Reintroducing Rationalization: A Study of Relational Goal Pursuit Theory of Intimate Partner Obsessive Relational Intrusion

Abstract

Obsessive relational intrusions (ORI) are the repeated and unwanted attempts by one person to initiate or maintain an intimate relationship with a specific, targeted, and unwilling other. To date, ORI has been commonly explained by relational goal pursuit (RGP) theory. Centrally, RGP theory posits that five clusters of goal-related cognitions and emotions explain ORI. These are goal-linking, self-efficacy, rumination, affective flooding, and rationalization. However, while the first four factors have been empirically investigated as predictors of ORI, rationalization has not. Thus, the current study aimed to reintroduce rationalization to the evaluation of the RGP model among heterosexual former intimate partners. Participants (N = 379; 45.6% female, M_age = 34.4) were recruited from North America and completed an online survey assessing the RGP factors and engagement in ORI. Overall, we found that, after the other factors from the RGP model were considered, the inclusion of rationalization increased the variance explained in ORI. Furthermore, the rationalization subfactor of distortion (as opposed to permissiveness) uniquely predicted ORI both at a broad level and across specific clusters of ORI behavior. These findings not only support the inclusion of rationalization within the RGP model but also provide initial evidence that this construct might be the most individually critical to the explanation of a wide array of ORI behaviors. In addition to having implications for the prediction and explanation of ORI, these findings can also be used to direct clinical treatment of ORI perpetrators toward addressing defenses of rationalization.

Keywords

obsessive relational intrusions stalking relational goal pursuit theory intimate partner violence rationalization

The end of an intimate relationship is typically followed by feelings of distress, lasting for a limited period of time (Sprecher, Felmlee, Metts, Fehr, & Vanni, 1998). Frequently, a member of the dyad attempts at least once to remain connected to their former partner, even when this desire is not reciprocated. Indeed, the majority of breakups include at least one unwanted pursuit behavior such as a phone call or asking friends about a former partner (Langhinrichsen-Rohling, Palarea, Cohen, & Rohling, 2000). The breakup process can, however, become more complicated when unrequited desire for intimacy with a former partner is enduring. This can result in unwanted and persistent pursuit behaviors known as obsessive relational intrusions (ORI; Cupach & Spitzberg, 1998). ORI can severely impinge on the targeted person’s daily life, and physical and mental well-being (Melton, 2007).

Relational goal pursuit (RGP) theory proposes ORI to be the result of five processes: goal-linking, self-efficacy, affective flooding, rumination, and rationalization (Cupach, Spitzberg, & Carson, 2000). Although there has been some empirical assessment of RGP, to date, rationalization has been excluded from such assessments. The current study addresses this by developing a measure of rationalization and including it in a comprehensive empirical assessment of the RGP model in the context of ORI by former intimate partners. In doing so, this study aims to consolidate the validity of RGP theory in relation to ORI and establish the importance of rationalization within the model.

ORI

The term ORI refers to repeated behaviors enacted in an attempt by one person (the pursuer) to establish or regain an intimate relationship with an unwilling other (the target). Severe ORI that induces a sense of threat in the target crosses the legal threshold and constitutes stalking (Cupach et al., 2000). Thus, ORI incorporates some, but not all, stalking behavior. Specifically, the term ORI only includes stalking behaviors motivated by a desire to form or maintain an intimate relationship with the target, as opposed to any other drive such as, for example, financial gain or malicious intimidation (e.g., stalking intended to rob a person). Thus, broadly, ORI incorporates a vast array of behaviors specifically focused on attaining an intimate relationship, ranging from those which might be mildly annoying to highly threatening.

Previously, ORI has been categorized into eight behavioral clusters (Cupach & Spitzberg, 2004). These are hyperintimacy (exaggerated and inappropriate courtship behavior), interactional contact (attempting interaction in the physical vicinity of the target), mediated contact (attempting interaction via some variation of media), surveillance (monitoring the target in the absence of interaction), invasion (intruding in the target’s symbolic or physical space), harassment and intimidation (vexatious behavior intended to incite fear or compliance), coercive threats (veiled warnings and overt threats toward the personal safety of the target, associates, and property), and coercive violence (physical or sexual acts that are violent or forceful). ORI can therefore be understood as either an overarching composite construct or as specific and separate clusters of unwanted pursuit behavior.

ORI is a common experience (Spitzberg, Cupach, & Ciceraro, 2010; Spitzberg, Cupach, Hannawa, & Crowley, 2014). Up to 51% of young adults report experiencing ORI (Fais, Lutz-Zois, & Goodnight, 2017) and, in the context of the broader community, 16% of women and 5% of men report being stalked during their lifetime (Black et al., 2011). This suggests women are at an elevated risk of being the targets of extreme ORI; however, when looking at ORI in young adult samples, little to no gender differences in prevalence has been found (e.g., Langhinrichsen-Rohling et al., 2000). Upon conducting meta-analyses of stalking and ORI behaviors, Spitzberg et al. (2010) concluded ORI was of concern to both men and women. Thus, the current study focuses on both genders.

Intimate Partner ORI

ORI can arise where the target and pursuer have a nonintimate prior relationship, or even no prior relationship at all (Cupach et al., 2000). However, the most common context in which ORI emerges is between former intimate partners (Spitzberg & Cupach, 2007). Indeed, in the case of ORI that crosses the threshold into stalking, the most common perpetrator is a former intimate partner (Tjaden & Thoennes, 1998). It has been theorized that ORI includes exaggeration of ordinary relationship behaviors that, if reciprocated by the target, could be deemed acceptable or even prosocial (Cupach et al., 2000). This may contribute to the common misperception that current or former intimate partner ORI behavior is less serious than ORI in which the pursuer is a stranger (Scott, Nixon, & Sheridan, 2013). Perhaps consequently, former partners, in comparison with stranger pursuers, are less likely to be convicted when ORI crosses the legal threshold to stalking (Sheridan & Davies, 2001). Given the high prevalence of intimate partner ORI, as well as correlations with other types of abuse (e.g., psychological; Davis, Ace, & Andra, 2000), and high levels of posttraumatic stress disorder among targets (Edwards & Gidycz, 2014), it is important to understand the specific mechanisms that may account for it. The present study therefore focuses specifically on intimate partner ORI.

