Romantic Relationships in Emerging Adulthood: Perception-Partner Ideal Discrepancies, Attributions, and Expectations

This study aimed to examine 1) emerging adults’ attributions for their subjectively perceived current romantic relationship as good or bad; 2) the role of their perception-partner ideal discrepancies in estimating their current romantic relationship as good or bad, and in the formation of attributions; and 3) the effects of all the three concepts on the generation of the expectations for the quality of the romantic relationship in the future. Undergraduate and postgraduate university students, from various faculties, ages of 18 to 25, both genders, participated in this study. The participants filled in, first, the scale of the romantic partner ideals, and, after one week, the scales of perceived of the quality of the current romantic relationship, attributions and expectations. The results regarding attributions revealed that 1) the perceived good romantic relationships were mainly attributed to internal and self-partner internal and controllable factors (love, passion, effective communication, honesty-loyalty, understanding each-other), whereas the perceived bad romantic relationships were mainly attributed to personal uncontrollable, external, unstable and partner’s internal factors (untrustworthiness, lack of passion, non honesty-loyalty, lack of love, and ineffective communication); and 2) locus of causality, followed by personal controllability and stability, was the most powerful attributional dimension in discriminating the two groups of the emerging adults. The findings regarding the perception-partner ideal discrepancy showed that 1) the students who had smaller perception-partner ideal discrepancy, compared to students who had larger perception-partner ideal discrepancy, estimated their current romantic relationships more favourable but only trustworthiness accounted for unique variance in it; and 2) perception-partner ideal discrepancy in trustworthiness, followed by warmth/intimacy, and attractiveness/vitality, was the most powerful factor in discriminating the two groups of the participants. The results with respect to effects of partner ideal discrepancy on attributions indicated that 1) the students’ perception-partner ideal discrepancies proved to be a significant and positive factor in the formulation of the attributional dimensions (expect for external controllability), mainly both stability and locus of causality, for the perceived current quality of their romantic relationship; and 2) only trustworthiness and warmth/intimacy had unique effects on locus of causality, and only trustworthiness uniquely contributed into stability, partner’s locus of causality, and self-partner interactive locus of causality. Finally, the three concepts, as a group, positively influenced the generation of expectations of the quality of the romantic relationship in the future but no one of the components of the perception-partner ideal discrepancy was unique contributor, while stability, self-partner interactive controllability, and, particularly, the perceived quality of the relationship accounted for unique variance in relationship expectations.


