The Brief Partner Behavioral Control Scale (B-PBCS): Development and Validation

Despite the large growth of close relationships literature, partner behavioral control has been rarely the primary focus of research. It is also evident that the instruments for measuring partner behavioral control are scarce. Thus, the present study aims to develop and validate a 9-item Brief Partner Behavioral Control Scale (B-PBCS). The validity and reliability of the B-PBCS were examined in a sample of emerging adults (N = 511; 81.8% females; Rangeage = 18-25) who are currently in a premarital romantic relationship. To assess psychometric properties of the B-PBCS, we conducted a series of tests examining factor structure (exploratory and confirmatory factor analyses), measurement invariance (configural and metric invariance), reliability (internal consistency and test-retest reliability), and predictive validity. Results demonstrated that the B-PBCS has a two-factor solution (overt and covert partner behavioral control), acceptable measurement invariance, internal consistency, test-retest reliability, and predictive validity. The overt partner behavioral control at baseline assessment has been linked to romantic relationship conflict and psychological distress at 14-week follow-up assessment even after controlling for sex, age, and relationship duration. In conclusion, the B-PBCS offers a valid and brief measure for assessing partner behavioral control.


Introduction
Over the last two decades, a large body of research has focused on understanding the nature and significance of romantic relation-ships in emerging adulthood. As a distinct period from adolescence and young adulthood, emerging adulthood, the period from the late teens through the late twenties, is characterized as the age of identity explorations, instability, self-focus, feeling in-between, and possibilities (Arnett, 2000). In the 21 st century industrialized countries, with changing societal trends and expectations, marriage rates tend to decrease and the mean age at first marriage tends to increase gradually (Eurostat, 2019). Instead of committing to adult roles and responsibilities, today's youth spend more time in romance and love before settling on permanent relationships (Arnett, 2000(Arnett, , 2004(Arnett, , 2007. Therefore, forming, maintaining, and ending romantic relationships appears as a critical developmental task in emerging adults' lives (Snyder, 2006). Achievement of this developmental task is believed to satisfy intimacy needs (Fincham & Cui, 2011), contribute to finding a meaningful life (Mayseless & Keren, 2014), and associate with progress in adopting adult roles and responsibilities (Barry, Madsen, Nelson, Carroll, & Badger, 2009).

Partner Behavioral Control and Its Expanded Operationalization
Partner behavioral control has already been conceptualized as the demands, directives, proscriptions, and prescriptions that a partner places on his/her intimate partner's behaviors (Rohner, 2005a(Rohner, , 2005b. Partner behavioral control refers to the quality of relationship from a permissive structure to enforcing a strict limit for restrictions on partner's behaviors. However, in our conceptualization, partner behavioral control involves not only overt (i.e., trying to change partner's behaviors) but also covert behavioral control (i.e., attempts to know what a partner does on a daily routine). We operationalize the overt partner behavioral control as limiting the partner's behaviors in a way that can be perceived by the romantic partner. For example, trying to change the decisions and behaviors of the partner is a prominent indicator of overt partner behavioral control. As a second dimension, covert partner behavioral control reflects control behaviors, which are undetected by the partner but still result in restrictions and limitations. For instance, attempts of a partner to demand from the other where he or she spends time, with whom and what does s/he do during the day would be presented as indicators of covert partner behavioral control. Furthermore, covert partner behavioral control, as a closely related construct to the partner monitoring, includes tracking the daily behaviors of the partner. In covert partner behavioral control, by monitoring daily activities of the partner, one tries to create a rule over his or her romantic partner. However, to date, partner monitoring has been frequently addressed in social networking site research (e.g., Belu, Lee, & Sullivan, 2016;Fox & Tokunaga, 2015;Van Ouytsel, Walrave, Ponnet, Willems, & Van Dam, 2019) and we still do not have strong evidence regarding the effects of partner monitoring in real life settings. Therefore, based on broader conceptualization of partner behavioral control, it seems important to fill the gap and develop a new valid and reliable instrument incorporating these two dimensions in the assessment of partner behavioral control. Furthermore, it is also crucial that assessing how overt and covert partner behavioral control differentially predicts individual and relational outcomes in couples would further advance close relationships literature.