RGP Theory and ORI

To date, a popular theoretical explanation for ORI has been RGP theory (Cupach et al., 2000). The theory operates on the premise that an intimate relationship can be conceptualized as the pursuit of a desired end state (i.e., a goal) in which ordinary relationship pursuit tactics (e.g., gift giving, proximity seeking) become persistent and obsessive by way of five processes: goal-linking, self-efficacy, rumination, affective flooding, and rationalization. The processes described within RGP likely occur in many typical breakups. For example, intense, noncontrollable negative affect is common after relationship dissolution (Davis, Shaver, & Vernon, 2003) and is representative of affective flooding. However, according to RGP theory, when specific processes co-occur, and escalate in persistence, the breakup process becomes complicated and ORI behaviors emerge.

According to RGP theory, the importance of the relational goal becomes exaggerated by the pursuer; the goal of reestablishing the relationship comes to be seen as vital to the achievement of higher order goals such as overarching life-success (i.e., goal-linking). The pursuer also believes the intimate relationship goal to be obtainable with sufficient effort (i.e., self-efficacy), despite the target’s unwillingness to engage in the intimate relationship. This rejection obstructs goal achievement, causing the pursuer to have excessive intrusive thoughts about the target (i.e., rumination). This is understood to occur via two main pathways: rumination persistence, which refers to the pursuer’s difficulty in ceasing thoughts related to their target, and rumination affect, which reflects emotional content associated with the pursuit process that centers around the possibility of reuniting with the target (Spitzberg et al., 2014). Furthermore, the pursuer experiences intense and overwhelming negative feelings (i.e., affective flooding). This includes breakup distress, which is emotional distress concurrent to the actual pursuit process, and residual distress, which is enduring emotional suffering that continues beyond the pursuit–rejection cycle (Spitzberg et al., 2014). Finally, pursuers adopt a cognitive style that justifies their unwanted pursuit behavior by downplaying its negative impact and misconstruing the target’s behavior as being conducive to intimacy (i.e., rationalization). That is, the pursuer unconsciously incorporates rationalization as a defense mechanism to relieve the discomfort of unrequited attempts at intimacy and to justify persistence in the face of rejection (Cupach et al., 2000; Spitzberg & Cupach, 2014). In the context of ORI, rationalization is understood to have two main underlying processes: permissiveness, or positive attributions of the pursuer’s own behavior, and distortions when interpreting the intentions of the target (e.g., seeing the target as desiring the relationship).

Existent empirical literature largely supports RGP theory (e.g., Cupach, Spitzberg, Bolingbroke, & Tellitocci, 2011; Johnson & Thompson, 2016; Spitzberg et al., 2014). For example, Johnson and Thompson (2016) measured rumination and affective flooding and found that 52% of convicted ex-intimate stalking offenders reported experiences of rumination, whereas 40% reported feeling angered and 59% reported feeling emotionally hurt by their target (i.e., affective flooding). Furthermore, Dardis and Gidycz (2017) found that RGP (formulated as a latent variable with rumination, goal-linking, and affective flooding indicators) predicted past engagement in minor unwanted pursuit behaviors both online and in person. Cupach et al. (2011) and Spitzberg et al. (2014) represented RGP by goal-linking, rumination, affective flooding, and self-efficacy. Both studies supported the unique effects of rumination, but only Cupach et al. (2011) found that self-efficacy independently predicted ORI. Furthermore, neither study supported affective flooding or goal-linking as independently predicting ORI. Overall, this literature provides support for the RGP model in general, while suggesting that some constructs could be more critical within the model than others.

It should be noted, however, that the studies conducted so far include only subsets of the five proposed facets of RGP, and none look at rationalization. Rationalization is a defense style that is incorporated to conceal the unconscious motives of conscious behavior (Abbass, 2015) and serves an obstructive purpose by providing an individual with a series of “reasons” that justify negative behavior and circumvent emotions (e.g., guilt, grief; Davanloo,1980). Although rationalization might broadly be common in many breakup scenarios, in terms of RGP theory, it manifests as the pursuer holding a suite of beliefs that the targeted person secretly desires the intimate relationship, irrespective of what that person says or does. When the target rejects the pursuer’s intimacy, this frustrates the pursuer’s goal and activates a set of cognitions that support these beliefs and facilitate the unwanted pursuit. Therefore, theoretically, in the case of ORI, a pursuer is likely to engage in some form of rationalization, such as minimizing target suffering or reframing the behavior as actually wanted, to be able to engage in ORI. Adding to this theoretical relevance, there is also some indirect clinical evidence that rationalization is key to understanding ORI. In qualitative work, Meloy (1998) reported on his interactions with clients who exhibited stalking behaviors. He noted rationalization was a commonly employed defense mechanism among these clients (e.g., a client stated, “she deserves it—look at what she did to me,” p. 9).

Irrespective of this, rationalization, in empirical literature at least, has been excluded. In some cases, rationalization is included in conceptual discussion of RGP theory, but is noted as not being of relevance to the specific study’s research questions (e.g., Cupach et al., 2011). Other research (e.g., Johnson & Thompson, 2016) describes RGP theory as comprising four main factors: goal-linking, self-efficacy, affective flooding, and rumination. As rationalization was not included in the early battery of RGP scales developed, the absence of a suitable measure of rationalization to isolate this construct in the context of RGP theory has likely also restricted empirical investigation of this factor. Thus, the exclusion of rationalization from investigations of RGP theory appears to be an oversight, given that it remains a central component of recent RGP theorizing (e.g., Spitzberg & Cupach, 2014) and that there is circumstantial evidence that rationalization is a core component of the mind-set of those engaging in ORI (Meloy, 1998). Consequently, the reintroduction of rationalization to the empirical evaluation of RGP theory seems vital to providing a comprehensive understanding of the application of this theory to ORI.

Finally, there has been little exploration of how RGP facets relate to each of the eight clusters of ORI behavior separately. Spitzberg et al. (2014) explored correlations between RGP factors (goal-linking, self-efficacy, rumination, affective flooding) and found significant and positive correlations between some RGP factors and certain ORI clusters (e.g., goal-linking, rumination, and affective flooding positively correlated with hyperintimacy, mediated contact, interactional contact, and surveillance) but not others (i.e., none of the RGP factors were associated with intrusion, harassment, coercive threats, or coercive violence). This suggests the RGP factors evaluated might relate more to less menacing behaviors (e.g., hyperintimacy) compared with more overtly hostile behavior (e.g., coercive violence). However, exploration of these relationships beyond the zero-order level appears absent, and, consequently, there is little information regarding the unique relationships between RGP factors and ORI clusters. Thus, we suggest it is important to assess all facets of RGP with each ORI cluster to better understand the unique relationships before conclusions can be confirmed.

