shared parenting: critical review of the research literature

SHARED PARENTING: CRITICAL REVIEW OF THE RESEARCH LITERATURE

Michael Benjamin, Ph.D.* Howard H. Irving, Ph.D**

This critical review of the shred parenting research literature highlights the need for sociolegal reform. This article was submitted to the Meyer Elkin Essay Contest and was selected for publication because of its quality and relevance.

INTRODUCTION

In recent years, traditional practices regarding child custody have been called into question. Presently, a preponderance of all custody awards in the United States and Canada involve a maternal sole custody arrangement (Furstenberg & Nord, 1985; Kanoy & Cunningham, 1984; McKie et al., 1983). An increasing proportion of the remainder involve a joint custody or shared parenting arrangement in which care and control of the child(ren) are shared equally between both parents (Folberg & Graham, 1981).

As of 1984, Folberg (1984a, 3984b) notes that over 30 states had legislation which recognized, presumed or mandated shared parenting. Since then, additional states have or are con- templating such legislation (Robinson, 1985; Coller, 1988). Such statutes take two forms: legal shared parenting in which care and control is shaed, but with a primary residence- typically maternal; and, a physical shared parenting in which there is no primary residence, the child(ren) moving between households. Of all custody awards in the United States, between 2% and 30% involve a shared parenting arrangement (Fursten- berg et al., 1983; Furstenberg & Nord, 1985; Phear et al., 1984; Weitzman, 1985), most of the legal shared parenting type (Coller, 1988).

In contrast, Canada has resisted shared parenting, with case law indicating considerable hesitance on the part of the judiciary (Payne, 1979; Payne & Patrick, 1979). However, as Ingram (1989) observes, “(d)uring (sic) the past decade both trial and appellate courts have ordered or allowed joint custody with increasing frequency,” even against joint parental opposition. Consequently, while current legislation (in Ontario, Manitoba, New Brunswick & Prince Edward Island) allows that custody

* Michael Benjamin is a family sociologist and works as a research consultant, professor and is research coordinator of the Student- Environment Study Group at the University of Guelph, Guelph, Ontanio, Canada. Howard H. Irving is a professor of social work at the University

of Toronto and is a well known authority in family mediation, which he practices, teaches and in which he does research.

**

be awarded to more than one person; in only the Yukon Temtory is shared parenting formally recognized (Ingram, 1989). While the incidence of shared parenting in Canada is unknown (McKie et al., 1983), anecdotal reports (Irving & Benjamin, 1987) suggest an increasing preference among divorcing parents.

As for the literature, interest in shared parenting has had two consequences: increase in the number of articles debating the merits of shared parenting, both in favor (eg. Robinson, 1983; Salji & Cassady, 1982) and opposed (Goldstein et al., 1973; Allen, 1984); and increase in the number of articles reviewing the research literature (Alexrod & Everett, 1983; Clingempeel & Repucci, 1982; Coller, 1988; Derdeyn & Scott, 1984; Scott & Derdeyn, 1984; Stahl, 1984; Steinman, 1983). The former highlight the salient issues. The latter should systematize the literature, subject it to critical appraisal, and integrate what is known about the subject. We say “should” because existing reviews have failed to do justice to the shared parenting research literature in at least three respects: omitted available reports; given little attention to theoretical or methodological issues, focusing instead on practice or policy; and, taken a global approach, thus ignoring much important detail.

Accordingly, this article reviews the 21 published studies that presently constitute the shared parenting research literature. For an earlier version, see Benjamin and Irving, (1985). Their discussion is divided into four sections concerned, re- spectively, with theory, methodology, analysis, and substantive results. We conclude by exploring implications for theory, research and sociolegal policy.

THEORY

One of the prime functions of scientific research is to develop theory which explains selected natural phenomena. This may be explicit or, as in literature about shared parenting, left implicit. In neither case, however, can the effort escape the constraints of theory. As Harris (1970) explains, “(t)heoretical conceptions permeate the entire process of thinking and facts always require interpretation, so that no sharp distinction can be drawn between theory and observation ....”

Accordingly, inspection of the shared parenting research literature indicates that authors operate within one of three theoretical frames of reference. The first involves a clinical model characterized by: (1) concern with intrapsychic phenomena, especially affective states; (2) interest in the discovery of pathology, particularly individual “adjustment”; (3) reliance

21

FAMILY AND CONCILIATION COURTS REVIEW/VOLUME 27, NUMBER 2/DECEMBER 1989

upon on qualitative data, typically the clinical interview; and (4) use of the individual as the unit of analysis.

The second model may be characterized as social psychological in character (Wehster, 1975). This approach is characterized by: (1) concern with intrapsychic states, especially sub- jective representations of role relations (eg. parental attach- ment); (2) some interest in interpersonal behavior, albeit from the individual’s perspective; (3) heavy reliance on quantitative data, especially the use of standardized instruments; and (4) use of the individual as the unit of analysis.

The final model is that of family systems theory (Benjamin, 1983). This approach is characterized by: ( 1 ) concern with interpersonal behavior, including associated affective and cog- nitive processes; ( 2 ) use of both qualitative and quantitative methods; and (3) reliance on a dyadic or triadic unit of analysis.

The distribution of authors by their theoretical preference is displayed in Table 1. Most authors favor a social psychological approach, with the clinical approach next in frequency and the systems approach last. This distribution is significant, for it immediately delineates the sorts of issues and questions that are deemed relevant and those that are excluded from examination. For example, it suggests that the primary unit of analysis is the individual respondent and his or her perception of and response to particular people and situations is the prime subject of inquiry. From a systemic perspective, these analytic concerns are highly problematic, with individual responses seen to reflect the larger family system of which respondents are members and within which their behavior is meaningful. It Sollows that, in the absence of data regarding that larger context, individual responses become difficult to interpret.

METHODOLOGY

If the development of theory is one Cornerstone of scientific research, the other concerns the scientific method (Denzin, 1978). Operationally, methodology translates into a concern for study design and data analysis. Study design, in turn: involves study sample (size, recruitment procedure), operational definition of shared parenting (i.e., sampling criteria), and the presence or absence of one or more comparison groups. Additional concerns specific to the shared parenting debate include the use of clinical or normative samples, and whether shared parenting groups involved legal or physical shared parenting. These data are displayed in Table 2.

Several aspects of this table stand out. First, with few exceptions, sample sizes tend to be small, generally ranging from 5-50; with the exception of Irving and Benjamin. studies with large samples used “cases” drawn either from court or clinic records. The problem with “case” data is that, originally constructed for judicial or clinical purposes, they may be subject to a variety of unspecified biases.

Second, study designs are markedly heterogeneous, vary- ing in terms of the parental status of the respondents, the presence or absence of children (typically under-represented),

the recruiting method used, and the presence or absence of a comparison group. Such variation is important because it reduces study comparability.

Third, heterogeneity is again evident in terms of the operational definition of shared parenting, with the same result. In some cases, shared parenting was self-defined, although closer examination of the two studies in question indicates that additional criteria (such as dual residences, the presence of minor children, separated or divorced parents) were given the status of default values. More often, definition by the courts was used. This is problematic, since legal definitions of shared parenting tend to vary across jurisdictions (Folberg, 1984b: Ingram, 1989). Most often, authors developed their own idiosyncratic selection criteria, which varied from the default values noted above (Ahurbanel, 1979) to rather esoteric definitions {Stein- marz, 1983).

Fourth, most authors have relied on normative samples, with respondents in most samples either involved in a physical shared parenting arrangement or with more than 50% of respondents so involved. The emphasis on physical shared parenting is noteworthy for two reasons. First, in the United States, i t misrepresents the preponderance of legal shared parenting cases, and thus what shared parenting really is in practice (Weitzman, 1985). Second, studies of legal shared parenting samples (Phew et al., 1984), due to their reliance on secondary data, have been quite limited in the questions addressed the only exception is an unpublished study by Walsh and Kalter (1986) using a normative sample.