Existing Partner Behavioral Control Measures and Empirical Findings
Researchers usually examine partner behavioral control either on the basis of Interpersonal Acceptance-Rejection Theory (IPARTheory; Rohner, 2014) or on the basis of Intimate Partner Violence and Abuse (IPVA; Fincham, Cui, Braithwaite, & Pasley, 2008). The IPAR-Theory, as an evidence-based theory of socialization and life span development, attempts to understand the antecedents and outcomes of interpersonal acceptance/rejection and control (Rohner, 2014(Rohner, , 2016. The IPARTheory suggests that the degree of acceptance, rejection, and control experienced in interpersonal relationships is significantly and panculturally related to individuals' personality dispositions and psychological adjustment (Rohner, 2016). Parental Acceptance-Rejection/Control Questionnaire (PARQ/Control; Rohner, 2005a) and Intimate Partner Acceptance-Rejection/Con-trol Questionnaire (IPAR/CQ; Rohner, 2005b) are two main self-report questionnaires used widely in IPARTheory research. While PARQ/ Control assesses individuals' perceptions of maternal and paternal accepting-rejecting and controlling behaviors, IPAR/CQ measures the same dimensions with reference to intimate partners. The results of an overview of 12 meta-analyses summarizing results from 31 countries and 149 440 participants (Khaleque & Ali, 2017) support the basic postulates of IPARTheory and confirm the cross-cultural robustness of the scales used in IPARTheory research. However, it must be noted that existing studies (e.g., Geitsidou & Giovazolias, 2016;Khaleque, Shirin, & Uddin, 2013;Malik & Rohner, 2016;Rohner, Melendez, & Kraimer-Rickaby, 2008) almost exclusively have focused on the warmth (acceptance-rejection) dimension of PARQ/Control and IPAR/CQ. There are comparatively fewer studies (e.g., Khaleque, 2012) assessing the effects of intimate partner behavioral control on individuals' psychological and relational outcomes. It is also evident that although data have been collected with regard to control dimension of PARQ/Control and/or IPAR/CQ, researchers usually did not use this dimension in the subsequent analyses (e.g., Khaleque & Rohner, 2013;Khaleque, Rohner, & Laukkala, 2008;Malik & Rohner, 2016). Nevertheless, limited empirical evidence (e.g., Khaleque, 2012) indicates that there is a significant negative link between perceived intimate partner behavioral control and psychological adjustment.
Furthermore, controlling behaviors perpetrated in romantic relationships are also addressed in the intimate partner violence and abuse (IPVA) research. IPVA includes violent and abusive physical, sexual, and/or psychological behaviors directed toward current or former partners (Fanslow & Robinson, 2011;Walker, Sleath, & Tramontano, 2021). Scholars (Arroya, Lundahl, Butters, Vanderloo, & Wood, 80 Studia Psychologica, Vol. 63, No. 1, 2021, 77-93 2017) argue that a desire to maintain power or exert control over a partner usually lies in the root of IPVA. As a serious public health issue, IPVA affects millions of people and occurs across the lifespan (Centers for Disease Control and Prevention, 2018). According to the World Health Organization's (WHO) technical report (Garcia-Moreno et al., 2013), the global prevalence of intimate partner violence (physical and/or sexual IPV) is 30%. Prevalence estimates are lowest in the high-income region (23.2%) and highest in South-East Asia (37.7%) and Eastern Mediterranean (37%) regions. Actually, recent findings (Lövestad, Löve, Vaez, & Krantz, 2017) highlight that as a form of IPVA, the prevalence of controlling behaviors is much more common than physical and sexual violence in intimate relationships. The Intimate Partner Violence Attitude Scale-Revised (IPVAS-Revised; Fincham et al., 2008), Revised Conflict Tactics Scale (Straus, Hamby, Boney-McCoy, & Sugarman, 1996), Revised Controlling Behaviours Scale (CBS-R; Graham-Kevan & Archer, 2005), and Multidimensional Measure of Emotional Abuse (MMEA; Murphy & Hoover, 1999) appear to be the most frequently used instruments in the IPVA literature. Nevertheless, it should be stressed that these measures focus mainly on assessing acts of violence and abuse within dating, cohabiting, and/or marital relationships. As controlling behaviors have been recently defined in the context of IPVA, there is limited number of measures assessing such behaviors comprehensively in the intimate relationships (Sleath, Walker, & Tramontano, 2018). For example, the CBS-R (Graham-Kevan & Archer, 2005) involves items related to economic, threatening, intimidating, emotional, and isolating behaviors. However, it is seen that the CBS-R assesses controlling behaviors in the context of coercive control.
Taken together, based on previous literature, it is seen that partner behavioral control has been investigated mostly either within the IPARTheory or IPVA literature. While IPARTheory focuses mostly on the effects of partner acceptance and rejection, IPVA literature focuses mainly on assessing acts of assault, threats, humiliation, intimidation or abuse of one partner over other. In addition, it is observed that while partner behavioral control is extensively assessed in the context of coercive partner behaviors in the IPVA literature, in the IPARTheory the partner behavioral control dimension is dropped from further analyses. Therefore, it may be concluded that there is a need for a new and psychometrically robust measure assessing partner behavioral control from a permissive to a strict structure in romantic relationships.
Moreover, it appears that there is a need for a broader conceptualization of partner behavioral control in close relationships literature. In our view, partner behavioral control not only includes overt but also covert acts. Developing an expanded measure of partner behavioral control in terms of overt and covert control and assesing how these two dimensions differentially predict relational and individual outcomes would further contribute to existing literature.