The Current Study

In this study, we argue that RGP theory provides an account of ORI arising between former heterosexual intimate partners and that the inclusion of rationalization in the RGP model will provide a more comprehensive account of such ORI. The decision to focus on heterosexual couples was the result of reported differences between heterosexual and same-sex ORI (Edwards et al., 2015).

To summarize, the current study had two main aims. First, we aimed to contribute to the growing body of literature examining RGP theory in the prediction of ORI behaviors, by exploring whether each of the RGP constructs previously considered in empirical research (i.e., goal-linking, self-efficacy, affective flooding, and rumination) makes a unique contribution to the explanation of ORI among heterosexual former intimate partners. Second, we aimed to reintroduce rationalization to the RGP model, with the expectation that this factor would contribute significantly to the explanation of ORI beyond the four other RGP factors.

Method

Participants

Participants were 470 adults from the United States, recruited via the online platform Amazon Mechanical Turk (MTurk). Study advertising stipulated that, to be eligible, participants must have been involved in a heterosexual intimate relationship (referred to as the identified relationship) that lasted at least 2 months and that ended either temporarily or permanently at least 4 weeks prior to completing the study. Data from 91 participants were deleted prior to analysis as they did not meet relationship inclusion criteria, failed attention checks, or were assessed as having spent inadequate time completing the survey (i.e., more than two standard deviations shorter than mean completion time). This left a final sample of 379 adults aged 18 to 72 years (M = 34.4, SD = 9.6).

The demographic and identified relationship information for participants is presented in Table 1. The final sample included slightly more males (54.4%) than females. Most participants identified as heterosexual (92.9%) and the remainder had previously been involved in at least one heterosexual relationship. At the identified relationship’s peak, the majority reported being in a serious dating relationship (71.0%).

Table 1.

Participant Demographic and Identified Relationship Information (N = 379).

	% or M (SD)
Demographics
Age (years)	34.4 (9.6)
Gender
Male	54.4%
Female	45.6%
Sexuality
Heterosexual	92.9%
Bisexual	6.3%
Gay or lesbian	0.8%
Ethnicity
Multiple	9.2%
Caucasian	75.2%
African American	10.8%
Asian	10.0%
Latino/a	8.4%
Native American	2.9%
Other	1.9%
Employment
Not working	14.8%
White collar	58.6%
Blue collar	41.4%
Education
University degree	60.0%
Diploma/high school certificate	39.5%
Current relationship status
Married	31.7%
Dating	26.0%
Single	47.7%
Identified relationship information
Breakup most desired by
My partner	39.6%
Myself	39.1%
Mutually desired	21.4%
Breakup initiated by
My partner	42.5%
Myself	33.2%
Equally initiated	24.3%
Strength of relationship at its peak
Married	11.9%
Engaged	8.4%
Dating seriously	71.0%
Dating casually	8.7%

Measures

Existing RGP measures

RGP scales were drawn from Spitzberg et al. (2014) to measure goal-linking, self-efficacy, rumination, and affective flooding. Each scale asked participants to use a 7-point response scale (1 = strongly disagree to 7 = strongly agree) to rate a series of statements preceded by the conditional phrase “After we broke up. . .,” with the exception of goal-linking which was preceded by “Before we broke up. . .”

Goal-linking (α = .86) included eight items (e.g., having this person in my life seemed essential to becoming who I wanted to be). Self-efficacy (α = .90) included eight items (e.g., I believed that persistence in trying to reestablish the relationship with my ex-partner would pay off). Rumination included 20 items in two subscales: Rumination Persistence (α = .92) included 12 items (e.g., I thought about this person even more when I tried not to) and Rumination Affect (α = .95) included eight items (e.g., I am worried that we might not ever get back together). Affective flooding included 11 items across two subscales: Residual Distress (α = .90) included seven items (e.g., even now I get upset thinking about this person), and Breakup Distress (α = .84) included four items (e.g., I felt intense emotions when this person wanted out).

Rationalization

A scale, consisting of eight items, was developed to measure an individual’s tendency to utilize defenses of rationalization. The items were preceded by the conditional phrase “After we broke up. . .” As this construct has not previously been measured in RGP scales, items were generated from the literature related to rationalization defense mechanisms (Abbass, 2015; Frederickson, 2013), RGP theory (Cupach et al., 2000; Spitzberg & Cupach, 2014), and an existing nonrelationship-specific rationalization measure (Defense Style Questionnaire–40 [DSQ-40], α = .80; Andrews, Singh, & Bond, 1993). All items are listed in Table 2. The two main subdomains of rationalization identified within the RGP literature (Cupach et al., 2000; Spitzberg & Cupach, 2014) were included as factors within the scale: permissiveness of pursuer’s own behavior and distortion of the target’s intentions and behaviors. Each factor consisted of four items and the 7-point response scale from existing RGP measures (i.e., 1 = strongly disagree to 7 = strongly agree). As this is a new measure, exploratory factor analysis with oblique rotation was conducted to assess the structure of the eight items (see Table 2). Two eigenvalues > 1.0 and the scree plot indicated a two-factor solution, with Item 2 and Item 3 presenting poor communality (<.4). In addition, Item 3 did not load on the expected factor (i.e., distortion). As Item 3 was more problematic, it was removed. Using the remaining seven items, a clean two-factor solution, with a Kaiser–Meyer–Olkin (KMO) of .8, accounted for 58.7% of the common variance. Communality of all items was acceptable (i.e., >.4) and all items loaded as expected. The first factor was labeled permissiveness (α = .70) and consisted of three items (e.g., If I behaved badly, it was because of the way I was feeling). The second factor was labeled distortion (α = .74) and consisted of four items (e.g., If my identified partner ignored me, it was because she or he was playing hard to get).

Table 2.

Rationalization Item Loadings on Permissiveness (Factor 1) and Distortion (Factor 2) (N = 373).

