Conflict with Friends, Relationship Blindness, and the Pathway to Adult Disagreeableness.

The ability to form and maintain relationships with friends and romantic partners is a major developmental task for adolescents. Disagreeable youth are likely to struggle with this task, yet little is known about how they maintain their oppositional style from adolescence to adulthood. The current study examines the long-term implications of disagreeableness in a diverse sample of 164 adolescents assessed repeatedly across a 10-year period along with their friends and romantic partners. Disagreeableness at age 14-15 was assessed in observation with friends. Disagreeableness was then examined as a predictor of both future relationship quality with friends at age 16 and romantic relationships at age 21. The results indicate that although disagreeable youth do not report any relationship struggles, both their friends and romantic partners see their relationships as being low in quality. Findings suggest a developmental process by which disagreeable adolescents maintain their oppositional style through a mechanism of relationship blindness, as they simply are unable to see the relationship issues that their friends and partners clearly perceive.

The ability to interact competently within voluntary intimate relationships gains importance in adolescent friendships and ultimately culminates in successful adult romantic relationships. Some adolescents, however, do not form successful relationships and it is important to understand the underlying reasons, as they are at risk for increases in depression and other health risk variables over time (Cavanagh, Crissey, & Raley, 2008). Adolescents who are prone to disagree with others repeatedly appear to represent one of these subgroups. Disagreeable youth are not simply unpleasant or disliked individuals; rather they are both oppositional and offensive in their interactions. There is ample evidence to suggest that being disagreeable, or low on the personality construct of agreeableness, has many short-term negative correlates including a lack of peer acceptance, along with more conduct problems and depression (Scholte, van Aken, & van Lieshout, 1997). There is also evidence suggesting that even after accounting for rejection and aggression, there is a sizeable subgroup of disagreeable youth that exhibit extensive adjustment problems (Laursen, Hafen, Rubin, Booth-LaForce, & Rose-Krasnor, 2010). Given this pattern of findings, it is surprising that there is a lack of research investigating the relationship profiles and developmental pattern of disagreeable youth as they move through adolescence and into adulthood.

During adolescence, teenagers become increasingly reliant on relationships formed outside of the family unit. These relationships, particularly friendships, differ from family relationships in many respects, but in particular because they involve choice. Adolescents are likely to develop patterns of interaction within these voluntary relationships that are likely to carry forward into their future relationships. Though, there is only a small amount of evidence that the quality of adolescent friendships is related to both concurrent and future romantic relationship quality (Connolly, Furman, Konarski, 2000). This evidence is based on the view that in voluntary relationships, there may be a working model that an individual carries forward with them from relationship to relationship which drives both their choice of future relationships and their interaction-style within those relationships (Collins, Welsh, & Furman, 2009).

Although there is a lack of strong empirical evidence extending findings about adolescent relationship patterns into adulthood, there is some evidence to suggest that personality and relationship patterns begin to stabilize as individuals move from adolescence to adulthood (Donnelan, Conger, Burzette, 2007). The case of disagreeable youth offers a prime opportunity to study this potential development. One might expect that disagreeable youth would receive negative feedback about their behavior and alter it accordingly. However, there is evidence to suggest that some individuals who are oppositional in nature do not pick up on the relational cues in a typical manner.

A classic study by Kobak and Sceery (1988) found that dismissing first-year college students were rated by their peers as more hostile, however their own self-reports of hostility did not differ from those of secure individuals. Discrepancies of this nature have been described as a self-protective or compulsive self-reliance mechanism whereby feelings of inadequacy are masked and avoided by an inflated perception of self-competence and functioning (Diener & Milich, 1997). Essentially, these individuals learn to overlook the ways their behavior is perceived by others. Given that individuals know it is socially frowned upon to be consistently oppositional and may cause interpersonal problems, highly oppositional individuals behavior may be maintained via a similar mechanism which allows them to continue to in an obliviously offensive manner, without ever acknowledging, self-correcting feedback from others. This mechanism may be the key to understanding the course of relationships for disagreeable youth, as their behavior of acting obliviously oppositional means that it is almost impossible for them to self-correct or learn from problematic relationships.

