;Waldinger, 1996).The current study therefore fills a gap in the literature by focusing on an emotional, non-instrumental consequence of social capital, namely, how approval from parents predicts the quality of their young-adult children's romantic relationships.
Of course, instrumental and expressive social capital from weak and strong ties may operate differently in various cultural and ethnic contexts.Whereas in our study men's and women's relationship distress appears relatively unaffected by the other dating partner's perception of their parents' relationship approval, this may not be the case in other countries (such as various Asian countries; Vaux, 1985), where a person ultimately may not just marry the partner, but may also end up marrying into the partner's family.In such circumstances, a person's relationship quality may very well depend on whether the partner's parents approve of the relationship or not.Social capital from strong and weak ties may also operate differently in marginalized couples (Lehmiller & Agnew, 2006), who ironically often receive much less social, legal, instrumental, and emotional capital, but at the same time have a greater need for it.Thus, future researchers may want to examine the concept of weak vs. strong ties in a variety of contexts.
measurement of parental approval to address four parents individually.Participants were asked to respond to the following statements on a 5-point Likert scale (1 = Not at all; 5 = A great deal), indicating the extent to which they agreed with the statements: "My mother approves of my dating relationship," "My father approves of my dating relationship," "My partner's mother approves of my dating relationship," and "My partner's father approves of my dating relationship."Items measuring individuals' perceptions of their own parents' approval were used as indicators of strong ties and those measuring perceptions of one's partner's parents' approval served as indicators of weak ties.Thus, the construct, Female Approval from Strong Ties had two indicators in all, asking about female participants' perceptions of their own mothers' and fathers' approval.The construct Female Approval from Weak Ties was represented by two indicators measuring females' perceptions of approval from her partner's mother and father.Parallel items were used for the Male Approval from Strong Ties and Male Approval from Weak Ties constructs.
, the effects from strong ties are represented in paths a and c′ for female relationship distress and paths a′ and c for male relationship distress, whereas the effects from weak ties are represented in paths b and d′ for female relationship distress and paths b′ and d for male relationship distress.The paths between individuals' own perceptions and relationship distress in our model represent actor effects (paths a and b for females and paths a′ and b′ for males), whereas the paths between individuals' partners' perception and individuals' own relationship distress represent partner effects (paths c′ and d′ for females and paths c and d for males).

Figure 1 .
Figure 1.Theoretical link between individuals' and partner's perception of approval from strong and weak ties and relationship distress.
partner's parents; (b) verify the unidimensionality of the measures to be used for the APIM model (i.e., obtaining correlations among indicators); and (c) ascertain that a single APIM model would fit the data from the two samples.With regard to the first set of preliminary analyses, four items assessed the strength of tie to the individual's own mother, own father, partner's mother, and partner's father: How close are you to your [mother, father, partner's mother, partner's father]?; How well do you know your [mother, father, partner's mother, partner's father]?; How well does your [mother, father, partner's mother, partner's father] know you?; and How much does your [mother, father, partner's mother, partner's father] Standardized coefficients of the APIM structural model are shown in Figure 2. Solid and dashed lines are interspersed simply to add visual contrast to the figure and thus aid viewing.

Table 1 .
Correlations between Latent Constructs