Finally, only nine of the 21 studies compare shared parenting and sole custody groups. As important, several of these studies are questionable as regards their ”fairness.” Those emp1oying“case” data (Iyeld et al., 1982: Phear et al., 1984) are, as previously noted, both limited in extent and vulnerable to charges of bias. Several studies focus exclusively on fathers (Gersick, 1979; White & Bloom, 1981;D’Andrea,1983;Luepnitr. 1982), a problematic choice since the experience of mothers and fathers differs markedly between shared parenting and sole custody samples (Inling & Benjamin, 1986). Similarly, only three studies (Shiller, 198686; Luepnitz, 1982; It-ving & Berz- jamirz. 1986) explore the experience of shared parenting and sole custody children. Finally, the work of Awad, Luepnitz, and Irving and Benjamin suggest shared parenting and sole custody groups are not internally homogeneous; rather, subtypes within each group may reliably be differentiated, in turn, calling in question the remaining studies which failed to make such distinctions.

These design characteristics render this literature metho- dologically suspect. Small sample size, for example, tends to limit generalizability while heterogeneity reduces comparability and reliability (Campbell & Stanley, 1963). Similarly, infre- quent use of comparison groups, and the dearth of studies that both compare across groups and differentiate subtypes within groups all serve to weaken claims regarding the viability or efficacy of the shared parenting alternative.

22

ANALYSIS

Given a sample of respondents, the next step in the research process involves collecting and analyzing the data(Anderson & Zelditch, 1968). The data collection and analysis procedures in use in the shared parenting literature are displayed in Table 3.

This demonstrates that data analysis in shared parenting research concentrates on “weak” statistical measures and relies inordinately on the interview method. Of the 21 studies examined, only seven employ any statistical test, typically the chi square test. While the latter is better than simple means or percentages, it can be problematic in the absence of measures of strength (such as gamma or lambda) without which spurious findings may be given weight (Nie et al., 1975).

Similarly, clinical or questionnaire-guided interviews remain a standard means of data collection; only nine studies supplement interviews with data from standardized instruments. While we have no quarrel with the interview method, it should be noted that such data are vulnerable to chaiges of bias. It follows that the use of independent measures can only strengthen whatever claims are advanced.

Here, it is noteworthy that the utility of “triangulation” (Denzin, 1978; see Baker- & Schulberg, 1973) to guard against bias has been well demonstrated. Conversely, variability across interviewers using clinical techniques can be reliably demonstrated, (eg. Sartorius et al., 1978). It thus appears that, taken as a whole, data analysis within the shared parenting literature can fairly be characterized as “weak.”

RESULTS

Our final concern is with the findings of shared parenting research. This involves two issues: the substantive areas of inquiry that are the foci of data collection; and, the findings in these areas. The first issue is important, for it speaks to the degree of consensus concerning the matters deserving attention. The second issue is important for a different reason. Given the methodological deficiencies in this literature, it is difficult for any single study to demonstrate the viability or efficacy of shared parenting. The same may not be true of the literature taken as a whole. From this perspective, methodological heterogeneity is (paradoxically) an advantage: if similar findings are consistently reported despite methodological variation, than these findings may be given considerable weight.

The data regarding areas of inquiry are displayed in Table 4. This indicates that attention is primarily centered on six topic areas. Further, with the exception of “satisfaction,” coverage of these areas is markedly uneven. Noteworthy here is the relative scarcity of studies concerning child adjustment; of 21 studies, only eight report such data and, of these, only five provide data taken directly from child respondents. Such scant interest in the children is remarkable.

As for the substantive findings, their summarization, below, is organized into the same six topic areas plus an “other” category.

Parental Adjustment Shared-parenting parents have been examined in one of

three ways: their social and psychological adjustment to separation or divorce; their feelings about themselves as persons and as parents; and, the extent to which they differ on either basis from sole-custody parents.

In all three respects, there was considerable consensus. Separation or divorce was universally experienced as a difficult, stressful, and emotionally painful experience. For this reason, parents who encountered no adjustment problems following the dissolution of their marriage were rare, although the issues perceived by parents as problems tended to vary by sex. Fathers, for example, reported feelings of guilt, anger, and depression as well as loneliness and a sense of loss. Mothers had to cope with many of the same feelings as well as the need to find employment (if they had not already done so) and the danger of becoming overburdened with simultaneous vocational and child care responsibilities.

Further, Gersick notes that the more fathers felt betrayed by the divorce, the more likely they were to seek child custody for the “wrong” reasons: anger and revenge. Irving and Benjamin report similar fiidings among “dissatisfied” shared parenting parents, especially mothers. Grief found that as the frequency of father-child contact increased, the rapidity of fathers’ emotional adjustment also increased. Related evidence comes from Roth- berg, and Irving and Benjamin both of whom found, first, that cordial relations among former spouses promoted parental adjustment, and, second, the involvement of new intimate partners was either neutral or promoted adjustment.

Shared-parenting parents appeared to fair somewhat better than their sole-custody counterparts. On the one hand, this relates to parental attributes prior to divorce. D’Andrea, and Steinman et al., for example, found higher self-esteem among shared parenting than custody parents. On the other hand, it appears directly attributable to the child custody arrangement. Several authors (Irving & Benjamin, 1984; Steinman et al., 1985; Shiller, 1986) found shared parenting as opposed to sole custody parents reported less anger, guilt, and depression, and were more satisfied with the continuity of parent-child and parent-parent contact. While most fathers with physical shared parenting had to make some adjustment to new parenting responsibilities, both their former wives and their children were available to assist them. Similarly, while mothers continued as before, the burdens associated with child care were now shared, providing them with more time alone.

Exceptions to these findings are reported. White and Bloom report that their shared parenting fathers adjusted poorly compared to sole custody custodial or non-custodial fathers. Given a sample size of n=3, however, it is hard to know how much weight to give these findings. Richards and Goldenberg report that fathers involved in physical shared parenting had problems with role overload, and hostile relations with their former spouses. In addition to caveats associated with sample size (n=10), the author’s findings are qualified spousal hostility did not interfere with father-child relations which, in turn, reduced oretiminatedany sense of loss orabandonment. Finally, McKinnon

23

and Wallerstein report both poor adjustment among shared- parenting parents, and a direct relationship between parental and child adjustment. However, these findings derive from a clinical sample who may be expected to have more problems than their non-clinical counterparts. Indeed, the authors themselves support such an interpretation by agreeing with Irving et al. (1984) concerning the existence dissatisfied shared parenting parents.

Child(ren) Two findings are consistent among the few studies which

report child data: a minority of children exhibit some degree of difficulty in adjusting to the demands of shared parenting; and, the proportion of such children and the severity of their problems were never worse than those reported by sole custody children, and were sometimes better.

For example, several authors note ‘‘good‘’ shared parenting child adjustment coupled with “high” shared parenting child satisfaction and only “minor” transition difficulties (Abar- banel, 1979; Nehls, 1980; Watson, 1981). The difficulties expected to arise from moving between households and living with parents having divergent lifestyles simply did not materi- alize in most cases.

Of the minority of shared-parenting children who experienced difficulty, Steinman found that 30% (11-32) were “hyper- loyal,” and 25% were confused by residential transition; in general, 30% felt “over-burdened‘’ by the demands of shared parenting. Shiller, who compared 20 shared parenting and 20 sole-custody boys, found that while both groups were above test norms on the Achenbach Child Behavior Checklist, shared- parenting children exhibited fewer problems, a finding consistent with parent and teacher reports. McKinnon and Waller- stein, who explored the experience of 26 preschool shared parenting children, found that four out of seven (57%) of the youngest children, and three out of 19 (16%) of the older children showed evidence of “distress,” with more severe distress among the older children. Finally, Irving and Ben- jamin, relying on parental reports, showed that, first, only a minority of shared parenting children were dissatisfied (10%- 25%); and, second, parental perception of child satisfaction was directly related to parental satisfaction with shared parenting.