The most unbiased source of information about the enduring patterns an individual displays in their interactions within close relationships is observation. Self-report measures are useful for capturing how an individual perceives themselves and their environment, but these perceptions can at times be misleading and as such are best used in concert with multiple methods. Research has consistently identified and conceptualized agreeableness only in terms of self-report measures (e.g. De Pauw & Mervielde, 2010), which has been useful for identifying the array of negative correlates of being disagreeable such as higher externalizing and internalizing problems and lower self-worth (Laursen et al., 2010). However, self-perceptions may not be ideal for understanding how disagreeable youth interact within close relationships, as these youth in many instances might not even know that they are acting in a disagreeable manner. Observations are useful for addressing this shortcoming and have been used to understand the enduring qualities of interaction-styles, particularly when trying to understand unique profiles across individuals. Further, understanding the interactional style in adolescence that lead to an enduring disagreeable personality-type has tremendous usefulness.

The current study utilized observational, multi-reporter data collected over a 10-year span to identify and track the development and relationships of disagreeable youth (see Figure 1 for conceptual overview). Observations of target youth and their friends at age 14 and 15 were used to assess early adolescent disagreeableness in terms of rudeness, lack of cooperation, and forcefulness. In order to track the relationship blindness of these disagreeable youth, reports from friends in middle adolescence and from romantic partners in emerging adulthood were collected. It was hypothesized that disagreeable youth would have more conflictual and poorer quality future friendships in adolescence, and that this would be evidenced by an element of relationship blindness in that their friends would report a poorer quality relationship but they themselves would not (Hypothesis 1). Second, it was hypothesized that this pattern would continue into target youth’s romantic relationships in emerging adulthood, such that their romantic partner’s would report a more negative relationship but the disagreeable youth would not (Hypothesis 2). Finally, to confirm that the developmental process outlined above truly results in the formation of a disagreeable adult, it was predicted that the disagreeable construct created from observations of target youth at age 14 and 15 would predict target youth’s selfreport of disagreeableness at age 25 (Hypothesis 3).

Figure 1

Conceptual Model for Proposed Development of Disagreeableness from Early Adolescence to Adulthood

Note. Disagreeable interactional style at age 14–15 was captured via observations of target youth and their best friends. Relationship blindness in mid-adolescence and emerging adulthood was captured by separately measuring self-report and partner-report.

Method

Participants

Participants included 164 (87 females and 77 males) teenagers along with their best friends (n = 156) and romantic partners (n = 111). This sample was drawn from a larger longitudinal investigation in the Southeastern United States of adolescent social development in familial and peer contexts. Students were recruited via an initial mailing to all parents of students in the school along with follow-up contact efforts at school lunches. The sample was racially/ethnically and socioeconomically diverse: 58% identified themselves as Caucasian, 29% as African American, 8% as mixed race/ethnicity, and 5% as being from other minority groups. As reported by the adolescents’ parents, the median family income was in the $40,000 – $59,999 range.

Target adolescents were assessed initially over a two-year period in mid-adolescence, at age 14 (M age = 14.27, SD = 0.62) and at age 15 (M age = 15.22, SD = 0.81). At each of these assessments, target adolescents were asked to nominate their closest friend at that time to take part in the study. Close friends were described as “people you know well, spend time with, and whom you talk to about things that happen in your life.” The targets and their close friends were asked to participate in a video-taped interaction in the lab setting and received compensation for their participation. The target individuals again nominated their closest friends one year later at an age 16 assessment (M age = 16.30; SD = 0.86), at which time they and their closest friends completed questionnaires. These target individuals were assessed again in early adulthood between ages 20–22 (M age = 20.99; SD = 1.10), this time with their romantic partners (M age = 19.06; SD = 3.10) of at least 3 months duration (M duration = 14.40 months, SD = 13.31).