Item	8-Item Solution		7-Item Solution
Item	Factor 1	Factor 2	Factor 1	Factor 2
6. My identified partner might have been upset by my actions but due to the circumstances it wasn’t my fault. (literature)	.78	−.32	.78	−.01
4. Even if my identified partner was upset by my behavior, I was always able to think of a good reason for what I did. (existing measure)	.76	−.38	.71	−.09
7. If things didn’t work out for me, there were good reasons why. (existing measure)	.72	−.19	.77	.14
2. If I behaved badly it was because of the way I was feeling (e.g., hurt, angry, sad). (literature)	.49	−.43	.54	−.20
3. My identified partner was likely wanting to get back together irrespective of what she or he said. (theory)	.49	−.29	−	−
5. I could find reasons to pursue the relationship, even when my identified partner seemed disinterested. (existing measure)	.34	.85	−.01	.83
8. Even if she or he appeared upset by it, there would not be any real harm if I kept trying to contact my identified partner. (theory)	.39	.83	.03	.82
1. If my identified partner ignored me, it was because she or he was playing hard to get. (theory)	.38	.72	−.01	.76

Note. The primary source from which the item was generated is displayed in brackets (i.e., literature = literature related to rationalization defense mechanisms; theory = RGP theory; existing measure = existing nonrelationship specific rationalization measure). RGP = relational goal pursuit.

ORI

ORI was measured with the ORI-42 Perpetration Inventory (Spitzberg et al., 2014). Participants rated the frequency with which they engaged in eight clusters of ORI behaviors using a 7-point scale (1 = never, 2 = once, 3 = 2-3 times, 4 = 4-5 times, 5 = 6-10 times, 6 = 11-25 times, 7 = 25 or more times). Mean scores were calculated for each cluster and, in addition, a total ORI score was generated by summing these subtotals (α = .98). The eight clusters included hyper-intimacy (α = .90; for example, leaving unwanted gifts), interactional contacts (α = .90; for example, showing up at the persons work, school, gym, or place of worship), mediated contacts (α = .85; for example, instigating unwanted phone calls, emails, chat, instant messages), surveillance (α = .90; for example, checking up on him or her through mutual friends), invasion (α = .88; for example, stealing/checking her or his mail/email), harassment and intimidation (α = .88; for example, seeking to be invited to social events through her or his friends, family, or co-workers), coercive threats (α = .95; for example, threats or vague warnings that something bad will happen to her or him), and coercive violence (α = .96; for example, holding the car door while she or her is in the car). Standard instructions were adapted slightly for the current study, as participants were directed to consider their behavior during/after the breakup with their identified partner.

Procedure

Following institutional ethical clearance, the Qualtrics survey link was loaded to the online scientific survey platform MTurk. The reliability of data sourced via this platform has been shown to be similar to data gathered via more traditional methods, and samples are reported to be representative of the general population (Buhrmester, Kwang, & Gosling, 2011). Note that the scales and items used to test the current research question were taken from a larger survey looking at ORI. The larger study included measures unrelated to the current research question (e.g., measures of gender-based attitudes). Participants were each paid US$3.50 for completing the study, which took approximately 20 min to complete. Demographic items were presented at the start and at the end of the survey, with the order of the scales randomized in between.

Results

Preliminary Analyses

During assumption testing, positive skew and extreme scores were identified in the outcome measure. Normality was achieved with a square root transformation; however, there was no variation to the direction or significance of results when these transformed data were used. Therefore, for ease of interpretation, untransformed data were retained and are reported herein. Missing data for each scale ranged from 0% to 1.3% and Little’s Missing Completely at Random (MCAR) test was nonsignificant, χ²(5,890, N = 379) = 5,896.43, p = .474, indicating that there was no identifiable pattern in missing values. As missing data were completely at random and represented a low proportion of the data (<5%), listwise deletion was used to deal with missing data.

Descriptive Statistics and Preliminary Analyses

Means, standard deviations, and bivariate correlations for RGP and ORI are displayed in Table 3. Each RGP variable was significantly and positively correlated with ORI. A series of one-way ANOVAs was conducted to determine if ORI differed depending on who initiated the breakup and, further, how serious the severed relationship was (i.e., casual dating—married). Results indicated there were no significant differences in ORI behavior between people who were the recipients or instigators of the breakup (all ps > .05). Neither did ORI differ according to relationship seriousness (all ps > .05).

Table 3.

Means, Standard Deviations, and Correlations Among Relational Goal Pursuit Variables (N = 378).

	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16
1. ORI	1.60	0.92
2. Hyperintimacy	2.20	1.33	.77**
3. Mediated contact	1.77	1.11	.91**	.76**
4. Interactional contact	1.74	1.14	.91**	.72**	.82**
5. Invasion	1.45	1.00	.90**	.58**	.75**	.82**
6. Harassment/intimidation	1.48	0.97	.92**	.58**	.79**	.80**	.87**
7. Surveillance	1.61	1.02	.91**	.64**	.79**	.83**	.83**	.87**
8. Coercive threats	1.30	0.92	.90**	.56**	.78**	.74**	.86**	.87**	.81**
9. Coercive violence	1.25	0.83	.87**	.52**	.74**	.72**	.83**	.84**	.79**	.95**
10. Goal-linking	4.20	1.76	.18**	.31**	.21**	.16**	.12*	.11*	.18**	.07	.06
11. Self-efficacy	3.77	1.54	.20**	.26**	.20**	.19**	.15**	.17**	.16**	.12*	.12*	.26**
12. Residual affective flooding	4.05	1.65	.26**	.27**	.27**	.28**	.21**	.21**	.27**	.15**	.13*	.39**	.12*
13. Breakup affective flooding	4.41	1.94	.16**	.26**	.20**	.19**	.11*	.10	.18**	.04	.01	.59**	.16**	.73**
14. Rumination persistence	4.27	1.81	.23**	.35**	.25**	.22**	.15**	.16**	.25**	.10	.09	.62**	.28**	.65**	.72**
15. Rumination affect	3.88	1.86	.26**	.39**	.28**	.24**	.16**	.17**	.26**	.12**	.11*	.69**	.32**	.57**	.76**	.83**
16. Rationalization distortion	2.87	1.52	.52**	.54**	.51**	.48**	.43**	.45**	.45**	.40**	.38**	.41**	.38**	.35**	.39**	.45**	.53**
17. Rationalization permission	4.01	1.43	.29**	.31**	.28**	.31**	.24**	.24**	.29**	.18**	.18**	.29**	.19**	.44**	.25**	.32**	.30**	.48**

Note. All measures had a possible range from 1 to 7. ORI = obsessive relational intrusions.

p < .05. **p < .01.