Introduction
Attaining love and romance ensures happiness and longevity and contributes to satisfy the need for positive long-term social bonds (Berscheid & Ammazzalorso, 2003;Carr, 2005;Fehr, 2003;Forgas & Smith, 2005;Hendrick & Hendrick, 2005;Rusbult, Arriaga, & Agnew, 2003).On the other hand, forming and maintaining a successful romantic relationship is a challenge as well as finding a desirable partner or any partner at all may be confronted with difficulties (Clark & Beck, 2010;Rusbult et al., 2003).Also, even finding a desirable partner, the things can go wrong, and happiness may disappear (Maner & Miler, 2010).These topics are particularly troublesome among emerging adults due to the fact that, among others, they have little experience with forming and maintaining a successful long-term relationship (Arnett, 2000;Fincham & Cui, 2010;Clark & Beck, 2010;Maner & Miler, 2010;Stanley, Rhoades, & Fincham, 2010).
Emerging adulthood has been defined as a specific stage of life between the ages of 18 and 25, during which individuals are neither adolescents nor adults with full responsibility for themselves and for others, and during which they evaluate possible romantic relationships, estimate world views, and seek and are preparing for future careers (Arnett, 2000(Arnett, , 2004;;Collins & Sroufe, 1999).Indeed, this is a distinct stage of life, characterizing "by the process of initiating goal pursuits and deliberating about goals regarding partners, world views, and work" (Clark & Beck, 2010: p. 190).However, as Clark and Beck (2010) emphasize, little literature has been devoted to "the intra-and interpersonal psychological processes involved in the tasks of this period" (p.190), while it has been mainly related to age, tasks, and a lack of commitments.
While romantic love is a complex phenomenon, there is a remarkable increasing research shift in defining and explaining it (Fletcher, 2002;Fitness, Fletcher, & Overall, 2005).Romantic love has been conceptualized as attachment, caregiving and sexual attraction, and it is crucial in the development of intimate relationships (see Adams, 1986;Fitness et al., 2005;Shaver, Morgan, & Wu, 1996).
One the other hand, people has beliefs and ideal standards about romantic partners and relationships through which they cognitively evaluate their partners and relationships (Higgins, 1987(Higgins, , 1989;;Regan, 1998;Rusbult, Onizuka, & Lipkus, 1993;Rusbult et al., 2003).More precisely, individuals have certain beliefs about the content and meaning of love, and the factors that cause relationship to fail or succeed (Fehr, 1999(Fehr, , 2003;;Fletcher & Kininmonth, 1992).For example, partners know whether they are in a good relationship by comparing their perceptions of their current relationship with their pre-existing ideals about a good relationship, and the more congruent the fit, the happier they are (for a review see Simpson, Fletcher, & Campbell, 2003).Furthermore, according to Simpson et al.'s (2003) ideals standard model in close relationships, which was originally developed by Fletcher, Simpson, Thomas, and Giles (1999), the relationship-based knowledge involve the components of the self, the partner, and the relationship that are likely to overlap, and have the functions of evaluation, explanation, and regulation.The partner and relationship ideals, in particular, "operate as chronically accessible knowledge structures that probably predate-and may casually influenceimportant judgments" (Simpson et al., 2003: p. 89), such as the perceived causes of relationship success (Fletcher & Kininmonth, 1992) and what is a "good" relationship (Hassebrauck, 1997), and decisions in relationships (see Fletcher et al., 1999).
People also have expectations about the partner, the relationship and the self in general that are based on previous experience, and influence their judgments, emotions and behavior in the current romantic relationship (Simpson et al., 2003;Simpson & Rhodes, 1998;Stephanou & Kyridis, 2011).These expectations are partly constructed through the cognitive appraisals the partners make in the relationship (Forgas & Smith, 2005;Trope & Gaunt, 2005).Furthermore, the dispositional attributions an individual makes to the partner reflect expectations of how he/she will behave in various situations (see Berscheid & Ammazzalorso, 2003;Karney et al., 2003;Trope & Gaunt, 2005).
This study, aiming to fill a research gap, focused on perception partner-ideal discrepancy, attributions and expectations in romantic relationships in emerging adult university students.This study was mainly based on Weiner's attribution theory, and on Simpson et al.'s (2003) ideals standard model in close relationships.

Romantic Relationship Appraisals and Expectation
People appraise a relationship by evaluating and by attributeing causes (Leary, 2000;Smith & Lazarus, 1990;Trope & Gaunt, 2005).The appraisals reflect what the stimulus-relationship means to them and whether it is good or bad (Fincham, 2003;Fitness et al., 2005).
People spontaneously attribute causes by the way they talk and think about their relationships (Fletcher & Fincham, 1991).Although, an intimate interpersonal relationship could be attributed to infinite number of attributions, self, other person, situation, environment, self-other person interaction, and relationship itself are the most prominent causes in describing positive and negative relationships (Argyle, 2001;Erber & Gilmour, 1995;Planalp & Rivers, 1996).However, the content of causes is not the crucial, as the location of the causes on attributional dimensions.Specifically, attributions are categorized into causal dimensions of locus of causality (internal/external to the person), stability (stable/unstable over time) and controllability (personal and external controllable/uncontrollable), which have psychological and behavioral consequences (Argyle, 2001;Berscheid & Ammazzalorso, 2003;Fletcher & Thomas, 2000;McAuley, Duncan, & Russell, 1992;Stephanou, 2007, in press;Weiner, 2002Weiner, , 2005)).
Previous researches indicate that the perceived quality of the relationship differentiates the attributional pattern (Fiedler, Semin, Finkenauer, & Berkel, 1995;Fincham, 2003).For example, there is strong correlation between the positivity of causal attributions and relationship satisfaction (Fletcher & Thomas, 2000).Furthermore, attributions maintain the preexisting relationship satisfaction regardless of the partner's behavior (see Fitness et al., 2005).Satisfied people with their relationships, tend to make charitable attributions for their partners negative behaviour as well as they make optimistic attributions for communication difficulties.Also, happy couples attribute their partners' positive behaviour to dispositional rather than situational factors, and they focus their disagreement on a specific issue, rather than globally criticise their partners (Carr, 2000;Gottman, 1993;Harvey et al., 2005).Yet, the longitudinal association appears to be mediated by the effects of attributions on efficacy expectations which, in turn, influence satisfaction (Fincham, Harold, & Gano-Phillips, 2000).In contrast, the more negative the interpersonal relationship the more the attributions to the other person's constant negative properties (Argyle, 2001;Gilbert & Malone, 1995;Hewstone & Antaki, 2001).
Concerning the effects of attributions on relationship expectations, the literature documents that the nature of attributions significantly influences how the relationship satisfaction changes or remain constant over time (Bradbury & Fincham, 1990;Harvey, 1987).For example, attributing successful romantic relationship to internal and stable factors enhances relationship expectations, and facilitates relationship engagement, while attributing negative romantic relationship to unstable is likely to improve the relationship and minimizes the feeling of hopelessness (see Fincham, 2003;Karney et al., 2003;Stephanou & Kyridis, 2011).In contrast, attributing negative relationship to internal, and particularly, stable factors reduces positive expectations, produces the feeling of hopelessness and can lead to learned helplessness, a sense that none effort can lead to good romantic relationship (see Fitness et al., 2005;Peterson & Steen, 2005;Seligman, 2002;Weiner, 2001).Also, partners who are satisfied with their relationships predict relationship longevity (see Fitness et al., 2005).