Measures

Disagreeableness with close friend

As described above, target adolescents indicated their best friend at both age 14 and age 15. At ages 14 and 15, each target adolescent-close friend dyad took part in an 8-min videotaped interaction in which they were presented with a revealed differences task (Strodtbeck, 1951). This disagreement task involved a hypothetical dilemma asking target teens and their friends to come to a consensus decision. At age 14, the dyad was asked to decide which 7 out of 12 fictional characters, with specific characteristics and ages, who are dying from a flu-like disease should be given the only 7 doses of the cure. At age 15, the dyad was asked to decide which 4 out of a possible 10 fictional individuals, with specific characteristics and skills, should be the first ones kicked off a deserted island as part of a “survivor” contest. For this study, participants’ scores for frequency and intensity of collaborating versus arguing, forcefulness/pressuring the peer to agree, and rudeness towards their friend were calculated. Scores were rated on a 0 to 4 scale, with higher scores indicating greater frequency and intensity of behaviors that are disagreeable and hinder free discussion. Each interaction was reliably coded by two trained raters blind to other data from the study (ICCs from .64 to .69, considered in the good range for this statistic. These three ratings were averaged across the two assessments. The construct of disagreeableness was operationalized as a latent variable with these three indicators (see Figure 1).

Close friendship competence

Both target adolescents and their close friends reported on the target adolescent’s competence in close friendships at age 16. Target adolescents completed the close friendship score from the Harter Self-Perception Profile for Adolescents (Harter, 1988). The format asks the reporter to choose between two contrasting descriptors and then rate the extent to which their choice is sort of true or really true about the target adolescent (e.g. “Some people are able to make really close friends/some people find it hard to make really close friends”). Item responses are scored on a 4-point scale and then summed, with higher scores indicating higher levels of peer-rated close friendship competence. The friendship competence subscale showed good internal consistency (α’s = .82 – .83).

Conflict and betrayal in friendships

Both target adolescents and their close friends completed the conflict and betrayal scale from the Friendship Quality Questionnaire (Parker & Asher, 1993) at age 16. The scale included seven items such as, “We argue a lot” and “S/he sometimes says mean things about me to other kids”. Item responses were reported on a five-point scale (1 = not at all true, 5 = very true) with higher scores indicating more betrayal and conflict within the relationship. In our sample, the internal consistency was good (α’s = .86 – .88).

Negative relationship interaction in early adult romantic relationships

Targets and their romantic partners each completed items about the negative interactions in their relationship using the Network of Relationships Inventory (NRI; Furman & Buhrmester, 1985). The scale included 6 items (e.g., “How much do you and this person disagree or quarrel?”). Each item was rated on a 5-point scale with higher scores indicating more negative interactions. The scale had high reliability (α’s = .82 – .84).

Adult Disagreeableness

Target youth completed self-report assessments of their own agreeableness at age 25 using the NEO-FFI (McCrae & Costa, 2010). The scale consists of 12 items rated on a scale from 1 (Strongly Disagree) to 5 (Strongly Agree). Agreeableness is typically assessed to capture compassion and cooperativeness on the higher end and antagonism on the lower end (e.g. I am not really interested in others). Since we are particularly interested in capturing disagreeableness, we averaged the total agreeableness score and then reverse scored the mean to create a disagreeableness score. The scale had high reliability (α= .78).

Procedure

Participating adolescents provided informed assent, and their parents provided informed consent until adolescents were 18 years of age, at which point they provided informed consent. The same assent/consent procedures were used for peers/romantic partners and their parents. Adolescents, close friends, and romantic partners were compensated for their participation. Participants’ data were protected by a Confidentiality Certificate issued by the U.S. Department of Health and Human Services, which further protects information from subpoena by federal, state, and local courts. If necessary, transportation and childcare were provided to participants.