Four factors appeared to predict child adjustment: continuity of parent-child interaction; regularity and predictability of the child movement schedule; coparental cooperation; and, child age (eg. older children were more likely than younger children to report distress associated with disruption of their peer relations and with enduring fantasies of parental reconciliation). While there is consensus as to the relevance of these factors, some disagreement persists as regards their relative salience. Luepnitz, for example, found that a low-conflict family climate was critical to the adjustment of the 91 children in her sample and that this, in turn, was independent of custody type.

Finally, as regards the comparative merits of shared parenting versus sole custody, authors using normative or clinical samples drew related conclusions. Among the latter, McKin-

non and Wallerstein suggest just as no evidence demonstrates shared parenting protects children from the stress of divorce, no evidence demonstrates children are better served by sole custody. Among the former, both Luepnitz, and Irving and Ben- jamin suggest shared parenting at its best serves the end of child adjustment better than sole custody at its best. Irving and Benjamin add shared parenting is not a panacea, for both shared parenting and sole custody can fail; under such circumstances, neither serves the best interests of the children. The primary advantage of shared parenting, based on available evidence, is it is less likely to fail, and thus provides a positive post-divorce experience to a greater proportion of the children involved in it.

Co-Parental Relationships Both shared parenting and sole custody involve the interac-

tion among spouses whose previous relationship has been such as to cause them to divorce. This suggests the possibility on- going mutual dissatisfaction will undermine their ability to cooperate with respect to child care.

Here, there is considerable agreement co-parental cooperation is more likely among shared parenting than sole custody parents. Abarbanel, for example, notes shared parenting parental agreement around scheduling, child rearing values, and mutual support. Ahrons, and Rothberg reports shared parenting joint decision-making, information sharing and general cordiality. Steinman found shared-parenting parents tolerant of differences and respectful of each other’s parenting skills, with mutual anger kept away from the children (see also Richards andGoldenberg, 1986). Similarly, comparative studies, such as Steinman et al., and Irving and Benjamin report less conflict and greater mutuality and cooperation among shared-parenting as opposed to sole-custody parents.

Related findings are reported by Ilfeld et al. who found, over a two year period, shared-parenting parents had a relitigation rate (16%) half that of the sole custody counterparts (32%) (Luepnitz, 1982). Even when shared parenting was ordered without the consent of both parents, theirrelitigation rate (33%) was no different than that of sole custody parents. Similarly, Irving and Benjamin report shared parenting parents were more likely to arrive at their custody arrangement through mutual agreement, and, once in place, were more likely to alter its terms informally rather than through litigation.

Exceptions include Frankel, and McKinnon and Waller- stein both of whom report intense conflict among shared parenting clinical parents, findings whose generalizability is likely limited to the 17% of dissatisfied shared parenting parents reported by Irving and Benjamin (Benjamin & Irving, 1989 in press). The work of Phear et al. is also exceptional: shared parenting and sole custody parents exhibited similarrelitigation rates. The basis for the difference between these findings and those of Ilfeld et al. is not clear, although variation in the distribution of shared parenting subtypes may reduce their comparability. Nearly all of Phear et al.’s cases, based in Massachusetts (shared parenting optional), involved legal shared parenting whereas Ilfeld et al.’s cases, based in California where shared parenting is given greater emphasis, may have

24

involved an unknown but possibility greater proportion of physical shared parenting.

Finally, cordiality among shared-parenting parents appears related to three factors: low pre-divorce conflict; intense com- mitment to child rearing; and, continuity in parent-child involvement. The first factor implies a typological difference between shared parenting and many sole custody families (Ahrons & Rodgers, 1987; Kressel et al., 1980; Vaughan, 1986). It may also relate to the finding shared parenting parents reported fewer and less severe pre-divorce marital problems, and higher education than sole custody parents (Irving & Benjamin, 1987). The second factor suggests greater mutuality regarding the salience and direction of child rearing values, what Irving and Benjamin call a “child orientation.” The third factor may be the most important, for it eliminates care and control of the children as a bone of contention, especially among sole custody parents (Irving & Benjamin, 1987).

Parent-Child Relations Three aspects of the parent-child relationship have been the

focus of research. The fmt, already noted, concerns the greater l ikel ihd of a “child orientation” among shared parenting as opposed to sole custody parents (Irving & Benjamin, 1987), further evidenced by the fact that choice of custody option was more likely mutual among shared parenting parents (Watson, 1981; Gersick, 1979).

The second is an extension of the fist-level of parenting involvement, especially among fathers. Here, there is near unanimity shared-parenting fathers were highly involved in parenting prior to the divorce (GrieJ 1979; Richards & Golden- berg, 1986), and that this was more likely of shared-parenting as opposed to sole-custody fathers (Irving & Benjamin). In addition, Shiller notes that shared-parenting mothers were more likely to perceive their former spouses as understanding and supportive, and to respect their parenting skills.

Finally, the third issue concerns the level of parenting involvement following divorce. Again, near unanimky is ap- parent: shared parenting fathers demonstrate high levels of postdivorce parenting involvement (GrieJ 1979); the level of involvement is significantly greater than that evidence by sole custody fathers (Irving & Benjamin, 1986: 65% vs 23%; Bowman & Ahrons, 1985); and, continuity of parenting involvement appears to be the norm (GrieJ 1979). Further, D’Andrea notes shared-parenting as opposed to sole-custody parents were more knowledgeable about their children, more influential in their lives, and more satisfied with their level of child influence; the likelihood of physical shared parenting was directly related to the level of parental involvement. Concomi- tantly, both Luepnitz, and Irving and Benjamin note the greater likelihood shared parenting mothers report a reduction in their child care burden. Finally, Shiller (1986b) found that shared- parenting children were likely to feel comfortable expressing their feelings (both positive and negative) while less preoccu- pied by reconciliation fantasies.

Logistics Shared parenting is a logistically complex custody arrange-

ment-including the movement of children between residences, variations in scheduling, and the sharing of parenting responsibilities-which invariably places great demands on parents who share parenting.

Research has focused on three concerns: the nature of the child-sharing schedule; the extent to which it is a source of conflict; and, the identification of factors which may facilitate or impede the demands associated with.

The expectation that physical shared parenting necessarily involves a 50/50 split has received little support. Rather, child- sharing schedules vary widely (Irving & Benjamin, 1986). Indeed, Ingram (1989) notes two Canadian cases in which shared parenting was awarded despite the fact that the parents in question lived on two different continents. This said, a weekly schedule is the single most common arrangement (eg. Abar- banel, 1979; Ahrons, 1981; Luepnitz, 1982; Richards & Gold- enherg, 1986; Shiller, 1986; Steinman et al., 1985; Watson, 1981).

Further, there is general agreement around two additional findings: scheduling is the most contentious issue among shared parenting parents (Nehls, 1980; Rothberg, 1983); and, shared parenting parents are characterized by flexibility and informal problem solving (exception: McKinnon & Wallerstein, 1987); conflict over scheduling seldom involves litigation (Aharbanel, 1979; A hrons, I981 ; Steinman, 1985). Given the intensity of conflict typical among sole custody parents (Irving & Ben- jamin, 1987), it is noteworthy shared parenting parents were significantly less likely than sole custody parents to perceive scheduling as a “serious problem”(Irving and Benjamin, 1986; see Benjamin & Irving, 1989 in press).