Association of Perceived Partner-Ideal Discrepancy with Romantic Relationship Appraisals and Expectation
According to Simpson et al.'s (2003) model of ideal standards model, the relationship and partner ideals have three functions: evaluation, explanation, and regulation.More precisely, a series of studies performed by Fletcher, Simpson and their colleague indicate that people compare their current partner or romantic relationship with their pre-existing standards and ideals about what is a good or a bad relationship or partner in order to estimate their partner or relationship as good or bad (see Flecher et al., 1999;Simpson et al., 2003).The finding is that the higher the matching, the happier they are with their relationships, and the higher the predictions of staying with their partners.In contrast, the larger actual-ideal discrepancies the more intense the dejection-related emotions, such as dissatisfaction, disappointment and shame, they experience, and the higher the changes of breaking up the relationship.
Individuals also use this comparison to explain and understand their relationship.For example, the partners by making attributions and giving causal accounts explain relationship satisfaction or problems (Fletcher & Kininmonth, 1992;Hassebrauck, 1997).Furthermore, the perceived causes of a relationship success, compared to global relationship concepts like love, is more similar to ideal standards (Simpson et al., 2003).
The flexibility of ideal standards affects the three functions in the romantic relationship.For example, men and women exhibit more flexibility in evaluating short-term relationship than long-term relationship (Kenrick, Groth, Trost, & Sadalla, 1993;Regan, 1998).It is reasonable to hypothesize that perception partner ideal discrepancy will account strongly in atributions about their long-term romantic relationships.
As above mentioned, the findings also document the ideal standards in close relationships comprise three interrelating components, the self, the partner, and the relationship (see Baldwin, 1992;Fletcher & Thomas, 1996).Ideal partner and ideal relationship overlap, since individuals prefer ideal partners who could contribute in forming their ideal relationships (see Fletcher et al., 1999).
Previous researches (e.g., Campbell, Simpson, & Orina, 1999;Gangestad & Simpson, 2000) suggest that people judge ideal partner on three basic dimensions: Warmth/ trustworthiness (intimacy, warmth, trust, and loyalty), vitality/attractiveness (attractive, energetic, and healthy), and status/resources (social status and resources).Researches on mate selection and mating strategies support that the three dimensions are related to successful mating (Buss & Schmit, 1993;Gangestad & Simpson, 2000;Simpson & Gangestad, 1992).For example, by focusing on intimacy and commitment, an individual may be more likely to acquire a cooperative and committed partner.The relationships ideals consist of two dimensions: intimacy/loyalty (intimacy, loyalty, stability-corresponding to partner intimacy dimension), and relationship passion (not clearly paralleling the partner other dimensions) (see Simpson et al., 2003).Little research has examined the content of ideal standards across various ages, while the list of the components does not end.
It could be hypothesized that holding high partner ideal standards for intimacy and trust may reflect a preference for relationship intimacy and commitment.In the case of confirmation of the preference, the relationship should be considered successful, and an optimistic attributional pattern will explain it, whereas, in the case of non confirmation of the preference, the relationship will probably be estimated as unsuccessful, and the attributional pattern may reflect self-serving (see Steele, 1988;Taylor, 1989).

