Explored factors that influence parenting behavior through examination of previously reported and more recently developed models for predicting parenting behavior. Patterson’s behavioral model, Abidin’s initial model of parenting stress, and Belsky’s process model of determinants of parenting are reviewed. A proposed model is presented to improve on previous models by integrating sociological, environmental, and behavioral factors, as well as personality characteristics of the parent, in predicting parenting behavior. Finally, description is provided on a new self-report measure (i.e., the Parenting Alliance Inventory) developed to focus on those aspects of the marital relationship that bear directly on parenting. Implications for future research are presented.
The role of parent and the acts of parenting have been of interest to philosophers, religious leaders, and society in general since ancient times. Empirical investigation of the role and functioning of parents, on the other hand, has a history of less than a hundred years, with most studies being of recent vintage. During the past two decades, the interest of psychologists and society in understanding the factors that influence parenting behavior has intensified given the widespread publicity surrounding the phenomenon of child abuse and concerns regarding the outcome of child development in the face of reduced mother-care and increased other-care.
In the area of child abuse, researchers have demonstrated that negative child outcomes and dysfunctional parenting are associated with a large number of sociological and environmental variables. A number of behaviorally oriented investigators have documented the relationship of certain specific parenting behaviors to cognitive deficiencies in children and child abuse.
More recently, researchers building on the work of behaviorally oriented psychologists and those who study parent–child interactions have developed transactional models which represent a bridge between the sociological, environmental, and behavioral perspectives. What appears to be missing at this point is a strong commitment of researchers to a consideration of parental belief and motivational systems as variables equally important to those previously described.
The deficit in our current perspective may be appreciated by considering the following analogy. Imagine, if you will, that our task as psychologists is to define the determinants of the behavior of a ship, in the same sense that psychologists are trying to define the determinants of parenting behavior. Further, assume that you are an alien researcher and know nothing about these ships. You begin your study by observing ships, and you note that a ship leaves Port A and crosses the ocean and reaches Port B. Some other ships leave Port A and arrive at Port C, whereas others break up or sink in transit. You quickly come to recognize that outcomes are diverse and you suspect that the determinants of these behaviors or outcomes are many. As a researcher you must decide which variables to study.
The environment seems a good choice: You check the weather, look for icebergs, and assess a large number of other environmental variables. You might then proceed to correlate these variables to the outcome of the ship’s voyage. That process would be similar to the way sociological environmental variables have been linked and demonstrated to have some utility as predictors of parenting behavior.
You may also choose to focus your research on observations of the ship’s behavior while in transit. From this behavioral perspective you notice that when a ship faces strong winds blowing from the north, the bow of the ship turns toward the north—at least it does in those ships that make it across the ocean. You also observe that of those ships that do not point their bow to the north, many capsize in rough weather. From these observations, we have a clear picture of a significant behavior of the ship as it attempts to cross the ocean. Thus, studying the relation of the ship’s behavior to external stressors or events reveals a strong relationship between behaviors and outcomes. Clearly, use of the behavioral approach has demonstrated that it is an informative way to predict the behavior of ships in certain circumstances.
Despite the utility of these approaches, full understanding of the determinants of the ship’s successful ocean crossing cannot be achieved without examining what goes on inside the ship. An examination of the internal mechanics, the reasoning, the beliefs, and the motives of whatever exists inside the ship is needed.
Psychology, as a science and profession, has developed some theories and models regarding parenting which have demonstrated heuristic value, but which need refinement. This refinement is, of course, taking place. One excellent behavioral model for predicting an area of parenting behavior is Patterson’s model, which suggests some of the determinants of antisocial and aggressive behavior in children. This model clearly describes relationships between parent and child behaviors, but does not inform us as to why the behaviors occur. What determines or causes the behaviors? One must ask: Why do some parents respond inappropriately at various points in Patterson’s model, whereas others do not? Why do some disengage from the process of increasing escalation whereas others do not?