Current Study
Given that previous research has either overlooked the behavioral control dimension or has provided limited number of findings regarding the role of partner behavioral control on relational and individual functioning (Khaleque, 2001;Khaleque et al., 2008;Parmar & Rohner, 2005;, 2010, the present study aimed to design a new questionnaire (The Brief Partner Behavioral Control Scale; B-PBCS) that may briefly and robustly capture the construct of partner behavioral control. By developing a brief, valid, and reliable instrument that targets the controlling behaviors experienced in romantic relationships, this research may expand future studies on close relationships. Understanding antecedents and outcomes of partner behavioral control would be an important step in developing intervention and prevention programs aimed at enhancing relationship functioning of couples.

Participants and Procedure
Turkish emerging adults (N = 511; 81.8% females), who were in a romantic relationship and unmarried took part in the research. We split overall research sample randomly into two subsamples. The first sample consisted of 256 participants (205 females, Range age = 18-25, M age = 20.98, SD age = 1.65) and was used for exploratory factor analysis. The relationship duration ranged from 1 month to 120 months (10 years), with a mean length of 26.22 months (SD = 24.96). The second sample consisted of 255 participants (213 females, Range age = 18-25, M age = 21, SD age = 1.68) and was used for confirmatory factor analysis. The relationship duration ranged from 1 month to 111 months (almost 9 years), with a mean length of 23.96 months (SD = 20.61). Of the total sample who had completed the surveys at baseline assessment (Time 1), 182 who were continuing the relationship with the same partner from Time 1 completed the surveys at the 14-week follow-up assessment (Time 2). Based on this data set, we calculated measurement invariance, test-retest reliability, and predictive validity of the B-PBCS. The data used in the current study was a portion of the Turkish Longitudinal Romantic Relationship Study (TLRRS; Sağkal & Özdemir, 2020). The participants were recruited from undergraduate education and psychology courses at a public university in western Tur-key during the spring semester of 2018. Prior to paper-and-pencil survey administration, we explained the nature of the research and obtained participants' informed consent. The research protocols were conducted according to the guidelines of Institutional Ethics Board Committee.

Development of the Brief Partner Behavioral Control Scale (B-PBCS)
The B-PBCS is a brief measure developed to assess perceptions related to partner's behavioral control. Partner behavioral control was conceptualized in two dimensions: (i) overt partner behavioral control and (ii) covert partner behavioral control. While overt partner behavioral control refers to the demands, directives, proscriptions, and prescriptions that a partner directly place on his/her intimate partner's behaviors, covert partner behavioral control refers to forms of behavioral control, which are masked in socially acceptable ways and undetected by the partner. Based on the Interpersonal Acceptance-Rejection Theory (IPARTheory), Intimate Partner Acceptance-Rejection/Control Questionnaire (IPAR/CQ), IPVA literature, IPVA measures as well as the cultural relevance of the items for Turkish emerging adults, we generated a total of nine items that capture facets of partner behavioral control. Items are rated on a three-point Likert scale (1 = Not like him/her, 2 = Somewhat like him/her, and 3 = A lot like him/her) and higher scores from the scale represent restrictive partner behavioral control. Instructions are as follows: "The following questions are about your experiences with your romantic partner. There are no right or wrong answers. Please choose the answer that best describes your relationship." Following item generation, to increase the content validity, three counseling researchers independently examined the rel-evance and representativeness of the B-PBCS items. Lastly, we administered the target scale to a small number of participants (n = 30) to test comprehensibility and unambiguousness of the items. After revisions and subsequent modifications, the nine-item initial pool was retained for empirical testing.