Aim and Hypotheses of the Study
This study aimed to examine 1) university students' attributions for their subjectively perceived current romantic relationship as good or bad; 2) the role of the perception-partner ideal discrepancy in the estimation of the current romantic relationship as good or bad, and in the formulation of the attributions; and 3) the effects of the perception-partner ideal discrepancies, the perceived current quality of the romantic relationship, and the attributions for the romantic relationship on the formulation of the romantic relationship expectations.
The hypotheses of the study were the following: The participants will attribute the perceived quality of their romantic relationships to various factors but among them partner, self, partner-self interaction and environment will be dominant (Hypothesis 1a).The perceived good or bad romantic relationship will be attributed to self-and others-related factors, respectively (Hypothesis 1b).Locus of causality will be the most powerful dimension in discriminating the group of students with the perceived good romantic relationships from the group of students with the perceived bad romantic relationships (Hypothesis 1c).The perception-partner ideal discrepancies will positively influence the perception of the quality of the current romantic relationship, and the subsequent attributions (mainly, stability) (Hypothesis 2a).While the perception-partner ideal discrepancy will be relevant to warmth, intimacy, trustworthiness, loyalty, and attractiveness and passion, their prevalence will differ in the formation of the perceived quality of the current romantic relationship and of the attributions (Hypothesis 2b).All the three sets of the concepts, the perception-partner ideal discrepancy, the perceived quality of the romantic relationship, and the attributions (mainly, stability), will account for unique variance in relationship expectations (Hypothesis 3).

Method Participants
The participants in this study were 176 male and 210 female undergraduate and postgraduate students.Their age ranged from 18 to 25 years (M = 21.30,SD = 2.93).They came from six Greek Universities, representing various Faculties of studies.The students had experienced 2 breakups on average, and the duration of their current romantic relationship ranged from six months to three and half years.Of the participants, 254 and 132 students perceived their current romantic relationship as positive and negative, respectively (see below).

Measurements
Perceived of the current romantic relationship good/ bad.Students' perceptions of the quality of their current romantic relationships were examined by responding to a sevenpoint four items scale (e.g., "How good is your romantic relationship?","How much satisfied are you with your romantic relationship?").Responses ranged from 1 = not at all to 7 = very much.The consistency of the scale was based on previous relevant to the topic literature (see Fincham, 2003;Fletcher, 2002;Schumm, Paff-Bergen, Hatch, Obiorah, Copeland, Meens, & Bagaighis, 1986;Stephanou, 2005Stephanou, , 2007, in press), in press).
The participants themselves defined their current romantic relationship as good or bad by filling in the romantic relationship scale twice.Specifically, they, first, filled it for the current quality of their romantic relationship, and, then, mentioned the lowest value in each item over which their romantic relation-ship would be good.The participants whom the mean score was lower than the indicated as good relationship formed the group of good romantic relationships, while those whom the mean score of their romantic relationships was equal or higher than the indicated one formed the group of bad romantic relationships.Cronbach's alphas were .88 and .79 for the positive and negative romantic relationship, respectively.scale, ranging from .70 for attractiveness/vitality thought .72 for warmth/intimacy to .79 for trustworthiness.
The perception-partner ideal discrepancy is the magnitude of the differences between current partner and ideal partner in attractiveness/vitality, warmth/intimacy, and trustworthiness.
Romantic relationship expectations.Partners' romantic relationship expectations were examined via the scale of the current romantic relationship.The wording of the questions for the two scales was the same except for verb tense (e.g., "How good will be your romantic relationship in the future?","How much satisfied will you be with your romantic relationship in the future?").Cronbach's alphas were .84 and .78for the positive and romantic relationship, respectively.
Personal factors.Personal information scale comprised of a set of questions relevant to personal factors, such as age, gender, type of studies and relationship experience.

Procedure
All of the participants, first, filled in the scale of the romantic partner ideals, and, after one week, completed the rest of the scales.The students completed the scales at their departments, in quite places, in front of the researchers' co-workers, who were female and male postgraduate students and had trained for this particular part of the research by the investigator.
To match the questionnaires that were responded by the same student, students were asked to choose a code name and use it on the response sheets.The students were assured of anonymity and confidentiality.