If we return to the ship analogy, it may be that ships typically turn into the wind in a storm. It is also observed that if they don’t, they usually capsize or don’t reach port. We have an orderly and predictable behavioral theory of ships’ behavior. Is our behavioral theory of ships useful? Certainly, it is. Theories and models of parenting behavior, however, need to be developed that go beyond the stimulus–response behavioral perspective and that integrate sociological and cognitive-psychological approaches with behavioral approaches.
The recent research of Ed Tronick and Tiffany Field and colleagues illustrates that we can identify some of the processes by which parent–child interactions are regulated by the beliefs and expectations of both parties during the child’s infancy. These studies are providing the information needed to build better models of the determinants of parenting.
In 1976, while developing the Parenting Stress Index, Abidin made an initial attempt at developing a model which integrated a range of variables that were believed to be central to the role of parenting. This model used stress as the central construct, with stress leading to dysfunctional parenting. At that time, Abidin believed higher levels of parenting stress led to increased dysfunctional parenting. Subsequently, both his research and that of others demonstrated that a simple linear relation did not exist between stress level and dysfunctional parenting. Abidin found that very low levels of parenting stress also were associated with dysfunctional parenting due to the disengagement of the parent and the subsequent low level of vigilant parental behaviors.
Recently there have been more elegant attempts at model building designed to illuminate the paths of the “determinants of parenting.” In 1984, Jay Belsky presented his model of the determinants of parenting in relation to child abuse. His model attempted to define the major global sociological and personality characteristics which related to parenting behaviors. In 1986, Belsky, Hertzog, and Rovine reported the results of an initial test of that model using some rather sophisticated statistical analyses. The model, which included personal historical, sociological, behavioral, and self-report data, produced challenges to some of Belsky’s previously held beliefs and to the conventional wisdom of the scientific community regarding some of the global family variables that were thought to have predictive utility in relation to parenting behavior. For example, the assumed importance of marital adjustment to parenting and child outcomes was challenged, whereas the importance of parental developmental history was increased as a direct predictor of parenting behavior. Belsky’s subsequent work has helped to expand the variables now considered important in models of parenting behavior, and his work has helped focus us on the interior of the parent. Although an improvement, Belsky’s model does not fully capture the parent as a thinking, planning, goal-oriented individual.
At this time, we need to develop more comprehensive and integrative models designed to incorporate sociological, behavioral, and personality characteristics of the parent. These models should suggest the measures that need to be developed to assess the conceptual components of the model. A model currently being developed by Abidin and his students is presented next, along with a brief description of a measure under development that is designed to assess one of the conceptual components of the model.
The proposed model is built on the work of many others and represents a distillation of many variables either known or suspected to be related to parenting behaviors. The variables included in the model are those that are suggested by Abidin’s interpretation of the literature as being the best bets for predicting parenting behavior. Clearly, the universe of variables is not present and others may propose different variables.
The model hypothesizes that parenting behavior and child adjustment are influenced by a number of sociological, environmental, behavioral, and developmental variables. The current model attempts to capitalize on those relationships, but suggests that the path of influence of those variables is through that component of the parent’s personality relating to the parenting role. Further parental cognitions and beliefs are seen as playing a key role.
The parenting role variable (commitment to parental role) in the model represents a set of beliefs and self-expectations serving as a moderator or buffer of more distal influences. Each parent is seen as having an internal working model of himself or herself as a parent. This model of “self-as-parent” is created out of the individual’s attachment history and includes the individual’s goals for himself or herself and his or her internalized expectations of others. Through this working model of self-as-parent (i.e., parenting role), parents assess the harm or benefit that confronts them in the role of parent. The result of that appraisal produces the level of stress the parent experiences.
Parenting stress is, thus, the result of a series of appraisals made by each parent in the context of his or her level of commitment to the parenting role. Conceptually, parenting stress is viewed as a motivational variable which energizes and encourages parents to utilize the resources available to them to support their parenting. The richness or paucity of resources available naturally plays a key role in the ultimate parenting behavior. The dynamics of this model are quite similar to those suggested by Lazarus and Folkman in their transactional model of stress. The current model represents an explication of a specific application of Lazarus and Folkman’s general theory.