Instruments
In order to assess predictive validity of the B-PBCS, we examined the links between B-PBCS, conflict in romantic relationship, and psychological distress. Participants who were continuing the relationship with the same partner from baseline assessment (Time 1) completed the surveys at the 14-week follow-up assessment (Time 2).
Conflict in romantic relationship. Conflict in romantic relationship was measured with the conflict properties dimension of the Children's Perception of Interparental Conflict Scale (CPIC; Grych, Seid, & Fincham, 1992;Ulu & Fışıloğlu, 2004). Although the original scale aims to measure perceptions regarding interparental conflict, in the present study we reworded the items so that participants responded to items with reference to romantic relationship conflict (e.g., "Even after we stop arguing, we stay mad at each other"). The scale consists of three subscales, namely, frequency, intensity, and resolution. Participants rated items on a 3-point Likert scale ranging from 1 = True to 3 = False, with higher scores indicating frequent, intense, and unresolved romantic relationship conflict. Previous research (Sağkal & Özdemir, 2019) demonstrated that the Turkish version, which assesses conflict in romantic relationship, has good validity and reliability. In the present study, the Cronbach's alpha coefficients were .84, .77, .76, and .90 for frequency, intensity, resolution, and total scale, respectively.
Psychological distress. Psychological distress was measured with the Kessler Psycho-logical Distress Scale (K10; Kessler et al., 2002). K10 is a unidimensional, 10-item, self-report instrument designed to assess psychological distress. Items are responded to on a 5-point Likert scale ranging from 1 = None of the time to 5 = All of the time, with higher scores indicating greater psychological distress. The validity and reliability of the Turkish version of the K10 has been confirmed in previous studies (Sağkal & Özdemir, 2017;Sağkal, Özdemir, & Koruklu, 2018). In the present study, Cronbach's alpha was .94.

Data Analysis
Data analyses were conducted using SPSS version 24.0 and AMOS version 24.0. The total sample was randomly split into two subsamples (Sample 1, n = 256; Sample 2, n = 255) in order to cross-validate the factor structure of the B-PBCS. We first conducted an exploratory factor analysis (EFA) with principal component analysis (PCA) and oblique rotation. Bartlett's test of sphericity and the Kaiser-Meyer-Olkin (KMO) measure were used to assess sampling adequacy of data for the factor analysis of the B-PBCS items. The eigenvalue > 1.0 criterion and the scree-plot were used to determine the number of factors to be retained in the model (Tabachnick & Fidell, 2007). Variables with factor loadings ≥ .40, communality values ≥ .50, and cross-loadings ≤ .32 were retained in the analyses (Worthington & Whittaker, 2006). To verify the two-factor structure identified in the EFA, we next conducted a confirmatory factor analysis using maximum likelihood estimation. The Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI), and the Root Mean Square Error of Approximation (RMSEA) were used to evaluate the goodnessof-fit of the proposed model. The TLI and CFI values above .90 and the RMSEA values below .06 were considered evidence for acceptable fit (Byrne, 2010). Furthermore, to assure the two-factor structure associated with the same items and the same meanings across time, we tested configural and metric invariance of the B-PBCS. We performed chi-square difference test (Δχ 2 ) to evaluate the relative fit of the nested models. For the reliability of the B-PBCS, we computed Cronbach's alpha and 14-week test-retest reliability coefficients. We also investigated the inter-item correlations and corrected item-total correlations of the B-PBCS. Lastly, we assessed predictive validity by investigating the links between partner behavioral control (overt and covert partner behavioral control), romantic relationship conflict, and psychological distress. A latent path model was used to estimate the predictive role of partner behavioral control at Time 1 (baseline assessment) in romantic relationship conflict and psychological distress at Time 2 (14-week follow-up assessment).