Attributions for the Perceived Positive and Negative Current Romantic Relationships
Examination of the university students' open-ended responses to the attribution scale with respect to most dominant cause of the given current romantic relationship revealed that these could be categorized into the general categories presented in Table 1.The reliability of this coding scheme was examined by asking two familiar with attribution theory judges to place each of the open-ended responses into one of the proposed categories.There was a total agreement of 87% of the re-sponses.Furthermore, self, partner, and self-partner interaction were the most prominent causes in describing the perceived as good and bad romantic relationships.Thus, Hypothesis 1a was confirmed.
Analysis of the frequencies of the attribution elements revealed significant effects.More accurately, the perceived good romantic relationships were mainly attributed to love (24.8%), passion (12.60%), effective communication (11.00%), understanding each-other (10.20%), and honesty-loyalty (9.10%), x 2 (12, Ν = 254) = 174.50,p < .01.Contrarily, the perceived bad romantic relationships were mainly attributed by the students to untrustworthiness (15.20%), lack of passion (13.60%), lack of honesty-loyalty (12.10%), lack of love (10.60%) and ineffective communication (9.10%), x 2 (12, Ν = 132) = 40.20,p < .01.In addition, the participants, who estimated their romantic relationships as good, compared to participants, who perceived their romantic relationships as bad, more often mentioned communication, x 2 (1, Ν = 40) = 6.45, p < .05,love, x 2 (1, Ν = 77) = 31.18,p < .01,passion, x 2 (1, Ν = 50) = 4.00, p < .05,understanding each-other, x 2 (1, Ν = 36) = 7.12, p < .05,and respect each-other, x 2 (1, Ν = 18) = 5.50, p < .05.The findings with respect to attributional dimensions were consistent with those of attribution elements.Specifically, the results from the repeated measures MANOVA (using the Wilks's lambda estimate) with the eight attributional dimensions as within-subjects factor and the perceived romantic relationship (positive/negative) as between-subjects factor revealed significant effect of the attributional dimensions, F(7, 378) = 41.22,p < .01,η 2 = .44,significant effect of the perceived romantic relationship, F(1, 384) = 245.00,p < .01,η 2 = .39,and significant multivariate effect, F(7, 378) = 23.15,p < .01,η 2 = .30.The results from subsequent two repeated measures Anovas, examining differences among attributional dimensions within each group of the students' (perceived good/ bad) romantic relationships, showed significant effect in attributional dimensions in the group with the perceived good romantic relationships, F(7, 242) = 48.33,p < .01,η 2 = .59,and in the group with the perceived bad romantic relationships, F(7, 125) = 19.40,η 2 = .52.Post hoc pairwise comparisons indicated that the students predominately attributed their good romantic relationships to locus of causality, and self-partner interactive both locus of causality and controllability, while they mainly ascribed their bad romantic relationships to partner's locus of causality and controllability, and self-partner locus of causality.The examination of the mean score and the results from subsequent Anovas, and Discriminant Function analysis (Table 2) showed that the participants made internal to themselves, personal controllable, external uncontrollable, stable, controllable by their partners, internal to their partners, internal to selfpartner and controllable in common with their partners attributions for the perceived good romantic relationships.In contrast, Copyright © 2012 SciRes.they mainly made external to themselves, unstable, external uncontrollable, personal uncontrollable, partner's internal, partner's controllable, self-partner's internal and self-partner's controllable attributions for the perceived bad romantic relationships.Additionally, locus of causality, discriminating power = .86,Cohen's d = 1.60, followed by personal controllability, discriminating power = .41,Cohen's d = 1.30, and stability, discriminating power = .33,Cohen's d = 1.21, was the most powerful factor in discriminating the one from the other group of the participants, while the rest of the attributional dimensions had no significant contribution in separating the two groups.
The above findings partly confirmed Hypotheses 1b and 1c.