Finally, there is agreement that residential proximity is a central facilitating factor (Abarbanel, 1979; McKinnon & Wallerstein, 1987; Steinman, 1983; Watson, 1981; Rothberg, 1983). Further, Irving and Benjamin found residential proximity was directly related to the absence of a primary residence, that is, to the choice of physical as opposed to legal shared parenting (no primary residence: 55% “c~os~,” 33% “far”). While important, this trend should not be construed to mean that residential distance contraindicates shared parenting. On the contrary, Irving and Benjamin report that a minority of shared parenting parents (and most of their children) who lived some distance apart were quite “satisfied” with their particular arrangement. Moreover, the benefits of proximity for the shared parenting arrangement may involve other, hidden costs, including limits to vocational advancement for both parents (Richards & Goldenberg, 1985).

Satisfaction The issue of satisfaction among shared-parenting parents

and children has been widely explored. The consensus is that both shared-parenting parents and children report a “high” degree of satisfaction, with proportions ranging from 67% (Rothberg, 1983) to 84% (Ahrons, 1980; Irving & Benjamin, 1989).

25

In addition, shared-parenting parents (especially fathers) were significantly more satisfied than soie-custody parents (Ahrons, 1980; D’Andrea, 1983; Griei 1979). While Irving and Benjamin found no difference in overall satisfaction, shared- parenting parents reported many fewer problems and problems of lesser severity. They note too shared-parenting fathers were significantly more satisfied than their sole-custody counterparts (87% vs 64%), and that those able to establish a working arrangement in a year or less were much more likely to be satisfied than those requiring more than a year and/or who had yet to achieve a working arrangement. Further, child satisfaction was highly correlated with parent satisfaction.

Finally, shared-parenting parents and children were often satisfied for different reasons: parents due to joint decision- making and positive child postdivorce adjustment; children due to continuity of peer relations, reduction of tension, and the sense that they were no longer part of the spousal battle. There was, however, one reason in which both parents and children concurred parenting continuity. For fathers, this meant continuity of parental involvement and parent-child interaction. For children, it meant feeling wanted and having two parents. Consequently, the majority of shared-parenting parents and children recommended shared parenting (Rothberg, 1983; Iw- ing &Benjamin, 1986).

Other Finally, three issues are noteworthy, but fit neatly into none

of the above categories: compliance as regards child and spousal support payments; the long-term durability of shared parenting; and typological variation among shared-parenting families.

Among women with sole custody, absent or irregular support payments are acommon problem (Weitzman, 1985; W-aller- stein & Blakeslee, 1988), and contribute to poverty among a significant proportion these women and their children (Irving & Benjamin, 1987). For this very reason, anumber of jurisdictions across North America have established programs in which paternal support payments are made directly to the court which then passes them on to the sole-custody mothers in question. Concomitantly, the court is empowered to pursue fathers who default or are in arrears.

In contrast, most studies of shared-parenting parents report consistent paternal support (Ahrons, 1979; Luepnitz, 1986; Steinman, 1985), although it could be a source of conflict (Rothberg, 1983). Irving and Benjamin found that shared- parenting were more likely than sole-custody parents to report “consistent” support (fathers: 88% vs 61 %; mothers: 83% vs 6 1 %). The exception, Phear et al., report no difference between legal shared-parenting and sole-custody parents as regard relitigation centered on financial matters.

In a related vein, Patterson (1982), using population data concerning household costs, projected additional expenses ranging from $12,354-$15,617 associated with shared parenting. This implies shared parenting may be restricted to middle-to-upper- class families. By contrast, Irving et al. (1984) note that one third of their sample reported annual incomes below $20,000; family income and outcome satisfaction were unrelated among

shared-parenting or sole-custody families. In the absence of longitudinal data, the absolute or relative

durability of single-parenting arrangements remain unclear. Only scattered data exist. Steinman et al., who assessed the “success” rate of 48 shared-parenting families after one year of a three year study, found that 27% were successful, 42% were “stressed” while 31% had “failed.” Within one year, 75% of Frankel’s “cases” returned to court, while McKinnon and Wallerstein found that coparental cooperation was “rare.” All three studies are suspect, since their clinical samples were likely biased in favor of single-parenting dissatisfaction.

In contrast, Irving and Benjamin found that only a minority of shared-parenting parents were dissatisfied with their arrangement (17%) and, of these, only two families (5%) of the dissatisfied subsample, less than 1% of the total sample) had returned to court seeking sole custody.

Finally, most studies fail to distinguish typological variation within shared-parenting families. This failure is curious given recent publications in the divorce literature (eg. Ahrons & Rodgers, 1987; Kressel et al., 1980; Vaughun, 1986). A few exceptions exist. Awad distinguishes between four “types” of child custody: type 3 refers to “non-alternating joint custody” (ie., legal shared parenting while type 4 refers to “alternating joint custody” (ie., physical shared parenting). Within both types, parents in agreement or in dispute are further distinguished. While Awad’s results indicate that type 3 parents fared better than type 4, the fiidings are problematic given the small sample size ( 1211 00 were of type 4) and the clinical character of his “cases.” Similarly, Phear et al. found few differences between shared parenting and sole custody cases, with 70% virtually identical). While problematic, given the use of court records, and sparse in the range of variables examined, the data do at least call attention to the difference between legal and physical shared parenting.

W e no study has systematically compared the two groups, an unpublished study by Walsh and Kalter (1986) has compared the experience in Michigan of 30 legal (maternal) shared- parenting and 30 maternal sole-custody children, all 7-12 years of age and equally split as to gender (18 boys, 12 girls). The authors found few differences between the two groups, either among the children (both were within the normal range, with about a third of each group “depressed”) or the mothers (both groups reported equal levels of [high] coparental hostility), while the differences that were found were unrelated to custody type and could be better explained in terms of pre-divorce parental differences. This bears some resemblance to Luepnitz’s finding that child adjustment was better predicted by the quality of postdivorce coparental and parent-child relations than by custody type (physical shared parenting, paternal sole custody, maternal sole custody).

Finally, Irving and Benjamin report three findings of spe- cial interest here: first, that satisfied and dissatisfied shared parenting parents differed systematically across arange of variables (Benjamin & Irving, 1989 in press); second, satisfied as opposed to dissatisfied shared-parenting parents were more liely to live in close proximity and to report no primary residence (ie.,

26

physical shared parenting; and, thiid, typologically distinct groups emerged when custody type, gender and satisfaction were interacted. The result of this analysis was an hierarchical array, with satisfied shared-parenting mothers and fathers at the top and dissatisfied sole-custody fathers at the bottom.

DISCUSSION

Having reviewed the shared parenting research literature and summarized the substantive evidence, we will close by exploring the implications of these data in three areas: theory, research, and sociolegal practice.

Theory Of the deficiencies in the shared-parenting research litera-

ture, none is more significant than the absence of any explicit effort at theory building. Virtually none of the studies examined, including our own, seeks either to include an existing model or to derive a model by induction. This absence main- tains shared parenting research on the periphery of the social sciences and restricts the sophistication of the research designs in use. It follows such theoretical efforts should be a primary future concern of shared parenting researchers.

Not to belittle the practical concems of policy-makers, and legal and human service personnel, the desire for guidance in legal decision-making, counseling service, or fiscal expendi- ture, is important. We contend, however, theory development and these pragmatic interests are complementary. New theoretical formulations can only contribute to the interests of service personnel by reducing the level of “indeterminacy” in the social sciences which Mnookin (1975) correctly sees as particularly problematic for policy makers and human service personnel.

Research If theoretical deficits are most signifcant, then methodo-

logical deficiencies are most prominent. Here, a half dozen changes are urgently required.

First, the ability to generalize is hampered by small sample sizes, the absence of measures of reliability and validity, and reliance on basic methods of data analysis. All of this needs to change, as this field of study traverses the same course as the divorce literature, from exploratory to analytical efforts.

Second, both family and couples research tend to get short shrift. The typical use of individual measures cannot subse- quently be employed as ad hoc measures of family process without serious distortion (Reiss, 1981).