Missing Data

Of the 164 individuals who participated at ages 14 and 15, 71% (n = 117) reported having a romantic partner and filled out reports about their relationship in early adulthood. Data was available from 95% (n = 111) of the identified romantic partners. A series of attrition analyses compared those individuals who did vs. did not participate in this portion of the study on all available demographic variables (gender, ethnicity, parental income) and all measures concerning their relationships with friends in adolescence. The results of these t-tests and chisquare analyses indicated that there were no significant differences on mid-adolescent study variables. Given that the amount of missingness was not related to demographic or other study variables, we used full information maximum likelihood (FIML) as the method to estimate our structural parameters in MPlus 6.12 (Muthén & Muthén, 2010).

Results

Overview of the Analyses

Disagreeableness was operationalized in early adolescence as a latent variable (see Figure 1), which included the observed collaboration, forcefulness, and rudeness of target individuals in a disagreement task with a friend at age 14 and age 15. As expected, the variables loaded together strongly, with collaboration loading negatively (β = −.78), and forcefulness (β = .69) and rudeness (β = .67) both loading positively. This latent variable was used in all subsequent analyses. Gender and income were included as covariates in all analyses and all interactions involving gender and income were tested though no interactions were significant.

Testing the Development of the Disagreeable Adult

To determine whether disagreeable youth are characterized by a distinct set of friendship difficulties in middle adolescence, a series of regressions in which the disagreeable latent construct was entered as a predictor of the target individual’s self-report and their friend’s report of the closeness and conflict within their friendship at age 16.

To determine whether disagreeable youth are characterized by a relationship pattern that carries over into emerging adult romantic relationships, a series of regressions was examined in which disagreeableness at ages 14–15 was entered as a predictor of the quality of romantic relationships in early adulthood (ages 20–22). Self-reports and romantic partner reports of negative relationship interactions were the dependent variables in these analyses.

Hypothesis 1: Disagreeable youth identified in early adolescence will have more conflictual and poorer quality friendships in middle adolescence

Table 2 summarizes the results of the regression analyses predicting age 16 friendship closeness and conflict from age 14–15 disagreeableness. Both target individuals and their friend reported on the closeness and conflict within their relationship. There were no interactions involving gender or income.

Table 2

Disagreeableness at Age 14–15 Predicting Friendship Quality at Age 16

	Age 16 Close Friendship Score				Age 16 Conflict and Betrayal Score
	Self-Report		Friend-Report		Self-Report		Friend-Report
	β	CI	β	CI	β	CI	β	CI
Step I
Gender (male = 1, female = 2)	−.12	[−.29, .06]	−.14	[−.30, .02]	−.24**	[−.45, −.04]	−.05	[−.19, .09]
Family Income	.09	[−.08, .26]	.10	[−.05, .25]	.01	[−.11, .14]	−.04	[−.14, .06]
Step II
Age 14–15 Disagreeableness	.03	[−.11, .14]	−.21*	[−.39, −.05]	.19*	[.04, .33]	.34**	[.09, .59]

Note.

p < .01.

p < .05.

Age 16 Closeness

As hypothesized, disagreeableness did not predict self-reports of friendship closeness at age 16 (β = .03); however, it did predict friend-reports of closeness at age 16 (β = −.21), such that the friends of disagreeable youth reported having a less close friendship with those youth than did friends of others. These coefficients were significantly different, χ2(1) = 5.26, p = .02, suggesting that disagreeableness predicted future friend-reports more strongly than future self-reports of closeness.

Age 16 Conflict

Disagreeableness predicted both self-report of friendship conflict (β = .19) and friend-reports of friendship conflict (β = .34) at age 16. Both disagreeable youth and their friends reported having more conflict within their friendship than did other youth. Further, the difference between these paths was not significantly different, χ2(1) = 3.33, p = .07.