Now this presentation shifts to a brief review of Abidin’s work on developing a self-report measure of one of the important conceptual variables in his model. The parenting alliance variable in Abidin’s model replaces the marital satisfaction variable included in Belsky’s model. This decision was based on the research of Belsky et al., Emery, and others who have researched the effects of conflict in marriage on child outcomes, and who found that marital satisfaction is not predictive of parenting behavior. The Parenting Alliance Inventory is a self-report measure which was developed to focus on those aspects of the marital relationship that bear directly on parenting. It is Abidin’s belief that a global measurement of marital satisfaction or adjustment, as is often used in family and child development research, introduces considerable error variance in relation to the prediction of parenting behavior. It is believed that this more specific measure (Parenting Alliance Inventory) will make a significant contribution to the prediction of parenting behaviors because it recognizes that both parents can be involved, and function well, in the parenting role, and yet not be very satisfied with their personal relationship with each other. Brunner, in his dissertation research, found that Parenting Alliance Inventory scores were able to predict significantly child behaviors as rated by teachers, whereas marital satisfaction scores were unrelated to the child outcome measures.
The research literature in recent years suggests that an unfortunate mistake was made during the past 25 years when many researchers, rushing to embrace behavioral methodologies, devalued, gave up on, or ignored self-report and personality measures designed to assess parental belief systems. Mischel’s landmark position that “situational” rather than personality factors by and large determine behavior was constructive in redirecting attention to environmental factors; however, it oversimplified the case.
Clearly, self-report methodology has had many problems associated with it, not the least of which is the use of global measures of personality characteristics to predict specific behaviors. An excellent review of methodological and conceptual issues related to self-report measures is presented by Holden and Edwards. In the old tradition of personality measurement, it was assumed that from a limited number of global personality traits one could predict specific behaviors, and attempts were made to do just that. The failure to demonstrate significant and substantial correlations between global attributes and specific (often micro) behaviors was seen in the 1960s and 1970s as demonstrating the invalidity of personality assessment, and by implication, self-report methodology.
More recent research has demonstrated that narrow-band personality measures consisting of questions which are directly tied to a specific belief system can predict and help define the determinants of parenting behaviors. An excellent explication of these issues may be found in the book edited by Irving E. Siegel entitled Parental Belief Systems: The Psychological Consequences for Children. Further, the recent work of Kochanska, Kuczynski, and Radke-Yarrow, using a more fine-grained assessment of parenting belief systems, demonstrates major and significant relationships of parental belief systems to observed parenting behavior.
Both the importance of parental belief systems and their paths of influence need to be emphasized. Parental belief systems as measured by self-report have both direct and indirect influences on parenting behavior and child outcomes. The indirect path is mediated by the dyadic interaction—thus the parent engages in interactions with the child based on his or her belief system, and child outcomes follow from these interactions. The direct path relates to the environment the parent creates or involves the child in and to the inferences the child makes about the parent’s belief system (expectations). This direct influence cannot be understood by observing a particular behavior, but can be determined only over a long history of parent–child interactions in a large variety of contexts. Thus behavioral observations are an inefficient and somewhat ineffective method of assessing some of the direct influences of the impact of parental belief systems on their children. Direct assessment of parental belief systems by self-report appears a more effective and practical methodology.
Many researchers and clinicians educated during the past 25 years need to overcome the prejudices against personality assessment in general, and self-report methodology in particular. Self-report of parental cognitions is one important data source for both theory construction and clinical assessment. Now we need to develop models of the determinants of parenting that will allow for and support the incorporation of various data sources to facilitate the prediction and understanding of parenting behavior. The emergence of new models and theories that integrate a wide range of data sources is a hopeful sign for developing a fuller understanding of the determinants of parenting behavior.