Similarly, more effort is required in selecting samples of couples rather than data derived from “parents” or “mother/ fathers.” The importance of such work is underscored by Ahrons (1981) who reports that divorced spouses not only maintain an on-going relationship, but that their respective perceptions of that relationship tend to diverge.

Third, as Irving and Benjamin, and Steinman et al. demonstrate, parent and child adjustment takes time. This points to the

salience of longitudinal studies. Interaction processes, for example, may vary significantly over time as a function of child and parent development, the addition of new family members (eg. birth, remarriage) or their loss (eg. serial divorce, departure of older children), the complications associated with forming binuclear family systems, and even statutory changes as shared parenting families increase in prevalence. Only Ahrons (1980b) is pursuing this line of inquiry. More such work is needed.

Fourth, if couples data can only be derived from research samples containing both ex-spouses, then the same is true with respect to child data. Too often, researchers have taken the easy (and less costly) route, relying on secondary parental reports rather than primary child data. This omission is curious, for the research instruments with which to conduct such efforts are plentiful (Kanoy & Cunningham, 1984).

A related concern is raised the merits of self-report measures versus those based on third-party observation. The former are widely favored over the latter. This choice is defensible. The experimental approach is much more complex than the use of a questionnaire or the completion of a face-to-face interview. That observational methods should be less widely used than its self-report counterpart is thus hardly surprising. Whether the results are comparable is another issue. Apart from the pitfalls attendant on questionnaire or interview methodologies (Sum- r-ious et al., 19781, it remains doubtful whether family members can accurately report on their interactional processes when they themselves may not be consciously aware of them. It follows that observation may be the only means to obtain such data.

Fifth, the comparison-group design has been little used in shared-parenting research. This is understandable in early exploratory work; single-group designs are no longer defensible. Concomitantly, future concern should not concern the viability of shared parenting-this has already been demonstrated-but rather should address the more challenging task of examining the extent to which shared parenting and sole custody are similar or different. Such specificity will require two forms of comparison-group design. The first will involve the comparison of shared-parenting and sole-custody groups matched on demographic and social indicators. The second, more potent design will f i t identify specific shared-parenting (eg. contested vs non-contested) and sole-custody (eg. maternal vs paternal custody) variants and then compare them, both across and within matched groups. This approach would seek to identify the intervening and predictive variables with respect to both.

Finally, whereas existing studies are hypothesis-generat- ing, more attention is required to make such hypotheses explicit and to subject them to empirical test. Accumulation in science does not consist merely of additional fmdings, but includes a selective process which eliminates unprofitable lines of inquiry and classes of research question. Such on-going refmement of formulations and designs profits us all by reducing wasted effort and making it unnecessary for new researchers to rein- vent the wheel.

Turning to analysis, we note an inordinate reliance on single methods and descriptive statistics. Both trends need to change.

27

Here, we echo Clingempeel and Repuccis’s (1982) call for more studies using multioperational and multidimensional methods.

In this context, a word about the intersection between statistical methods and theoretical formulations: all statistical procedures make specific assumptions about linear causality. Since these accord with the assumptions underlying the social psychological formulation, there is little trouble about “fit” between the two. Such is not the case with family system formulations. Its recursive assumptions are simply not ame- nable to existing statistical procedures such that authors (such as Ahrons, 1979; Irving & Benjamin, 1989) are forced to employ linear causal models in their statistical analysis. This supports the use of multioperationalism, for only some combi- nation of quantitative and qualitative methods will provide the data to substantiate the claims of systems theory proponents.

Similarly, the interpretation of results relates to the data base in use. Shared-parenting reviewers (Derdeyn & Scott, 1984; Steinman, 1983) have typically focused on the individual study. Moving the analysis up a level, to the body of literature taken as a whole, yields a different result. Here, the literature’s heterogeneity becomes an advantage. The logic of triangulation holds that if studies using divergent samples, focusing on different variables and employing different methods all gener- ate similarfindings, then confidence is warranted in the empirical trends thus revealed. Despite our caveats regarding methodological variation in shared-parenting research literature, reliable trends in the substantive findings are unmistakable, in turn providing a secure empirical base from which to draw conclusions.

Sociolegal Policy The reliable trends noted above include: ( 1 ) high coparental

agreement regarding chil- rearing values and the salience of child rearing; (2) high parental involvement in child rearing; (3) high parenting continuity; (4) good postdivorce parental adjustment; (5) good postdivorce child adjustment; (6) high coparental cooperation regarding child care issues; (7) high compliance regarding scheduling and support; (8) informal resolution of disputed issues, including a low rate of relitigation; and, (9) high parental and child satisfaction.

These trends point to an optimal pattern of postdivorce spousal and parent-child relating. While this pattern does not describe all shared-parenting families, it does establish an ideal against which all custodial arrangements may be assessed. In turn, we suggest that existing sociolegal policy be consistent with such an ideal. At present, relevant statutes vary widely across North American jurisdictions. In light of the above review, shift to a uniform sociolegal policy should having the following features:

Implementation of a (1) rebuttable ( 2 ) presumption of shared parenting involving three components: (3) a detailed and specific parenting plan, (4) a specified trial period, and (5 ) the routine provision of support services during the latter period.

We conclude by elaborating on these changes.

Presumption of Shared Parenting Available evidence consistently indicates that most parents

and children adjust well to a shared-parenting arrangement. Consequently, they make fewer demands on the state in terms of court or other costs associated with failure to provide child and spousal support. Moreover, comparative data indicates shared parenting at its best is at least as good and often better than sole custody. However, as Coller (1988) notes, these data notwithstanding, mostjudges prefer sole custody (Kelly, 1984). The answer, Coller suggests, and we and others (eg. Everett, 1984; Schwartz, 1984) agree, is to limit judicial discretion by creating a statutory presumption of shared parenting.

Presumption Rebuttable While most shared parenting families adjust well, a minor-

ity do not. Further, shared parenting with specific subgroups of divorcing parents would conflict with the principle of child best interests. Accordingly, the presumption of shared parenting must be rebuttable, with statutes specifying classes of parental conduct that would exempt them from the presumption. Here, Elkin (1987) and Coller (1988) tender the following list: inabil- ity to care for the child(ren), whether mentally, emotionally, or physically; significant substance abuse; physical abuse of spouse or child; intractable overt hostility over time between spouses despite mediation; great geographic distance in the case of very small children when frequent changes of residences are unman- ageable; and, expressed desire of parents not to participate in joint custody. We concur, but find the last two items problematic.

Parenting Plan Debate often centers on the viability or efficacy of shared

parenting versus sole custody. Data reviewed indicate that shared parenting is both viable and comparatively more effica- cious than sole custody. However, this line of argument misses the point, for the data also show that physical shared parenting, legal shared parenting and sole custody (with or without access) also serve the best interests of some children. Thus, the critical issue concerns creating a parenting plan which best matches the needs of the child(ren) in question (compare with Weitzman, 1 985).

In turn, this suggests a preferred hierarchy of parenting arrangements or plans, with both physical and legal shared parenting fist, legal shared parenting only second, and sole custody (with or without access) third. Matching plan to family should be guided by both best interest and the above parenting ideal: maximum parental involvement, parenting continuity, and mutual decision-making regarding child care.

Parenting plans, while practical, also address an ideological objective: to replace the legal notion of “custody” with one which highlights the integral character of the parent-child relationship. Parents do not own their children, and so should not be forced to struggle over who gets custody of them (Everett, 1984). Rather, what is at stake is the relationship between family members and the sort of postdivorce relationship that serves the children’s best interests. It is that relation-

28

ship that is specified by aparenting plan, not the strength of each parent’s claim to child ownership (“custody”).

Trial Period Related issues concern (1) whether shared parenting should

be restricted to voluntary couples or extended to those in which one or both partners prefer sole custody, and (2) how to deal with inappropriate behavior by one or both parents sharing parenting.