Factor Structure
Exploratory Factor Analysis (EFA). The first sample (256 participants, Range age = 18-25, M age = 20.98, SD age = 1.65) was used for the exploratory factor analysis. The results of the Kaiser-Meyer-Olkin measure (.84) and the Bartlett's test of sphericity (χ 2 (36) = 1101.590, p < .001) indicated that the adequacy of the data set for factor analysis is appropriate. The underlying structure of the B-PBCS was evaluated through a principal component analysis with oblique rotation (direct oblimin). Inspection of the eigenvalues and the scree plot suggested a two-factor structure for the B-PBCS items. The eigenvalues for the first and second factors were 4.097 and 1.852, respectively. The first factor consisted of 5 items that capture the perceptions related to a partner's attempts at controlling his/her partner's daily activities (e.g, My partner wants to know where I am and what I do during the day) and was labeled "covert partner behavioral control". The second factor consisted of 4 items that capture the perceptions concerning a partner's attempts at controlling his/her partner's behaviors (e.g., My partner always tries to change my behaviors) and was labeled "overt partner behavioral control". Factor loadings ranged from .81 to .86 for covert partner behavioral control and from .73 to .80 for overt partner behavioral control. Findings revealed that all the items evidenced strong loadings on their respective factors without cross-loading problems. Communalities were between .50 and .75 (see also Table 1). The two-factor model accounted for 66.10% of the variance (dimensions of covert and overt partner behavioral control contribute 45.52% and 20.58%, respectively).
Confirmatory Factor Analysis (CFA). The factor structure of the 9-item B-PBCS was also examined through the application of CFA on the data from the second sample (255 participants, Range age = 18-25, M age = 21, SD age = 1.68). Maximum likelihood estimation was used to test the two-factor model. The initial CFA results showed that model fit indices, except TLI and RMSEA, were adequate for the two-factor solution: χ 2 (26) = 105.970, χ 2 /df = 4.076, p < .001, CFI = .91, TLI = .88, and RM-SEA = .11 [.09, .13]. The modification indices suggested the addition of an error covariance term between items 1 and 3 ('My partner wants to know who my friends are' and 'My partner wants to know whom I spend time with') to improve model fit. As these two items overlap in content (loading on the same factor) and adding modification substantially improves model fit, we allowed the error variances of items 1 and 3 to correlate. After adding covariance, the reestimated model yielded good fit for the two-factor solution: χ 2 (25) = 34.547, χ 2 /df = 1.382, p > .05, CFI = .99, TLI = .98, and RMSEA = .04 [.00, .07]. For the 9-item B-PBCS, standardized factor loadings ranged from .52 to .84 for covert partner behavioral control and from .59 to .79 for overt partner behavioral control (see Figure 1). All the t values were significantly different from zero at the .001 level. The factor correlation was .42 and significant at the p < .001 level.

Reliability
Internal consistency reliability. The internal consistency reliability for the B-PBCS was calculated for both subscales and total scale in each sample. The Cronbach's alpha values were .90 and .86 for covert partner behavioral control, .76 and .76 for overt partner behavioral control, and .84 and .82 for total scale in Samples 1 and 2, respectively. Inter-item correlation matrixes revealed that the inter-item correlations were moderate to high in magnitude within factors in both data sets. Corrected-item total correlations ranged from .68 to .78 and .57 to .74 for covert partner behavioral control and ranged from .46 to .60 and .53 to .64 for overt partner behavioral control in Samples 1 and 2, respectively. As a result, all these findings indicate that the B-PBCS has a good level of internal consistency.
Test-retest reliability. In order to assess the test-retest reliability, the B-PBCS was administered to the sample across a 14-week time in-terval. Test-retest reliability coefficients were calculated .59 (p < .01, n = 182) for covert partner behavioral control, .69 for overt partner behavioral control (p < .01, n = 182), and .63 (p < .01, n = 182) for total scale. The results demonstrated that the B-PBCS scores are stable over time.

