The effort within the social science research literature that seeks to preserve the duality of typifying IPV as a singular construct while simultaneously delineating behavioral variance has resulted in two broad, overlapping categorizations – clinical and relational.  Clinical studies of IPV with a view towards psychopathological indicators, behavioral patterns, or genetic predispositions would tend toward a monolithic conception of IPV, with each sub-strata (e.g., stalking behavior, sexual coercion, or emotional aggression) being emblematic of a larger, internalized propensity to violence. Relational studies tend to view IPV sub-strata “more as a dysfunction of the interactional and relational processes of courtship and relationship evolution, rather than a disorder of the individual’s attachment system” (Spitzberg & Cupach, 2003, p. 348).

Researchers such as Johnson (2008) and Stark (2007) have advanced such typological delineation under the label of Controlling Coercive Violence (CCV). CCV, which as a model of both clinical traits and relational characteristics unites previously disparate sub-strata of violent or aggressive behaviors under a definitional “pattern of emotionally abusive intimidation, coercion, and control coupled with physical violence against partners” (Kelly & Johnson, 2008, p. 478).

Specifically included in this review are the findings from research relative to prevalence rates for emotional abuse/control, sexual coercion, stalking, combined physical assaults and emotional abuse/control, sexual coercion and stalking, and controlling/coercive violence or combined rates of physical and non-physical abuse, what Michael Johnson (2008) now calls “Controlling Coercive Violence” (CCV). Studies included rates for males and females and were all conducted in industrialized English speaking countries. A differential/deviant case relational organizing framework of emotional abuse/control, sexual coercion, and stalking behavior within Johnson’s (2008) schema of CCV was used for summarizing the studies given that the assumptions of CCV are, ostensibly, that these occur WITHIN (and EVOLVE from) a once normative relational context, namely through increase in incidence or prevalence. A variety of search engines were used to identify empirical work including PsychINFO, WebofScience, ERIC, EBSCO (Social Work Abstracts; Criminal Justice Abstracts), ProQuest Research (Social Services Abstracts; Sociological Abstracts), and Google Scholar.
 
More than 300 studies were reviewed, 204 studies met the identified criteria and are included in the extensive tabular listing of the reviewed research that can be found on-line at the Partner Abuse journal website. Included in the summary tables is the full reference for the study (authors, year, title, and journal information), the sample size of the study, characteristics of the sample (e.g., gender, socio-demographic information, age range), study method and design, measures used, and resulting conclusions. Studies excluded from this review dealt with non-English speaking populations and had some translation as part of the methodology, were unrelated to the purview of the current manuscript (animal abuse relative to IPV, etc.), or covered in other Partner Abuse State of Knowledge summaries.
 
Notable findings derived from this review are reported for each of the three aspects of CCV. For emotional abuse, prevalence rates might average around 80%, with 40% of women and 32% of men reporting expressive aggression (i.e., verbal abuse or emotional violence in response to some agitating or aggravating circumstance), and 41% of women and 43% of men reporting some form of coercive control. For sexual coercion, national samples demonstrated the widest disparity by gender of victim, with 0.2% of men and 4.5% of women endorsing forced sexual intercourse by a partner. By far the largest selection of highly variable studies, stalking and obsessive behaviors showed a range from 4.1% to 8% of women and 0.5 to 2% of men in the United States have been stalked at some time in their life. Women were reported as having a significantly higher prevalence (7%) of stalking victimization than men (2%).  For all types of violence except being followed in a way that frightened them, strangers were the most common perpetrators, as reported in approximately 80% of cases. Overall, intimate stalkers seem to comprise somewhere between one-third and one half of all stalkers.  In stalking involving obsessive following behaviors, women were most often victimized by men they knew, most frequently their current or former intimate partners.  Among women who reported repeated unwanted contact, current (15.9%) and former (32.9%) intimate partners were the perpetrators in nearly half of the most recent incidents and the largest subdivision of reports came from college or university student samples. Within studies of stalking and obsessive behaviors, gender differences are much less when all types of obsessive pursuit behaviors are considered, but more skewed toward female victims when the focus is on stalking

Four broad conclusions can be drawn from this review.

1. As Johnson (2008) contends, there does seem to be two types of IPV, one which traditionally manifests in physical violence, and another, more nebulous, multifaceted, or perhaps stochastic type that comports with elements of power, control and coercion. This review demonstrates that the two types are not, however, altogether conterminous.
2.  IPV and CCV behaviors are generally supported by international studies from other English-speaking nations. None of the CCV sub-strata under review showed marked variance in the U.K., Australia, Canada, or New Zealand studies contained as part of the literature reviewed.
3. Studies of IPV tend to vary, as Spitzberg and Cupach (2003) noted, in their perspective of combined coercive violence as having a clinical or relational locus. This can also be clearly evinced in the overlap of studies in this review from one aspect of CCV behavior and another (i.e., emotional abuse and stalking behavior). The data reported from single studies seemed to capture elements of both discrete CCV categories, which, while separately significant to the purposes of this review, complicates the ability to categorize a single study as capturing only one discrete form of IPV. 
4. With regard to emotional and psychological abuse, sexual coercion, and stalking, studies which utilized the same or similar methodologies, instrumentation, and measurement reported a much higher variance, as seen with the community-based and clinical studies, than those which diversified the means of obtaining respondent data. In interview or mixed-methodological studies, variances in prevalence were often comparable and, in many cases, much higher.  

This review highlights the need for increases in reportage, adjudication, and assessments of prevalence. With regard to policies, advocates, practitioners, and researchers alike must sound the call for uniform definitions, legislation, and law enforcement standards that specifically address emotional and psychological abuse, sexual coercion, and stalking behaviors. Uniform policies are the stable basis from which victim-appropriate, empirically rigorous and accurate prevalence studies are conducted, interventions designed, and programs initiated.

Research into the three facets of CCV in this review have uncovered several potential areas of interest for researchers and avenues for further development where only one or a few studies have been conducted to date, but which reported significant findings. Recommendations for further research are grouped in the manuscript according to the CCV category to which they apply.