Minimal evidence concerning the first issue suggests that the extended option may operate as well or better than sole custody. Recall Ilfeld et al.’s finding that the relitigation rate of shared-parenting and sole-custody parents were no different. In our own work, while dissatisfied shared-parenting and sole- custody fathers did not differ in most respects, the former alone maintained parenting continuity (Kelly, 1983). Further, Wil- liams (1988), based on his clinical experience, argues that, “joint custody provides one of the best methods of stimulating a degree of significant and meaningful cooperation in warring parents.. . ,” adding “that cooperation should not be a criteria for joint custody vs. sole custody schedules for children.” h our experience, the same applies to geographic distance, with child best interest and parental involvement central criteria, not child age or frequency of child movement between residences.

Thus, imposing shared parenting on reluctant parents will, at worst, likely do no more harm than sole custody while offering some of them an unexpectedly positive custody experience. It is likely, however, these couples will be at higher risk than voluntary couples. Accordingly, all couples should enter shared parenting for a trial period of between six and 12 months. At the end of that time, their experience would be reviewed and shared parenting either made permanent, modified or aban- doned altogether.

As to inappropriate parental behavior, a trial period would provide the opportunity to adjust terms and conditions in response to parental complaints, with the Court available to adjudicate intractable conflict. More importantly, with the exception of contraindicated couples, shared parenting should fist be seen to fail before sole custody is regarded as an appropriate option.

Support Finally, having become involved in shared parenting, couples

are more or less on their own. With respect to voluntary couples, this poses no problem; within 12 months most will have a working arrangement. For “at risk” couples, the absence of support may prove more problematic. In turn, this data suggest shared parenting be routinely coupled with some form of support. While several options come to mind, our own bias would favor the use of family mediators (Irving & Benjamin, 1987). They would serve three functions: monitor the couples’ progress during the trial period; intervene as needed (for example, help them work through conflict); and, report to the Court describing the operation of the shared parenting plan and recommend its continuation, modification or replacement.

29

TABLE 1

THEORETICAL MODELS IMPLICIT IN THE SHARED PARENTING RESEARCH LITERATURE.

THEORETICAL MODELS

Author(#

Keshet2 & Rosenthal

AbarbaneP

Ahrons4

Gersick?

Griefs

Nehls6

Steinman

Watson

White & Bloom

Ilfeld et al.

Awad

Rothberg

D’Andrea3

Phear et al.

Steinman et a1.7

Frankel

Richards* & Goldenberg

Shille139

Luepnitzio

McKinnon’ ’ & Wallerstein

Irvingi2 & Benjamin

Year

1978

1979

1979/80a

1979

1979a

1979

1981

1981

1981

1982

1983

1983

1983

1984

1985

1985

1986

1986a

1986

1987

1989

Clinical S.Psych. Fam.Sy s.

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

W

KEY: S.Psych=social psychological; Fam.Sys.=family systems.

NOTES: 1 Shared parenting has been the subject of graduate work research; see Shavin (1976 cited in Grief, 1979a), Pojman (1981 cited in Ilfeld et al., 1982),

2 See Rosenthal & Keshet (1978,1981). 3 Study taken from author’s Ph.D. thesis. 4 See Bowman and Ahrons (1985). 5 See Grief (I979h). 6 See Morgenbesser & Nehls (1981) and Nehls & Morgenbesser (1980). 7 See Zemmelman et al. (1987). 8 See Richard & Goldenberg (1985). 9 See Shiller (1986b).

10 See Luepnitz (1982). 11 See also Kolata (1988). 12 See Irving et al., 1984; Irving & Benjamin, 1986; Benjamin & Irving (in press).

Nunan (1980 cited in Richard & Goldenberg (1986), Cowen (1982 cited in Shiller, 1986b) and Handley (n.d. cited in Ingram, 1989).

30

TABLE 2

RESEARCH DESIGN IN THE SHARED PARENTING RESEARCH LITERATURE.

RESEARCH METHODS

Author($

Keshet & Rosenthal

Abarbanel

Ahrons

Gersick

Grief

Nehls

Steinman

Watson

White & Bloom

Ilfeld et al.

Awad

Rothberg

D’Andrea

Phear et al.

Steinman et al.

Frankel

Richards & Goldenberg

Shilller

Luepnitz

McKinnon & Wallerstein

Irving & Benjamin

Sample’ Subjects

29/128 Fa

4 Fam

41 Pa

15/40 Fa

8/40 Fa

1% Pa

24 Fam

1 1/17Pa3

3/40 Fa

138/414 Ca

100 Ca

30 Pa

24/46 Fa

109/500 Ca

48 Pa

32 Pa

10 Fa

20140 Pa

1 1 143 Fa5

25 Fam

20 1 /395 Pa

Recruit

N/I

N/I court

court

Snowball

court

Snowball

court

court

court

Clinic

Snowball

Snowball

court

Clinic

Clinic

court

Snowball

Snowball

Clinic

Snowball

CL

N

N

N

N

N

N

N

N

N

N

Y

N

N

N

Y

Y

Y

N

N

Y

N

Define s c a

.-¤ - - B

- - B

- - B

- - B

- .- - - B

- B -

- - B

- B -

- -. - - B

- - B

- .- - -. - -. - - B

- - B

- -. - - B

B - -

cs

P

P+

L

P+

P+

P

P+

P

P+

L

P

P+

L

P+

P

P

P+

P+

P+

P+

Groups Compared

No

No

No2

Yes

No

No

No

Yes

Yes

Yes

No4

No

Yes

Yes

No

No

No

Yes

Yes

No

Yes

KEY: CL=clinical case (Y=yes, N=No); Define=author definition of shared parenting (s=defined by subject; c=defined by court; a=defined by author); CS=custody status (L=legal shared parenting [primary residence]; P=physical shared parenting [no primary residence]; P+=more than 50% of subjects physical shared parenting).

NOTES: 1

2 3 4

5

Sample size is described in terns of the number of shared parenting respondents, the larger sample total of which they were a part (if applicable), and their parental status: Fa=fathers, Pa=parent pairs, Fam=family (father, mother, child), Ca=cases (based on Court or Clinic records). Three different forms of custody are differentiated shared parenting, alternating custody, and split custody. Sample inciudes three children of the parents: one sole custody, one shared parenting, and one split custody. Two variants of shared parenting are distinguished non-altemating shared parenting (one residence, typically maternal) with or without dispute between parents, and alternating shared parenting (two residences) with or without dispute between the parents. Includes both maternal and paternal sole custody.

31

TABLE 3

DATA COLLECTION AND ANALYSIS IN THE SHARED PARENTING RESEARCH LITERATURE.

DATA COLLECTION STATISTICS

Keshet & Rosenthal

Abarbanel

Ahrons

Gersick

Grief

Nehls

Steinman

Watson

White & Bloom

Ilfeld et al.

Awad Rothberg

D’ Andrea

Phear et al.

Steinman et al.

Frankel


Shiller

Luepnitz


Irving & Benjamin

Interview‘

H b

H a

H b

H b

Ma

Mb

H b

H a

H b

H b

H b

Mb H b

H a

H b

H S

H S

H

H

H *7

*8

H S

*7

H S

*‘)

H S

Ms

N

H

DJ

H

I

H

H

H

H

H

H

H

TS4

H

* 5

*b

KEY: N=None (ie., no numerical analysis); D=descriptive only; Tx=one or more statistical tests.