Prior to the main regressions, moderation analyses were conducted to assess if there was a moderating effect of gender for the associations between RGP variables and ORI. Gender was dummy coded and individual interaction terms between gender and each RGP variable were created. In separate moderated regressions, each RGP variable was entered with gender in Block 1 of the model, and the interaction term entered at Block 2. There were no significant interactions (i.e., all ps > .05), which indicated that RGP variables similarly predicted ORI for men and women. Consequently, results reported below refer to a gender-pooled data set that does not control for breakup instigation or previous relationship seriousness.

Primary Analyses

A series of hierarchical regressions was used to examine the unique contribution of RGP factors to the explanation of ORI, to determine if the inclusion of rationalization in the model added to the variance explained in ORI and assess if this model differentially related to certain clusters of ORI behaviors. Due to a large range in age (18-72 years) and time elapsed since the most recent breakup (1 month to >5 years), these, along with gender, were entered as control variables at Block 1 in each regression. Existing RGP measures (goal-linking, self-efficacy, rumination, affective flooding) were included at Block 2, and the newly created rationalization measure was entered at Block 3, to identify if any additional variance could be accounted for by its inclusion in the RGP model. Tables 4 and 5 display beta values and change statistics for the hierarchical regressions.

Table 4.

Predicting IP-ORI as a Function of Control Variables and RGP Variable Subscales (N = 378).

Block	IP-ORI
Block	1	2	3
Controls
Age	−.07	−.06	−.07
Gender	−.03	−.05	−.01
Time	−.21***	−.19***	−.10*
RGP variables
Goal-linking	—	.03	−.01
Self-efficacy	—	.11**	.01
Rumination
Persistence	—	−.07	−.05
Affect	—	.26**	−.05
Affective flooding
Residual distress	—	.27***	.20**
Breakup distress	—	−.20**	−.17
Rationalization
Distortion	—	—	.47***
Permissiveness	—	—	.02
R ² _Ch	.06***	.11***	.14***
Adjusted R²	.06***	.15***	.29***
F _Ch	8.37***	8.17***	37.61***

Note. “Time” refers to time since breakup. IP-ORI = Intimate Partner-Obsessive Relational Intrusion; RGP = relational goal pursuit; R²_Ch = R² Change; F_Ch = F Change.

p < .05. **p < .01. ***p < .001.

Table 5.

Predicting Intimate Partner Obsessive Relational Intrusion Behavior Clusters as a Function of Control Variables and RGP Variables.

Block	Hyperintimacy(N = 378)			Mediated Contact(N = 378)			Interactional Contact(N = 378)			Invasion(N = 378)			Harass/Intimidation(N = 378)			Surveillance(N = 377)			Coercive Threats(N = 378)			Coercive Violence(N = 378)
Block	1	2	3	1	1	2	3	1	1	2	3	1	1	2	3	1	2	3	1	2	3	1	2	3
Control variables
Age	−.05	−.05	−.05	−.08	−.08	−.08	−.06	−.06	−.06	−.08	−.08	−.08	−.05	−.04	−.03	−.05	−.05	−.05	−.08	−.07	−.07	−.06	−.04	−.04
Gender	−.01	−.02	.01	−.07	−.09	−.06	−.01	−.03	−.01	−.01	−.04	−.04	−.04	−.06	−.03	−.01	−.02	−.01	−.03	−.05	−.01	−.03	−.05	−.02
Time	−.14**	−.13*	−.05	−.19***	−.18**	−.09	−.19**	−.16**	−.08	−.20***	−.17**	−.09	−.18**	−.16**	−.07	−.22***	−.19***	−.13*	−.20***	−.19***	−.11*	−.20***	−.18**	−.10
RGP variables
Goal-linking	—	.07	.03	—	.05	.01	—	.01	−.05	—	.04	.01	—	.01	−.02	—	.02	−.02	—	.01	−.02	—	.01	−.02
Self-efficacy	—	.13**	.04	—	.11*	.02	—	.12**	.03	—	.10	.01	—	.11*	.01	—	.07	−.01	—	.07	−.02	—	.07	−.01
Rumination
Persistence	—	−.20	.01	—	−.10	−.08	—	−.06	−.04	—	−.06	−.04	—	−.07	−.04	—	.01	.03	—	−.12	−.10	—	−.09	−.07
Affect	—	.36***	.18	—	.27**	.09	—	.19	.01	—	.11	−.07	—	.20	.01	—	.19	.04	—	.23*	.05	—	.24*	.06
Affective flooding
Residual	—	.13	.05	—	.27***	.21**	—	.27***	.18*	—	.25**	.19*	—	.28***	.24**	—	.24**	.16*	—	.25**	.23**	—	.24**	.21**
Breakup	—	−.14	−.10	—	−.16	−.13	—	−.12	−.08	—	−.14	−.11	—	−.23*	−.20*	—	−.16	−.12	—	−.24*	−.23*	—	−.29***	−.28**
Rationalization
Distortion	—	—	.40***	—	—	.42***	—	—	.42***	—	—	.42***	—	—	.46***	—	—	.35***	—	—	.44***	—	—	.42**
Permissiveness	—	—	.05	—	—	.02	—	—	.07	—	—	.01	—	—	−.01	—	—	.07	—	—	−.05	—	—	−.03
R ² _Ch	.03*	.19***	.11***	.07***	.12***	.12***	.05**	.10***	.13***	.06***	.06**	.11***	.05**	.08***	.13***	.06***	.10***	.09***	.07***	.05**	.10***	.05***	.05**	.10***
Adjusted R²	.02*	.20***	.31***	.06***	.17***	.28***	.04**	.13***	.26***	.05***	.10**	.21***	.04**	.10***	.23***	.05***	.14***	.23***	.06***	.09**	.20***	.05***	.08**	.18***
F _Ch	3.61*	14.65***	29.90***	8.76***	9.29***	30.01***	6.04**	7.50***	32.13***	7.62***	4.11**	26.98***	5.88**	5.24***	31.18***	7.54***	7.14***	22.47***	8.60***	3.47**	25.82***	6.97***	3.43**	23.14***

Note. “Time” refers to time since breakup. RGP = relational goal pursuit.

p < .05. **p < .01. ***p < .001.