Predictive Validity
Using structural path modeling and data from 182 emerging adults, we examined the predictive role of the B-PBCS measured at Time 1 (baseline assessment) in romantic relationship conflict and psychological distress measured at Time 2 (14-week follow-up assessment). The latent predictor variables were covert and overt partner behavioral control, and the latent outcome variables were romantic relationship conflict and psychological distress in the model. The latent variables of covert and overt partner behavioral control were represented by corresponding B-PBCS items. The romantic relationship conflict latent variable was represented by frequency, intensity, and resolution subscales. Psychological distress latent variable was represented by three parcels. Parcels were created using an item-to-construct balance approach (Little, Cunningham, Shahar, & Widaman, 2002). Sex, age, and relationship duration were used as covariates in the analyses. The measurement model, comprising four latent variables with 15 indicators, indicated a good fit to the data: χ 2 (83) = 111.817, p < .05, χ 2 /df = 1.35, CFI = .98, TLI = .97, and RMSEA = .04 CI [.02, .06]. The factor loadings ranged from .57 to .88 for covert partner behavioral control, .64 to .82 for overt partner behavioral control, .67 to .76 for romantic relationship conflict, .90 to .94 for psychological distress, and all were statistically significant (p < .001). The fit indices for structural model showed a good fit to the data as well: χ 2 (118) = 204.531, p < .001, χ 2 /df = 1.73, CFI = .94, TLI = .92, and RMSEA = .06 CI [.05, .08]. The results of the path analysis are depicted in Figure 2. As expected, overt partner behavioral control at Time 1 had a statistically significant positive effect on romantic relationship conflict (β = .51, p < .001) and psychological distress (β = .22,  Figure 2 Structural links between the covert partner behavioral control, overt partner behavioral control, romantic relationship conflict, and psychological distress (n = 182).
Note. The covariates (gender, age, relationship duration) were included in the analyses but are omitted to enhance clarity of the figure. Standardized regression weights are presented. T1 = Time 1. T2 = Time 2. ** p < .01, *** p < .001 p < .01) at Time 2. Counter to expectations, covert partner behavioral control at Time 1 was not significantly linked with romantic relationship conflict (β = .03, p > .05) and psychological distress (β = -.04, p > .05) at Time 2. The squared multiple correlations indicated that the perceived partner behavioral control at Time 1 accounted for 32% of the variance in romantic relationship conflict and 11% of the variance in psychological distress at Time 2 in emerging adults. Taken together, the results confirmed the predictive validity of the B-PBCS on individual and relational processes.