NOTES: Two types of interviews are differentiated: clinical (a) or questionniare guided (b). either structured or semi-structured. Instrument(s) is taken here to mean either a standardized measure of some variable (eg. self-esteem) or one created especially for the study in question. Descriptive statistics are taken here to mean such things as mean, median, mode, standard deviation, frequency and simple percentage, with the last two the most common. Statistical tests are taken here to mean any statistical procedures which intended to meet a conventional alpha of .05, including, most typically, chi square, t test and analysis of variance (ANOVA). The author refers to the use of the Mann-Whitney “U” test, but reports no such data in the published report. The authors employ both cluster analysis and analysis of variance, but these do not apply to the shared parenting subsample due to the small sample size (n=3). No instrument was employed here other than a data collection form used to tabulate court records. The author reports his ‘‘clinical impressions” only. No instrument was employed here other than a data collection form used to tabulate psychiatric records.

32

TABLE 4

AREAS OF SUBSTANTIVE INQUIRY IN THE SHARED PARENTING RESEARCH LITERATURE.

SUBSTANTIVE AREA

Author(s)

Keshet & Rosenthal

Abarbanel

Ahrons

Gersick

Grief

Nehls

Steinman

Watson

White & Bloom

Ilfeld et al.

Awad

Rothbeg

D’Andrea

Phear et al.

Steinman et al.

Frankel


Shiller

L u e p n i tz


Irving & Benjamin

Parents

W

W

W

W

W

W W

W

W

1 W

W

W

Child’

Wd

Wi Wd

Wi

I d

Wd

W d

W i

co-P

W

W

W

W

W

W

W

W

W

W

W

W

P-c

W

W

W

W

W

W

W

W

W

Sat

W

W

W

W

w W

W

W

W

W

W

W

W

W

W

KEY: Parents=parental adjustment; Child=child(ren)’s experience; Co-P=coparental relations; P-C=parent-child interaction (eg. involvement); Log=logistics (eg. scheduling); Sat=satisfaction.

NOTE: 1 Child data have been obtained in one of two ways: directly (d), by interviewing or testing the child(ren) or indirectly ( i ) based on parental report.

33

REFERENCES

Abarbanel, A. “Shared parenting after separation and divorce: A study of joint custody.”American Journal of Orthopsychiatry 1979,49, 320-329.

Ahrons, C.R. “The coparental divorce: Preliminary research findings and policy implications.” In A.L. Milne (ed.) Joint Custody: A Handbook for Judges, Lawyers, and Counsellors. Portland, OR Association of Family Conciliation Courts, 1979.

Ahrons, C.R. “Joint custody arrangements in the postdivorce family.” Journal of Divorce 1980a, 3,189-205.

Ahrons, C.R. “Divorce: A crisis of family transition and change.” Family Relations I980b, 29 (#4), 533-540.

Ahrons, C.R. “The continuing coparental relationship between divorced spouses.” American Journal of Orthopsychiatry 198 1,5 1, 41 5-428.

Ahrons, C.R. & Rodgers, R.H. Divorced Families: A Multidiscipli- nary Developmental View. NY: Norton, 1987.

Allen, N. Joint custody: “A long awaited solution or a mere promise?’ Conciliation Courts Review 1984,22,39-48.

Anderson, T.R. & Zelditch, M. Jr. A Basic Course in Statistics: With SociologicalApplication. 2nd Ed. NY: Holt, Rinehart & Winston, 1968.

Awad, G.A. “Joint custody: Preliminary impressions.” Canadian Journal of Psychiatry 1983,28,41-44.

Axelrod, P. &Everett, C.H. “Joint custody.”In A.M. Haralambie (ed.) Handbook ofchild Custody Cases. Colorado Springs, C O Shep- ards/McGraw-Hill, 1983.

Baker, F. & Schulberg, H.L. “A system model for evaluating the changing mental hospital.” In F. Baker (ed.) Organizational Systems. Homewood, IL: Irwin, 1973.

Benjamin, M. “General systems theory, family systems theories, and family therapy: Towards an integrated model of family process.” In A. Bross (ed.) Family Therapy: A RecursiveModel of Strategic Practice. NY: Guilford, 1983.

Benjamin, M. & Irving, H.H. “Comparison of the experience of satisfied and dissatisfied shared parents.” Journal of Divorce 1989,14, in press.

Benjamin, M. & Irving, H.H. “Shared parenting: A critical review of the research.” Canadian Social Work Review 1985,84, 13-29.

Bowman, M.E. & Ahrons, C.R. “Impact of legal custody status on father’s parenting postdivorce.” Journal of Marriage & the Fam- ily 1985,47,481-488.

Campbell, D.T. & Stanley, J.C. Experimental and Quasi-Eperimen- tal Designs for Research. Chicago: Rand McNally, 1963.

Clingempeel, W.G. & Repucci, N.D. “Joint custody after divorce: Major issues and goals for research.” Psychological Bulletin

Coller, D.R. “Joint custody: Research, theory, and policy.” Family Process 1988,27,459-469.

D’Andrea, A. “Joint custody as related to paternal involvement and paternal self-esteem.” Conciliation Courts Review I983,21,8 1- 87.

1982,91, 102-127.

Denzin, N.K. The Research Act. 2nd Ed. NY: McGraw-Hill, 1978. Derdeyn, A.P. & Scott, E. “Joint custody: A critical appraisal.”

American Journal of Orthopsychiatry 1984,54, 199-209. Elkin, M. “Joint custody: Affirming that parents and families are

forever.” Social Work 1987,32, 18-24. Everett, W.J. “Shared parenthood in divorce: The parental covenant

and custody law.” Journal of Law & Religion 1984,2,85-99.

Folberg, I. “Custody overview.” In J. Foiberg (ed.) Joint Custody and Shared Parenting. Washington, D.C.: Bureau of National Affairs & Association of Family & Conciliation Courts, 1984a.

Folberg, J. “Joint custody statutes and judicial interpretations” (as of January 1, 1984). In J. Folberg (ed.) Joint Custody and Shared Parenting. Washington, D.C.: Bureau of National Affairs & Association of Family Conciliation Courts, 1984b.

Folberg, J. & Graham, M. “Joint custody of children following divorce.” In H.H. Irving (ed.) Family Law: An Interdisciplinary Perspective. Toronto: Carswell, 198 I.

Frankel, S.A. “Joint custody awards and children: A theoretical frame- work and some practical considerations.” Psychiatry 1985, 48,

Furstenberg, F.F. Jr. & Nord, C.W. “Parenting apart: Patterns of child rearing after marital disruption.” Journal of Marriage & the Family 1985,47,893-904.

Furstenberg, F.F. Jr., Peterson, J.L., Nord, C.W. & Zill, N. “The life course of children of divorce: Marital disruption and parental contact.” American Sociulogical Review 1983,48,656-668.

Gersick, K.E. “Fathers by choice: Divorced men who receive custody of their children.” In G. Levinger & O.C. Mole (eds.) Divorce and Separation: Contexts, Causes, and Consequences. NY: Basic, 1979.

Goldstein, J., Freud, A. & Solnit, A.J. Beyond the Best Interests of the Child. NY: Free Press, 1973.

Grief, J.B. “Joint custody: A sociological study.” Trial 1979a, 15 (#5), 32-33,65.

Grief, J.B. “Fathers, children, and joint custody.”American Journal of Orthopsychiatry 1979b,49,3 1 1-3 19.

Harris, E. Hypotheses and Perception. London: George Allen & Unwin, 1970

Ilfeld, F.W. Jr., Ilfeld, H.Z. & Alexander, J.R. “Does joint custody work? A first look at outcome data or relitigation.” American Journal ofpsychiatry 1982,131 (#l), 62-66.

Ingram, A.P. “Joint custody since Baker and Kruger.” Institute of Continuing Legal Education. Toronto: Canadian Bar Association (Ontario), March 4,1989.

Irving, H.H. &Benjamin, M. “Sharedparentingproject: Overview and implications.” J. Folberg (ed.) Joint Custody and Shared Parent- ing. 2nd Ed. Washington, D.C.: Bureau of National Affairs and Association of Family & Conciliation Courts, 1989, in press.