Predicting ORI with RGP theory

As can be seen in Table 4, the overall model was significant and accounted for 29% of the variance in ORI. At Block 1, time since the most recent breakup accounted for a significant amount of variance in ORI, such that a greater passage of time was related to less ORI behaviors. Neither of the other control variables (i.e., age and gender) uniquely accounted for variance within ORI. When the RGP variables were introduced at Block 2, self-efficacy, rumination affect, and both affective flooding subscales (i.e., residual and breakup distress) each uniquely accounted for variance in ORI. Specifically, higher levels of self-efficacy, rumination affect, and affective flooding residual distress were related with more ORI behaviors. Conversely, individuals that reported more contemporaneous emotional distress related to the breakup (i.e., affective flooding breakup distress) reported fewer ORI behaviors, despite the opposite relationship presenting at the zero-order correlation level. With the addition of the two rationalization subscales (Distortion and Permissiveness) at Block 3, rationalization distortion uniquely accounted for a significant 11.56% of the variance in ORI, with those individuals who distorted the intentions of their identified partner engaging in more ORI behavior. No other RGP variable remained a unique predictor of ORI, with the exception of affective flooding residual distress.

Predicting Intimate Partner Obsessive Relational Intrusion (IP-ORI) behavior clusters with RGP theory

To understand how RGP theory in general, and rationalization in particular, relates to the various subtypes of ORI, correlations between all RGP variables and ORI clusters were examined (Table 3). Several differences across the ORI clusters at the zero-order correlation level were present, suggesting sufficient complexity across the relationships to warrant further examination. Therefore, a series of hierarchical regressions with control variables at Block 1, existing RGP variables at Block 2, and rationalization subscales at Block 3 were used to examine all RGP variables as predictors of each of the ORI clusters.

As can be seen in Table 5, for each of the eight ORI behavior clusters, the overall model was significant and accounted for between 18% (coercive violence) and 31% (hyperintimacy) of variance. At Block 2, self-efficacy, rumination affect, affective flooding residual, and affective flooding breakup each uniquely accounted for a significant amount of variance for several of the clusters of ORI behaviors. With the introduction of rationalization distortion and rationalization permissiveness at Block 3, however, of the RGP variables, only the variance accounted for by affective flooding subscales (Residual Distress and Breakup Distress) remained significant. As can be seen in Table 5, rationalization distortion significantly accounted for unique variance in each of the eight ORI clusters (hyperintimacy = 8.30%, mediated contact = 9.43%, interactional contact = 9.18%, invasion = 9.49%, harassment/intimidation = 11.20%, surveillance = 6.55%, coercive threats = 10.11%, coercive violence =9.12%). There were no unique significant effects of rationalization permissiveness for any of the ORI clusters. Of note, rationalization distortion was the only unique predictor of hyperintimacy at Block 3, and the only variable to uniquely account for variance in each of the ORI clusters of behavior.

Discussion

RGP theory posits that ORI can be understood by a series of processes: goal-linking, rumination, affective flooding, self-efficacy, and rationalization (Cupach et al., 2000). A small number of studies have empirically evaluated the theory, but none have included the full five factors; prior research empirically evaluating RGP theory in relation to ORI has excluded rationalization. A key purpose of the current study was to reintroduce rationalization and determine if its inclusion strengthens the application of RGP theory to ORI. In line with our predictions, the inclusion of rationalization in the RGP model provided a more comprehensive account of ORI. These results add to the support of RGP as a valid and important theory of ORI and highlight that rationalization is relevant, and critical, to this model.

Consistent with other research (e.g., Cupach et al., 2011; Spitzberg et al., 2014), we found a four-factor model of RGP predicted ORI. However, in current findings, the accuracy of the RGP model’s account of ORI increased from 15% to 29% of variance explained with the introduction of rationalization. Consistent with past research (Cupach et al., 2011; Johnson & Thompson, 2016; Spitzberg et al., 2014), we found self-efficacy, rumination affect, and affective flooding uniquely predicted ORI. However, in current findings, a negative beta weight was displayed by affective flooding breakup distress (see Table 4), despite the presence of a positive zero-order correlation with ORI. This could indicate there is an avoidance component of affective flooding breakup distress when other predictors are taken into account. It could also be the result of a statistical artifact, as Spitzberg et al. (2014) has mentioned that such negative associations may be due to complex multivariate relationships within the RGP model. Future research may benefit from evaluating the exact nature of these complex relationships.

More critically, we found rationalization could be of greater relevance to ORI than other RGP variables, as it was the strongest and most consistent unique predictor. Specifically, individuals who viewed their identified partner’s behavior through a distorted lens to perceive signs of encouragement for the pursuit (rationalization distortion), but not those who tended to downplay the negative consequences of pursuit (rationalization permissiveness), were most likely to report also having engaged in ORI. Significantly, rationalization distortion was also the only RGP variable that uniquely predicted ORI for each of the eight categories of ORI behavior, from the seemingly benign behaviors (e.g., mediated contact) to behaviors that are more evidently hostile (e.g., coercive violence). Thus, individuals who justify their behavior through distorted perceptions of their former partner’s true feelings, wants, and needs are more likely to engage in unwanted pursuit behaviors. At a theoretical level, it is possible that some degree of rationalization is necessary to be able to engage in ORI (e.g., people may need to think that advances are secretly welcomed to engage in them). Although no work has explicitly looked at rationalization predicting ORI within the RGP framework, our findings are broadly consistent with work showing that offenders of ex-intimate partner stalking feel hurt by their former partner, rather than seeing their own behavior as hurtful to their former partner (Johnson & Thompson, 2016). It is possible the rationalization distortion process is so common because it allows individuals to escape from negative affect (e.g., guilt) which would otherwise challenge their persistence in pursuit.

Furthermore, rationalization negated the unique contribution of some RGP variables (e.g., self-efficacy, rumination affect, affective flooding breakup distress) in relation to ORI broadly, and within several clusters. This suggests the effects of some RGP variables in predicting ORI may be mediated by rationalization. For example, in the case of self-efficacy, the belief that pursuit attempts can be successful might lead to the development of complex cognitions that distort perception of the target’s behavior as encouraging the pursuit (i.e., rationalization), and this, in turn, predicts ORI. Further work is required to investigate the exact nature of the complex relationships between RGP variables and ORI.