Hypothesis 2: The patterns exhibited by disagreeable youth in adolescent friendships will carry over into their romantic relationships in emerging adulthood

Negative Relationship Interaction

Table 3 presents results predicting self-report and romantic partner report of negative relationship interaction in emerging adulthood (age 21) from age 14–15 disagreeableness. Again, disagreeableness did not predict self-report of negative relationship interaction (β = .11). However, disagreeableness did predict romantic partner report of negative relationship interaction (β = .27), such that romantic partners of disagreeable youth reported having more negative interactions with targets than did romantic partners of other youth. These coefficients were significantly different, χ2(1) = 4.26, p = .04, suggesting that disagreeableness predicted future romantic partner-reports more strongly than future self-reports of negative relationship interaction.

Table 3

Disagreeableness at Age 14–15 Predicting Romantic Relationship Negativity in Emerging Adulthood

	Age 20–22 Romantic Relationship Negativity
	Self-Report		RP Report
	β	CI	β	CI
Step I
Gender (male = 1, female = 2)	−.11	[−.29, .08]	−.30**	[−.44, −.16]
Family Income	−.05	[−.19, .10]	−.08	[−.20, .04]
Step II
Age 14–15 Disagreeableness	.11	[−.03, .20]	.27**	[.11, .44]

Note.

p < .01.

p < .05.

RP = romantic partner.

Hypothesis 3: Being disagreeable in observations with friends in early adolescence will predict identifying as disagreeable in young adulthood

Disagreeableness in early adolescence, as captured by observations of interactions with friends, significantly predicted self-report of disagreeableness in young adulthood (β = .38, p < .01). Those individuals who were disagreeable in their interactions with friends at age 14–15 reported being more disagreeable at age 25.[1]

Discussion

Disagreeable youth are not just unpleasant; they are contentious and offensive in their relationships. This study confirmed hypotheses that individuals who are disagreeable have relationship partners who report poorer quality relationships and more conflict than relationship partners of individuals who are not disagreeable. Perhaps even more importantly, we found that disagreeable youth view and report their relationships differently than their friends and romantic partners. This raises the possibility that the enduring pattern of negative quality across relationships of disagreeable youth is maintained as a result of a perceptual bias that these individuals hold concerning their relationships. While disagreeable youth are aware of the conflict in their relationships, they do not report the relationship to be negative overall, although their partners (both friends and romantic partners) repeatedly do. An unfortunate potential consequence of this discounting may be that it keeps these youth from recognizing and potentially improving their relational style. The findings from this study highlight the breadth and duration of the implications of this disagreeable interaction style, as it is evident in future adolescent friendships and also carries over into adult romantic relationships six years later. These results provide empirical evidence for the theory that there may be an internal mechanism that accounts for stability in relationship difficulty for certain youth (Collins, Welsh, & Furman, 2009).

One of the striking features of this study is the consistency with which disagreeable youth view their relationships distinctly differently than do their friends and romantic partners. We had predicted that disagreeable youth may exhibit a pattern similar to the self-protective mechanism that others had uncovered in preoccupied adults (Kobak & Sceery, 1988). The pattern of relationship blindness described in the current study is particularly troubling because it suggests that disagreeable youth will struggle to form healthy relationship interactions, as they may not work to improve upon their maladaptive interaction style. Adopting this relationship blindness approach to social relationships dramatically reduces their opportunity to alter their behavior, potentially condemning them to the long-term pattern of negative interactions observed in this study.

It follows that a disagreeable friend would be a disagreeable romantic partner. A key finding of the current study, however, suggests that those prone to disagreeableness in adolescence not only continue to be disagreeable in their early adult romantic relationships, but that they continue to find relationship partners. There is a wealth of literature suggesting that similarity is a key factor in the choice and maintenance of voluntary relationships throughout adolescence and early adulthood (Collins, Welsh, & Furman, 2009). Disagreeable individuals may find interactions with other partners prone to conflict and an argumentative style because they find it rewarding and reinforcing of their own behavior. It is possible that this leads individuals prone to disagreeableness limited in their partner choices in early adulthood, thus only finding enduring relationships with other young adults who have similar interaction styles. This would only serve to further their unhealthy relationship expectations.