Εffects of the Perception-Partner Ideal Discrepancies on Perceptions of the Quality of the Romantic Relationship, and on Attributions
The results from ANOVAs revealed significant differences between the two groups of students in perception-partner ideal discrepancies.In addition, discriminant analysis, with stepwise method, was conducted to determine the set of variables that best discriminated the group of students with the perceived good romantic relationships from the group of students with the perceived bad romantic relationships.The perception-partner ideal discrepancies in attractiveness/vitality, warmth/intimacy and trustworthiness were the predictor variables, and the perceived (good/bad) romantic relationship was the grouping variable.Results confirmed the findings from ANOVAs, and, additionally, showed that trustworthiness, F(1, 384) = 65.35,p < .01,discriminating power = .96,Cohen's d = .86,was the most powerful factor in discriminating the two groups of students, followed by warmth/intimacy, F(1, 384) = 61.35,p < .01,discriminating power = .93,Cohen's d = .80,and attractiveness/ vitality, F(1, 384) = 39.81,p < .01,discriminating power = .74,Cohen's d = .63.Furthermore, examination of the mean scores showed that the students, who estimated their romantic relationships as good, compare to students, who estimated their romantic relationships as bad, had smaller perception-partner ideal discrepancies.
Also, the results from regression analysis, with the perception-partner ideal discrepancies as predictor variables and the perceived quality of the romantic relationship as predicted variable, showed that the perception-partner ideal discrepancies was a significant factor in the generation of the perception of the current quality of the romantic relationship.Specifically, the students who had smaller perception-partner ideal discrepancy, compared to students who had larger perception-partner ideal discrepancy, rated their relationships more favourable, R 2 = .40,F(3, 382) = 75.60,p < .01.However, only trustworthiness had unique effect on it, b = .44,t = 5.04, p < .01.
The above findings partly confirmed Hypothesis 2a, and totally confirmed Hypothesis 2b.

Effects of the Perception-Partner Ideal Discrepancies, Perceived Quality of Romantic Relationship and Attributions on Romantic Relationship Expectations
The results from regression analysis indicated that the three concepts, namely perception-partner ideal discrepancies, attributions (external controllability was not entered into analysis because of lack of correlation with the other variables) and the perceived current quality of the romantic relationship, as a group, positively influenced the generation of expectations of the quality of the relationship in the future, explaining 44% of the variance, F(11, 374) = 34.73,p < .01.Furthermore, stability, b = .17,t = 2.73, p < .05,self-partner interactive controllability, b = .25,t = 3.85, p < .01,and, mainly, the perceived quality of the relationship, b = .54,t = 20.15,p < .01,accounted for unique variance in the expectations for the quality of the relationship in the future, suggesting that the more stable and self-partner controllable the attributions, and the more favourable the perceptions of the romantic relationship the higher the expectations for good relationship in the future.However, perception-partner ideal discrepancies in attractiveness/vitality or warmth/intimacy or trustworthiness did not uniquely contribute into relationship expectations.
Hypothesis 3 was partly confirmed by the above findings.