To gain a thorough understanding of the RGP relationships, it appears there is merit in examining these at the subscale and ORI cluster levels to detect subtle nuances. Our results suggest the overall RGP model may be more applicable to some aspects of ORI (e.g., hyperintimacy) than others (e.g., coercive violence), and individual RGP constructs specifically and uniquely relate to varying isolated components of ORI (e.g., affective flooding residual distress was related to all clusters except hyperintimacy).

At a surface level, these results bear some consistency with recent findings (Dardis & Gidycz, 2017) that suggest RGP is related to “minor” (e.g., unsolicited contact, following) but not “severe” (e.g., threat, kidnapping) unwanted pursuit behaviors, as both findings support the differential nature of relationships between RGP factors and ORI clusters. However, seemingly, there can be both “minor” and “severe” cases of each ORI cluster (e.g., minor surveillance: driving past the target’s workplace; severe surveillance: breaking into and hiding in the target’s home to covertly monitor him or her). As such, behaviors classified as “minor” in previous research have the potential to be severely distressing to the target. This is particularly pertinent to note as both the current study and comparative research (Dardis & Gidycz, 2017) incorporated behavioral frequency measures of unwanted pursuit behaviors that did not explicitly capture severity. Therefore, relationships between RGP and ORI severity cannot be assumed on these findings alone.

Strengths, Limitations, and Future Directions

Current findings support the conclusion that RGP theory gives a more comprehensive account of ORI when rationalization is included, and provide a strong justification for rationalization to remain included in the RGP model as originally proposed by Cupach et al. (2000). Current findings also better our understanding of the complex nature of associations between RGP constructs and ORI clusters of behavior. Several methodological strengths were also present. First, by isolating ORI directed at a target with whom there has formerly been an intimate relationship, the current study controlled for the extraneous influence of the nature of the relationship between target and pursuer. Second, in comparison to many other studies in the field that draw from college populations (e.g., Cupach et al., 2011; Spitzberg et al., 2014), the current sample was drawn from an online North American community population and displayed diversity across ethnicity, age, employment, and education. This provides the benefit of current results being broadly generalizable to the wider community.

A limitation of current findings is the sole focus on ORI within heterosexual relationships; the current results are not able to be extended to sexual minority groups (i.e., lesbian, gay, bisexual, transsexual, queer/questioning [LGBTQ]). To the knowledge of the authors, RGP theory has not yet been empirically applied specifically within sexual minority populations. Given ORI among LGBTQ intimate relationships is reported to be more severe compared with ORI within heterosexual relationships (Edwards et al., 2015), future research could examine whether the explanatory power of RGP theory and, specifically, rationalization extend to LGBTQ populations. Furthermore, given that rationalization had not previously been measured in the context of RGP, there was no existing RGP rationalization measure available. Thus, the current study is limited by the use of an as yet unvalidated measure of rationalization. We encourage future research to further validate this measure. In addition, factors not addressed in the current study (e.g., co-habitation, shared children, and financial assets) could affect behavior in the context of relationship break down. Future research could address this by gathering comprehensive relationship information for analysis. Finally, the current study is limited by the cross-sectional nature of the data. Although RGP theory proposes that its constructs predict ORI, we cannot infer causality from cross-sectional data. Longitudinal and experience sampling studies may be valuable in establishing if the relationships are cyclical in nature, such that specific sequences of RGP factors instigate ORI and others perpetuate such behavior.

Clinical Implications

Given that rationalization appears integral to predicting ORI, it could be a key focus for clinical interventions. Defense mechanisms are pervasive in nature, and thus, a rationalization defense is not likely to self-correct. Rather, clinical intervention would generally be required to assist the individual to confront their feelings (Frederickson, 2013). Previously, the predominant treatment for ORI type behaviors has been Dialectical Behavior Therapy (DBT; Rosenfeld et al., 2007), supported by research that suggests intolerance of distress plays a role in ORI perpetration for some populations (Reilly & Hines, 2017). The current findings, however, suggest there could also be clinical utility in treatment such as Davanloo’s (1990) Intensive Short Term Dynamic Psychotherapy that assists patients to first identify the unhelpful impact of defenses, separate from them, and then address them. This is an important area of development considering the suffering of targets can be alleviated by pursuer rehabilitation, and pursuer focused therapy might also reduce pursuer distress. Although the priority remains the well-being of the target, pursuer care should also be considered of importance given stalking offenders are at significantly greater risk of suicide in comparison with the general population (McEwan, Mullen, & MacKenzie, 2010).

Conclusion

Current findings confirm that the reintroduction of rationalization to RGP theory improves the theory’s applicability to ORI, at least among heterosexual former intimate partners. They also highlight intricate relationships between each broad RGP variable and RGP subfactors, with each cluster of ORI behavior. Critically, our work suggests that one of the most central constructs within the RGP model is the very construct that has previously been excluded: rationalization. A distorted view of the intentions of the target as a means to rationalize the behavior of the pursuer may play an integral role in the perpetration of ORI across each cluster of behavior. Our findings provide a platform for future research to incorporate rationalization in RGP theory evaluation, as well as having implications for future research in risk assessment and clinical interventions for ORI perpetrators.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Jessica Brownhalls is a PhD clinical psychology candidate at Griffith University, and a registered psychologist. Her research interests include gender-based and intimate partner violence, with a focus on offenders’ maladaptive perceptions and behaviors. Her clinical interests lie in attachment trauma, character pathology, and intimate relationship behavior.

Amanda Duffy, PhD clinical psychology, is a senior lecturer in the School of Applied Psychology at Griffith University. Her research focuses on the dysfunctional interactions that can occur between individuals, with a particular focus on aggression and victimization. She is particularly interested in the individual and social contextual factors that contribute to such outcomes.

Li Eriksson, PhD criminology, is a lecturer with the School of Criminology and Criminal Justice at Griffith University. Her research interests include violence, intimate partner homicide, filicide, and criminological theory. Prior to joining Griffith University, she worked as a research analyst for the Swedish National Council of Crime Prevention.

Fiona Kate Barlow, PhD psychology, is an associate professor and Australian Research Council future fellow with the School of Psychology at the University of Queensland. She is a social psychologist with primary research interests in intergroup relations, prejudice, and discrimination.

References

Abbass

(2015). Reaching through resistance: Advanced psychotherapy techniques. Kansas City, MO: Seven Leaves Press.

Andrews

Singh

Bond

(1993). The Defense Style Questionnaire. The Journal of Nervous and Mental Disease, 181, 246-256. doi:10.1097/00005053-199304000-00006

Black

Basile

Breiding

Smith

Walters

Merrick

. . . Stevens

(2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Buhrmester

Kwang

Gosling

S. D.

(2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3-5.