Individuals with a disagreeable interaction style are not just unpleasant; they also carry their elevated level of contentiousness in their interactions with them from relationship to relationship. The fact that this behavior is at least in part maintained by the inability to acknowledge the resulting discomfort this creates is noteworthy. It suggests developmental processes by which youth that have an oppositional style in early adolescence use conflict within a relationship to justify their disagreeable nature. This developmental process is also indicated in the literature on trait aggressiveness, whereby individuals who are prone to aggressive perceptions actively contribute to creating consistently aggressive environments and responses (Anderson, Buckley, & Carnagey, 2008; Dill, Anderson, Anderson, & Deuser, 1997). Understanding the nature of forming and maintaining a disagreeable style of interaction is an underappreciated developmental phenomenon worthy of future study (Laursen & Richmond, 2014).

Although strong in many respects, a few limitations are worthy of note. First, although the use of longitudinal data is sufficient to refute causal hypotheses, they cannot directly support causal claims. Second, the analyses presented were spread out across several years of development. While this is a strength in that it suggests an enduring continuity from disagreeableness to friendship quality to romantic relationship quality, we can only speculate about the mechanisms or incremental steps that lead the disagreeable individuals to consistently form and maintain poor quality relationships. Third, though we specify the differences between individuals and their relationship partners as reflecting “relationship blindness”, we would need to directly model this blindness over time to understand its role and function. Future work with larger samples would help add clarity to these novel findings. Finally, the longitudinal and multiple informant nature of the data made it logistically impractical to follow a large sample, and thus there was relatively little power to detect potential moderators. Thus, while gender of the target youth was included in all analyses as a moderator, the lack of results should not be interpreted as a concrete statement of null effects.

The findings of this study are noteworthy in many respects. We found that an individual’s interaction style with friends in adolescence is directly related to their interaction style years later in romantic relationships. In particular, disagreeable youth form unhealthy friendships in adolescence and continue to have unhealthy relationships in adulthood that seem to stem from their interaction style. This pattern may be related to the observed tendency for disagreeable youth to be obliviously offensive in their interactions, acting aggressively but not noticing its negative impact on their relationships. As such, if these findings are confirmed in further research, they suggest that disagreeable youth might benefit greatly from interventions designed to improve social interactions and, particularly, social awareness. Without intervention, these individuals are likely to experience conflictual, relationally aggressive, and unhealthy relationships throughout development as they consistently ignore or rationalize their unhealthy patterns.

Table 1

Correlations Among Variables in the Study

Variable	1	2	3	4	5	6	7	8	9	10	11
1) Family Income	-
2) Sex (1 = Male, 2 = Female)	.01	-
Age 14–15 Observed Disagreeableness
3) Collaborating vs. Arguing	.03	.07	-
4) Pressuring Peer to Agree	−.02	−.02	−.48*	-
5) Rudeness Toward Friend	−.04	−.06	−.45*	.43*	-
Age 16 Friendship
6) Close Friendship (Self-Report)	.03	.05	.06	.04	.04	-
7) Close Friendship (Friend-Report)	.05	.08	.11	−.15	−.17*	.24*	-
8) Conflict (Self-Report)	−.07	−.03	−.17*	.10	.16*	−.23*	−.33*	-
9) Conflict (Friend-Report)	−.08	−.05	−.15	.28*	.28*	−.12	−.37*	.29*	-
Age 20–22 Romantic Relationship
10) Negativity (Self-Report)	.02	.05	.04	.02	.15	−.17	−.13	.19*	.11	-
11) Negativity (Partner-Report)	−.01	.04	.13	.25*	.19*	.00	−.02	.28*	.13	−.20*	-
12) Age 25 Disagreeableness	.02	−.07	−.14	.18*	.20*	−.08	−.15	.11	.18*	.09	.15

Note.

p < .05.

p < .01.