Discussion
This study mainly aimed at examining emerging adults' attributions for their subjectively perceived current romantic relationship as good or bad, the role of their perception-partner ideal discrepancies in evaluating and attributing causes for the same relationship, and, in turn, the effects of the three concepts on romantic relationship expectations.
The findings regarding attributions were in the main consistent with our predictions.The primary finding that the perceived good and bad romantic relationships were attributed to various causes is not surprising given that romantic relationship is important in people's life, particularly in the stage of emerging adulthood (Arnett, 2004(Arnett, , 2007;;Fincham & Cui, 2010;Maner & Miler, 2010).Also, corresponding to previous research evidence (e.g., Fincham, 2003;Planalp & Rivers, 1996;Rusbult et al., 2003), the participants stressed self-, partner-, self-partner-and situational-related factors in explaining their romantic relationships.Furthermore, by stressing intimacy-, trustworthiness-, honesty/loyalty-, and commitment-related factors in their current romantic relationships, the emerging adults confirmed previous researches about the high importance of these specific components in close relationships, and they expressed their willingness to maintain and enhance their relationships (see Barry & Lawrence, 2008;Fehr, 2003;Fitness et al., 2005;Holmes & Rempel, 1989;Johnson, 1991;Jones, Couch, & Scott, 1997).Also, as previous investigations documented, particularly at the early stages of romantic love (see Dion & Dion, 1996;Fletcher, 2002;Hatfield, 1988), the students considered love and passion as significant components of their romantic relationships.The students mentioned geographical distance as a cause for their perceived good or bad current romantic relationship.However, as Myers (1999) suggests, the critical factor is not the geographical proximity as "functional distance".Individuals, by frequent interaction, become familiar with one another and explore themselves and their potential similarities that contribute to good relationship.
The attributional pattern for the perceived good or bad romantic relationship is consistent with the above mentioned findings about happy/unhappy couples, and it is related to confirmation or non confirmation of a desirable good relationship (see Harvey et al., 2005;Herbert & Popadiuk, 2008;Weiner, 2002Weiner, , 2005)).In addition, by attributing the perceived good romantic relationships to personal positive properties along with the self-partner interactive positive dispositions (love, passion, effective communication, honesty-loyalty, understanding each-other), the emerging adults boosted themselves, enhanced the possibility of being good their relationship in the future, and multiplied the longevity of their relationship (Fincham, 2003;Gottman, 1994Gottman, , 1995;;Harvey et al., 2005;Rusbult et al., 2003;Weiner, 2001).In contrast, by ascribing the perceived non good romantic relationships to personal uncontrollable, external, unstable, parter-self internal and partner's internal and controllable factors (untrustworthiness, lack of passion, non honesty-loyalty, lack of love, and ineffective communication), the students protected themselves and their desire for being better their relationship in the future (see Peterson & Steen, 2005;Seligman, 2002;Stephanou, in press;Weiner, 2001).However, considering the partner as responsible for the bad relationship does not facilitate future positive relationship (Karney et al., 2003;Weiner, 1995Weiner, , 2001)).Furthermore, attributing the non satisfactory romantic relationship to partner's dispositional negative factors minimizes the chances of enhancing the relationship in the future (see Fletcher, 2002;Fincham et al., 1998;Planalp & Rivers, 1996).Also, it seems that the Berscheid and Walster's (1978) concept of reciprocity of attraction involved in attributions of the self-partner internal factors, according to which people are attracted to individuals who like them and who are responsive to them.Probably, the partners were less attractive each other than they were in past.
Also, locus of causality, followed by personal controllability and stability, was the most powerful factor in discriminating the two groups of the emerging adults, while none attributional dimensions further differentiated the two groups.This specific finding probably reflects the emerging adults' tendency of self-focus in identifying themselves (see Clark & Beck, 2010).However, research needs to clarify this speculation.
To summarize, the findings regarding the perceptions of partner ideals discrepancy were in the main consistent with our expectations, and previous researches (see Campbell & Baumeister, 2003;Rusbult et al., 2003;Simpson et al., 2003).Specifically, the perceptions of partner ideals discrepancy had a significant effect on the evaluation of the current romantic relationship.More precisely, the emerging adults who had smaller perception-partner ideal discrepancy, compared to the emerging adults who had larger perception-partner ideal discrepancy, estimated their current romantic relationship more favourable, and used positive attributional pattern.
Apparently, the perceived good current romantic relationship had stabilized and reached relatively high levels of commitment, and, therefore, the motive of relationship enhancement and partner/relationship idealization process was powerful (see Murray et al., 1996aMurray et al., , 1996b;;Simpson et al., 2003).On the other hand, the larger perception partner-ideals discrepancy in the perceived bad current romantic relationship seemed to motivate the emerging adults to engage in more information processing about the partner/relationship, and to use the motive of selfserving to maintain positive self-views than views of the current partner/relationship (see Simpson et al., 2003;Taylor, 1989;Weiner, 2005).The specific stage of the participants' life might be also related to this specific finding, since emerging adulthood is an age of instability during which the individuals, aiming to explore the word and clarify their identity, ask and wonder about the person they should find as a partner through life (Arnett, 2000(Arnett, , 2004;;Fincham & Cui, 2010).Yet, the high importance of the romantic relationship for the students may be an influential factor in it, since self-evaluation is particularly enhanced under such conditions (Tesser, 1988).
It was also found that the perceptions of partner ideals discrepancy explained a greater amount of the variability in the perceptions of the quality of the current romantic relationship than in the subsequent attributions.This is an indication of the notion that ideals standards are more associated with self-concepts than beliefs about the relationship, and perceptions of the relationship are related to self-factors (Campbell & Baumeister, 2003;Simpson et al., 2003).However, research needs to examine in depth this.
Interesting, the components of the perception-partner ideal discrepancy differed in the generation of the perceptions of the quality of the current romantic relationship.Specifically, the perception-partner ideal discrepancy in trustworthiness proved the critical factor in the formulation of the perceptions of the quality of the relationship, and in discriminating the group of students with the perceived good current romantic relationships from the group of students with the perceived bad good current romantic relationships.This finding is in agreement with empirical evidence showing the important role of relational trust and betrayal in romantic relationships (Jones et al., 1997).Relational trust means that the individuals have expectations that the partner is predictable and dependable, believe that they will be so in the future, and sustain relationship (Carr, 2005;Holmes & Rempel, 1989).
Perception-partner ideal discrepancy in warmth/intimacy, closely to trustworthiness, contributed into separation of the two groups of the participants.This finding also confirmed other studies, documenting the high importance of perceiving and displaying positive types (e.g., acceptance, respect, listening, care being, interesting in partner) of social behavior in attaining and maintaining a good romantic relationship (Gottman, 1994(Gottman, , 1995;;Rusbult, Zembrodt, & Gunn, 1982).Furthermore, as Harvey et al. (2005) suggested, acceptance implies trustworthiness and discretion, and, probably, therefore, it had not unique effect (beyond that afforded by trustworthiness) on the generation of the estimation of the quality of the whole romantic relationship.
The perception partner ideal discrepancy in attractiveness/ vitality, in comparison to other two factors, contributed to a lower extent in separating the two groups of the students, suggesting its significant role in the early stage of forming a romantic relationship (Fitness et al., 2005).
The findings with respect to the effects of the three components of the perception partner ideal discrepancy on the attributional dimensions for the perceived quality of the current romantic relationship were similar to that of the evaluation of the relationship, and the same speculations could be applied.More precisely, trustworthiness uniquely contributed into the generation attributional dimensions of stability, and locus of causality of self, partner and self-partner, warmth/intimacy had unique and complimentarily effect on self locus of causality, while attractiveness/vitality had no effect.It is interesting that the perception-partner ideal discrepancy in trustworthiness influenced personal/relationship dispositional attributes, suggesting its crucial role in relationship development and stability, and in partners' happiness, self-esteem and identity (Campbell & Baumeister, 2003;Simpson et al., 2003).Further, idealization of the relationship/partner is associated to a more satisfying relationship and to robust self-view (Murray et al., 1996a(Murray et al., , 1996b)).
The results from the present study also, confirming in the main our predictions and other findings (see Berscheid & Am-mazzalorso, 2003;Fincham, 2003;Karney et al., 2003;Stephanou & Kyridis, 2011;Trope & Gaunt, 2005;Weiner, 2002), showed that the expectations the participants hold about their romantic relationships were partly constructed through the cognitive appraisals they made.The more stable the attributions the higher the relationship expectations, in agreement with Weiner's (2002Weiner's ( , 2005) ) model.The positive effect of the self-friend interactive controllability on expectations is another indication of the significant role of the sense of "being together" in relationship development and enhancement.
Τhe fact that the intuitive appraisal of the romantic relationship was the most powerful formulator of relationship expectations is not a surprise since the literature appears to be consistent about the importance of the evaluation of the relationship in its development (Fletcher, 2002;Harvey et al., 2005).Besides as the romantic relationship develops, the partners are forming expectations for the partner dispositional behaviour and the content of the relationship (Berscheid & Ammazzalorso, 2003;Fincham, 2003).The participants in the present study had already formed an impression about their current romantic relationship since the duration of it ranged from six months to three and half years.
However, contrarily to our hypothesis, perception-partner ideal discrepancies in attractiveness/vitality or warmth/intimacy or trustworthiness did not uniquely contribute into romantic relationship expectations.Probably, perception-partner ideal discrepancies influenced the relationship expectations indirectly through the cognitive appraisals of the relationship.Research needs to clarify this finding, taking into consideration self-factors and relational experience.