Cupach

W. R.

Spitzberg

B. H.

(1998). Obsessive relational intrusion and stalking. In Spitzberg

B. H.

Cupach

W. R.

(Eds.), The dark side of close relationships (pp. 233-263). Hillsdale, NJ: Lawrence Erlbaum.

Cupach

W. R.

Spitzberg

B. H.

(2004). The dark side of relationship pursuit: From attraction to obsession and stalking. Mahwah, NJ: Lawrence Erlbaum.

Cupach

W. R.

Spitzberg

B. H.

Bolingbroke

C. M.

Tellitocci

(2011). Persistence of attempts to reconcile a terminated romantic relationship: A partial test of relational goal pursuit theory. Communication Reports, 24, 99-115. doi:10.1080/08934215.2011.613737

Cupach

W. R.

Spitzberg

B. H.

Carson

C. L.

(2000). Toward a theory of obsessive relational intrusion and stalking. In Dindia

Duck

(Eds.), Communication and personal relationships (pp. 131-146). Chichester, England: John Wiley.

Dardis

C. M.

Gidycz

C. A.

(2017). Reconciliation or retaliation? An integrative model of postrelationship in-person and cyber unwanted pursuit perpetration among undergraduate men and women. Psychology of Violence. Advance online publication. doi:10.1037/vio0000102

10.

Davanloo

(1980). Short-term dynamic psychotherapy. Northvale, NJ: Aronson.

11.

Davanloo

(1990). Unlocking the unconscious. Chichester, UK: John Wiley.

12.

Davis

Shaver

P. R.

Vernon

M. L.

(2003). Physical, emotional, and behavioral reactions to breaking up: The roles of gender, age, emotional involvement, and attachment style. Personality and Social Psychology Bulletin, 29(7), 871-884. doi:10.1177/0146167203252884

13.

Davis

K. E.

Ace

Andra

(2000). Stalking perpetrators and psychological maltreatment of partners: Anger-jealousy, attachment insecurity, need for control, and break-up context. Violence and Victims, 15, 407-425.

14.

Edwards

K. M.

Gidycz

C. A.

(2014). Stalking and psychosocial distress following the termination of an abusive dating relationship: A prospective analysis. Violence Against Women, 20, 1383-1397. doi:10.1177/1077801214552911

15.

Edwards

K. M.

Sylaska

Barry

Moynihan

Banyard

Cohn

. . . Wards

(2015). Physical dating violence, sexual violence, and unwanted pursuit victimization: A comparison of incidence rates among sexual minority and heterosexual college students. Journal of Interpersonal Violence, 30, 580-600. doi:10.1177/0886260514535260

16.

Fais

C. R.

Lutz-Zois

C. J.

Goodnight

J. A.

(2017). Mediators and moderators of the association between stalking victimization and psychological distress. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517699951

17.

Frederickson

(2013). Co-creating change: Effective dynamic therapy techniques. Kansas City, MO: Seven Leaves Press.

18.

Johnson

E. F.

Thompson

C. M.

(2016). Factors associated with stalking persistence. Psychology, Crime & Law, 22, 879-902. doi:10.1080/1068316X.2016.1197225

19.

Langhinrichsen-Rohling

Palarea

R. E.

Cohen

Rohling

M. L.

(2000). Breaking up is hard to do: Unwanted pursuit behaviors following the dissolution of a romantic relationship. Violence and Victims, 15, 73-90.

20.

McEwan

Mullen

MacKenzie

(2010). Suicide among stalkers. The Journal of Forensic Psychiatry & Psychology, 21(4), 514-520.

21.

Meloy

J. R.

(1998). The psychology of stalking. In Meloy

J. R.

(Ed.), The psychology of stalking: Clinical and forensic perspectives (pp. 2-21). San Diego, CA: Academic Press.

22.

Melton

(2007). Stalking in the context of intimate partner abuse: In the victims’ words. Feminist Criminology, 2, 347-363. doi:10.1177/1557085107306517

23.

Reilly

Hines

D. A.

(2017). Distress tolerance as a mediator of the association between borderline personality symptoms and obsessive relational intrusion: An exploratory analysis. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517712274

24.

Rosenfeld

Galietta

Ivanoff

Garcia-Mansilla

Martinez

Fava

. . . Green

(2007). Dialectical behavior therapy for the treatment of stalking offenders. International Journal of Forensic Mental Health, 6, 95-103. doi:10.1080/14999013.2007.10471254

25.

Scott

A. J.

Nixon

Sheridan

(2013). The influence of prior relationship on perceptions of stalking: A comparison of laypersons, nonspecialist police officers, and specialist police officers. Criminal Justice and Behavior, 40, 1434-1448. doi:10.1177/0093854813494183

26.

Sheridan

Davies

G. M.

(2001). Violence and the prior victim-stalker relationship. Criminal Behaviour and Mental Health, 11, 102-116. doi:10.1002/cbm.375

27.

Spitzberg

B. H.

Cupach

W. R.

(2007). The state of the art of stalking: Taking stock of the emerging literature. Aggression and Violent Behavior, 12, 64-86. doi:10.1016/j.avb.2006.05.001

28.

Spitzberg

B. H.

Cupach

W. R.

(2014). The dark side of relationship pursuit: From attraction to obsession to stalking (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.

29.

Spitzberg

B. H.

Cupach

W. R.

Ciceraro

L. D. L.

(2010). Sex differences in stalking and obsessive relational intrusion: Two meta-analyses. Partner Abuse, 1, 259-285. doi:10.1891/1946-6560.1.3.259

30.

Spitzberg

B. H.

Cupach

W. R.

Hannawa

A. F.

Crowley

J. P.

(2014). A preliminary test of a relational goal pursuit theory of obsessive relational intrusion and stalking. Studies in Communication Sciences, 14, 29-36. doi:10.1016/j.scoms.2014.03.007

31.

Sprecher

Felmlee

Metts

Fehr

Vanni

(1998). Factors associated with distress following the breakup of a close relationship. Journal of Social and Personal Relationships, 15, 791-809.

32.

Tjaden

Thoennes

(1998). Stalking in America: Findings from the National Violence Against Women Survey (NCJ 169592). National Institute of Justice Research in Brief. San Francisco, California: US Department of Justice.