Irving, H.H. & Benjamin, M. Family Mediation: Theory and Practice of Dispute Resolution. Toronto: Carswell, 1987.

Irving, H.H. & Benjamin, M. “Shared parenting in Canada: Questions, answers, and implications.” Canadian Family Law Quarterly

Irving, H.H., Benjamin, M. & Trocme, N. “Shared parenting: An empirical analysis utilizing a large data base.” Family Process

Kanoy, K.W. & Cunningham, J.L. “Consensus or confusion in research on children and divorce: Conceptual and methodological issues.” Journat of Divorce 1984,7 (M), 45-71.

Kelly, J.B. “Further observations on joint custody.” University of California at DavisLaw Review 1983, 16,762-770.

Kelly, J.B. ”Examining resistance to joint custody.” In J. Folberg (ed.) Joint Custody and Shared Parenting. Washington, D.C.: Bureau of National Affairs & Association of Family & Conciliation Courts, 1984.

Keshet, H.F. & Rosenthal, K.M. “Fathering after marital separation.” Social Work 1978,23 (#I), 11-18.

318-328.

1986,1,79-103.

1984,23,561-569.

34

Kolata, G. “Child splitting.”Psychology Today 1988,22 (#1 l), 34-36. Kressel, K., Jaffe, N., Tuchman, B., Watson, C. & Deutsch, M. “A

typology of divorcing couples: Implications for mediation and the divorce process.” Family Process 1980, 19, 101-1 16.

Luepnitz, D. “A comparison of maternal, paternal, and joint custody: Understanding the varieties of post-divorce family life.” Journal ofDivorce 1986,9 (#3), 1-12.

Luepnitz, D.A. Child Custody: A Study of Families After Divorce. Lexington, MA: Lexington, 1982.

McKie, D.C., Prentice, B. &Reed, R. Divorce: Law and the Family in Canada. Ottawa: Statistics Canada, Research & Analysis Divi- sion, 1983.

McKinnon, R. & Wallerstein, J.S. “Joint custody and the preschool child.” Conciliation Courts Review 1987,25,39-47.

Mnookin, R.H. “Child custody adjudication: Judicial functions in the face of indeterminacy.” Law & Contemporary Problems 1975,39

Morgenbesser, M. & Nehls, N.M. Joint Custody: An Alternative for Divorcing Families. Chicago: Nelson-Hall, 1981.

Nehls, N.M. “Joint custody of children: A descriptive study.” In A.L. Milne (ed.) Joint Custody: A Handbook for Judges, Lawyers, and Counsellors. Portland, OR Association of Family Conciliation Courts, 1979.

Nehls, N.M. & Morgenbesser, M. “Joint custody: An exploration of the issues.” Family Process 1980,19, 117-125.

Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. & Bent, D.H. Statistical Package for the Social Sciences. 2ndEd. NY: McGraw- Hill, 1975.

Patterson, M. “The added cost of shared lives.” Family Advocate 1982,

Payne, J.D. “Co-parenting revisited.” Family Law Review 1979, 2,

Payne, J.D. & Patrick, J.B. “Divided opinion on joint custody.” Family LawReview 1979,2 (#3), 163-172.

Phear, W., Beck, J., Hauser, B., Clark, S. & Whitney, R. “An empirical study of custody agreements: Joint versus sole legal custody.’’ In J. Folberg (ed.) Joint Custody andShared Parenting. Washington, D.C.: Bureau of National Affair, and Association of Family & Conciliation Courts, 1984.

Reiss, D. The Family’s Construction of Reality. Cambridge, MA: Harvard University Press, 198 1.

Richards, C.A. & Goldenberg, I. “Joint custody: Current issues and implications for treatment.” Journal of Family Therapy 1985,13, 33-40.

Richards, C.A. & Goldenberg, I. “Fathers with joint physical custody of young children: A preliminary look.” American Journal of Family Therapy 1986, 14,154-162.

Robinson, H. “Joint custody: Constitutional imperatives.” Cincinnati L w Review 1985,54,27-65.

Robinson, H. “Joint custody-An idea whose time has come.”Journul ofFanzilyLaw 1983,21,64-685.

Rosenthal, K.M. & Keshet, H.F. “The impact of child care responsibilities on part-time and single fathers.” Alternative Lifestyles

Rosenthal, K.M. & Keshet, H.F. Fathers Without Partners: A Study of Fathers and the Family Afer Marital Separation. Totowa, NJ: Roman & Littlefield, 1981.

Rothberg, B. “Joint custody: Parental problems and satisfaction.” Family Process 1983,22,43-52.

(#3), 226-293.

5 (2), 10.

243-252.

1978, 1 (4), 465-491.

Salfi, D.J. & Cassaday, N. “Who owns the child? Shared parenting before and after divorce.” Conciliation Courts Review 1982,20, 3 140.

Sartorius, N., Jablensky, A., Stromgren, E. & Shapiro, R. “Validity of diagnostic concepts across cultures: A preliminary report from the international pilot study of schizophrenia.” In LC. Wynne, R.L. Cromwell & S. Mathysse (eds.) The Nature of Schizophrenia. NY: Wiley, 1978.

Schwartz, S.F.G. “Towards a presumption of joint custody.” Family Law Quarterly 1984,18,225.

Scott, E. & Derdeyn, A. “Rethinking joint custody.” Ohio State Law Journal 1984,45,455.

Shiller, V.M. “Joint versus maternal custody for families with latency age boys: Parent characteristics and child adjustment.” American Journal of Orthopsychiatry 1986a, 56,486-489.

Shiller, V.M. “Loyalty conflicts and family relationships in latency age boys: A comparison of joint and maternal custody.” Journal of Divorce 1986b, 9 (M), 17-38.

Stahl, P.M. “A review of joint and shared parenting literature.” In J. Folberg (ed.) Joint Custody and Shared Parenting. Washington, D.C.: Bureau of National Affairs & Association of Family & Conciliation Courts, 1984.

Steinman, S.B. “Joint custody: What we know, what we have yet to learn, and the judicial and legislative implications.” University California at Davis Law Review 1983,16,739-762.

Steinman, S.D. “The experience of children in a joint custody arrangement: A report of a study.” American Journal of Orthopsychiatry

Steinman, S.B., Zemmelman, S.E. & Knoblauch, T.M. “A study of parents who sought joint custody following divorce: Who reaches agreement and sustains joint custody and who returns to court.” Journal of the American Academy of Child Psychiatry 1985,24, 545-554.

Vaughan, D. Uncoupling: Turning Points in Intimate Relationships. NY: Oxford University Press, 1986.

Wallerstein, J. S. & Blakeslee, S. SecondChances:Men, Women,and Children a Decade After Divorce. NY: Ticknor & Fields, 1988.

Walsh, M.C. & Kalter, N. “Joint legal custody versus sole maternal custody: A comparison of the effects of custody arrangements on the post-divorce adjustment of children.” Paper presented at the Annual Conference fo the American Orthopsychiatric Associa- tion, 1986.

Watson, M.A. “Custody alternatives: Defining the best interests ofthe children.” Family Relations 198 1,30,474-479.

Webster, M. Jr. Actions and Actors: Principles of Social Psychology. Cambridge, MA: Winthrop, 1975.

Weitzman, L.J. The Divorce Revolution: The Unexpected Social and Economic Consequences for Women and Children in America New York: Free Press, 1985.

White, S.W. & Bloom, B.C. “Factors related to the adjustment of divorcing men.” Family Relations 198 1,30,349-360.

Williams, F.S. “Child custody and parental cooperation.”Joint Custo- dian 1988, January, 1-5.

Zemmelman, S.E., Steinman, S.B. & Knoblauch. T.M. “A model project on joint custody for families undergoing divorce.” Social Work 1987,32 (#l), 32-37.

1981,51,403-414.

35

shared parenting: critical review of the research literature

Documents