Implication of the Findings into Emerging Adulthood and Future Research
Good romantic relationship proved important for the emerging adults.Emerging adults should be helped develop the capacity to make and maintain stable and satisfying romantic relationships.This capacity is partly acquired through attributional retraining programs (Seligman, 1998(Seligman, , 2005)).
The findings from the present study indicate that identifying the content and the effects of partner ideals on attributing causes and forming expectations help to understand romantic relationships in emerging adulthood.However, examining self-, partner-and relationship-ideals in association to perceived causes and 'success' of relationship is more helpful in it.Research also needs to investigate the role of self-factors, such as relationship self-relevant self-esteem, self-efficacy and selfperceived mate value, and relational experience in the observed associations.Yet, research should be expanded to general population, beyond the university students.Finally, the dyad-level (both partners) process and effects on relationship are needed to be investigated.

Table 1 .
Frequency of students' attributions for their perceived good/bad romantic relationships.
Note: The nature of attribution is positive and negative in good and bad romantic relationships, respectively.

Table 2 .
Results from discriminant analysis for the attributional dimensions for the perceived (good/bad) romantic relationships.
Note: All F(1, 384) values are significant at the .01level of significance; -: Non significant at the .05level of significance; --: Attributional dimensions did not to further differentiate the one group from the other group of students.