Discussion
The primary purpose of the present study was to develop and validate the psychometric properties of the Brief Partner Behavioral Control Scale (B-PBCS). Results from the research indicate that the B-PBCS is a brief, valid, and reliable measure for assessing partner behavioral control in premarital romantic relationships of emerging adults. The findings obtained from exploratory and confirmatory factor analyses confirmed that the B-PBCS involves two distinct dimensions: overt and covert partner behavioral control. The results of the measurement invariance analyses demonstrate that the factor structure and loadings of the B-PBCS is the same across time. Cronbach's alpha and test-retest reliability indices ensure that the B-PBCS has stable and reliable psychometric properties. The results also confirm the predictive validity of the B-PBCS.
Particularly, in the present research, we found that overt partner behavioral control was a significant predictor of both romantic relationship conflict and psychological distress over time (14-week). Indeed, this finding was consistent with Khaleque's (2012) finding that perceived partner behavioral control associates with lower levels of psychological adjustment. However, unexpectedly, covert partner behavioral control was not a significant predictor of either relationship conflict or psychological distress. Based on relationship maintenance literature, some scholars (Goodboy, Dainton, Borzea, & Goldman, 2017) emphasize that allowing control could be used as a relationship maintenance strategy to keep the less committed partner in the relationship. As such, allowing covert partner behavioral control serves the function of keeping the relationship in existence. Dainton and Gross (2008) suggest that romantic partners experiencing passionate love are also likely to allow control in order to maintain passion in their relationships. However, Dainton and Gross (2008) also caution that although partners using such strategies meet their personal needs and/or feel better about their relationships, negative maintenance behaviors (e.g., allowing control) involve some level of dysfunction. Moreover, from the perspective of self-determination theory (Deci & Ryan, 2014), it would also be likely that partners allowing covert behavioral control may be using this relationship maintenance strategy in order to fulfill their basic psychological need satisfaction. Finally, the perception of behavioral control would be seen as a social construction to some degree, and thus it is likely that partners possess distinct interpretations of its nature. Particularly, covert control behaviors including tracking the daily behaviors and routines of the partner are likely to be perceived as a manifestation of interest, warmth, and love by the partner, particularly considering the collectivistic nature of Turkish culture. Indeed, empirical evidence (e.g., Toplu-Demirtaş, Murray, & Hatipoğlu-Sümer, 2019) indicates that aggressive/controlling behaviors are highly prevalent in Turkish emerging adults' dating relationships and they are commonly perceived as an indicator of possessiveness in the romantic relation-ships (Toplu-Demirtaş & Fincham, 2018). Toplu-Demirtaş, Hatipoğlu, and Fincham's (2017) study also supports that abusive behaviors like jealousy (doing something just to make partner jealous) are perceived as a sign of love by Turkish young adults. As a result, the present finding regarding the nonsignificant links between covert partner behavioral control and outcome variables (romantic relationship conflict and psychological distress) could be influenced by individual, relational as well as cultural factors. But, further research is needed to elaborate on this finding.
The present study extends the partner behavioral control research in important ways. Although scholars (Dush & Amato, 2005;La Guardia & Patrick, 2008;Patrick et al., 2007;Sağkal & Özdemir, 2020) have endorsed the significance of romantic interactions on individual and relational outcomes, limited research has paid attention to the construct of partner behavioral control in the close relationships literature. In addition, the existing measures (e.g., Fincham et al., 2008;Graham-Kevan & Archer, 2005;Rohner, 2005b) either mostly addressed overt partner behavioral control or targeted coercive behaviors. However, to the best of our knowledge, no research to date has captured the covert control dimension in the assessment of partner behavioral control. Thus, the present study, offering a brief and comprehensive measure of partner behavioral control, provides a significant step in addressing this gap in the literature. In addition, the B-PBCS, providing a framework for identifying patterns of partner behavioral control, hopefully inspires future research in the close relationships field.
The results of this study can also help counselors and therapists to improve their assessment and treatment protocols in the helping process. Based on Rigazio-DiGilio, Ivey, Kunkler-Peck, and Grady's (2005) descriptions, the B-PBCS including the features of having emphasis on perceived behaviors, subjective experiences, unrecognized topics, and multiple perspectives seems to meet several criteria needed to be identified as a measure best adapted for clinical use. The B-PBCS as a self-report survey assesses the extent to which partners are perceived to place overt and covert limits, restrictions, proscriptions, and/or prescriptions on behaviors. By using the B-PBCS, therapists could help couples to examine the quality of relationships they currently experience with one another. The B-PBCS would also help couples to explore the effects of perceived partner behavioral control on the intrapsychic and interpersonal problems that led them to seek a professional treatment. According to Rigazio-Digilio and Rohner (2015), with the use of microcounseling skills, therapists could invite partners to understand their subjective perceptions and experiences regarding partner behavioral control, determine therapeutic goals, and establish a collaborative relationship for a therapeutic change. Based on Rigazio-Digilio and Rohner's explanations, we recommend therapists to use the B-PBCS as a brief and psychometrically sound assessment tool in therapy sessions and interpret the results interactively with their clients. We believe that using such an interactive and collaborative clinical application of the B-PBCS promises to increase therapeutic alliance and shared responsibility between therapists and their clients and to promote treatment processes and outcomes in therapeutic interventions.
The present study has some limitations that need to be considered when interpreting the findings as well. First of all, the psychometric properties of the B-PBCS were tested with emerging adults living in a western city of Turkey. Further research is needed to validate the scale across different regions of Turkey. Moreover, in order to ensure the generalizability of the findings, researchers are encour-aged to validate the B-PBCS across cultures in future studies. Secondly, in this study, we assessed the psychometric properties of the B-PBCS in premarital romantic relationships of emerging adults. But, partner behavioral control is also experienced in all relationship types. Therefore, future studies would be well-served to test the psychometric properties of the B-PBCS across different relationship types, such as cohabiting and/or marital relationships. Thirdly, as the data were not dyadic, we only tried to understand the actor effects of partner behavioral control on relationship conflict and psychological distress. By using a dyadic research design, future studies could also aim to explore the cross-partner effects. Furthermore, future researchers could also longitudinally investigate the antecedents and outcomes of partner behavioral control. Finally, as the sample comprised a disproportionate number of females, further research is needed to conduct measurement invariance of the B-PBCS across gender.
In conclusion, the present study is the first to validate a brief and comprehensive measure of partner behavioral control. The empirical findings obtained from exploratory and confirmatory factor analyses, measurement invariance tests, Cronbach's alpha and test-retest reliability analyses as well as predictive validity study highlight that B-PBCS is a brief, valid, and reliable measure of partner behavioral control. These results suggest that researchers and practitioners can validly and reliably utilize the B-PBCS in assessing perceived partner behavioral control in premarital romantic relationships of emerging adults. The present study provides an avenue for research, training, and practice. But, further exploration of the construct of partner behavioral control is also needed to make greater contribution to the field of close relationships.