An Empirical Look at Carl Rogers, Thomas Gordon,
and PET Claims and Promises
By
Tatia Nelson and Paul Robinson
Historical Review of PET’s Creators
In the mid-twentieth century, a paradigm shift influenced mainstream psychology in favor of non-punishment parenting models such as Parent Effectiveness Training (Gordon, 1970). Regardless of substantial empirical evidence showing the ineffectiveness of non-punishment parenting approaches and the efficacy of punishment-reinforcement parenting models, Biglan (1998) points out that the main message the mental health profession is giving society is that non-punitive parenting is effective while parenting that includes the use of punishment is not. This presentation focuses on one specific parenting system that condemns the parental use of reinforcement and punishment—Parent Effectiveness Training. It particularly includes a review of the history of the people and practices involved in PET’s development, a look at the basic PET principles and claims, and a review of the scientific evidence for and against PET.
While Thomas Gordon officially created Parent Effectiveness Training (PET) as a parenting system, the parenting principles underlying PET essentially came from Gordon’s mentor, Carl Rogers. According to Kirschenbaum’s (1979) biography of Carl Rogers, he was born January 8, 1902, in Oak Park, Illinois, to an upper middle class Christian family. He received his college education from the University of Wisconsin and Columbia University, earning his doctorate in 1931. During his education Rogers enjoyed being on debate teams, and that is perhaps the reason Rogers was well respected for his ability to present and defend his psychological beliefs.
After attending the World Student Christian Federation’s Conference in Peking, China at the age of 22, he was touched by the Zen tradition of insisting that the individual find answers within himself as this quote explains:
“The Perfect man…does not interfere in the life of beings,
He does not impose himself on them,
But he helps all beings to their freedom.” Lao-Tse
Prior to going to the conference Rogers was planning on entering the ministry. Rogers’s experience in the Far East led him to conclude Jesus Christ was just a man, not the Son of God. Rogers left organized religion and became a champion of the humanist movement in Psychology, which asserts that the answers to a person’s problems in developing lie within the person, not in God or science. He believed free agency could not coexist in a world where rewards and punishments influence one’s behavior. Having lost his belief in a living God and having accepted Lao-Tse’s belief of noninterference in others’ lives, Rogers felt the Bible’s emphasis on rewards and punishments guiding behavior was incorrect.
Rogers developed five main parenting principles upon his humanistic beliefs. The first is the rejection of the value of reinforcement and punishment. The second principle, based on his Zen concept of free agency explained that a man’s decisions and choices were based upon free agency, excluding the influence of consequences for one’s actions. The Rogerian parental role is to be supportive of a child’s decisions rather than directing the child with consequences. The third principle, unconditional love, came from Rogers’s belief that each person contained within him/herself all the necessary knowledge at birth to guide their own life toward becoming a fulfilled individual. Like a peach seed internally knows it is to become a peach tree and only needs a supportive environment to become the tree, Rogers argued a child just needs a supportive family environment of unconditional love, and that child will inherently move in the direction of personal fulfillment. The fourth principle, active listening, came from Rogers’s belief that the main role of a good therapist or parent was as an ‘active listener’ who allows the child to talk out his/her problems and reach his/her own solutions. The fifth principle, children’s choice parenting, advocates a family system where children are allowed to make their own decisions and parents take the role of a consultant who gives advice when asked. According to Rogers, parents serve as models that children will then emulate. The Rogerian parental role is thus permissive, and basically relies on children responding to their desires.
Carl Rogers did not carry out any scientific investigations to validate any of his ideas about parenting. The research he did carry out focused mainly on analyzing therapist-client therapy sessions. He received a great deal of criticism for lack of scientific soundness in his research, and in 1954, openly admitted his research methodology was weak (Kirschenbaum, 1979).
Rogers never fully implemented his talk-it-out parenting approach in raising his own children. Biographer Kirschenbaum (1979) interviewed Rogers’s children and asked them what happened in their home when they misbehaved. The children unanimously responded that the talking it out, listening approach was not used in their home. When they did something wrong, they were hit or sent to their room until their attitude changed. They all said they had a strong love and respect for both of their parents.
Thomas Gordon, a graduate student of Carl Rogers, decided to create a complete parenting approach built on Carl Rogers’s theories and ideas. In 1962, Gordon carried out his first parenting workshops and eventually published Parent Effectiveness Training in 1970. Gordon presented PET as the new, scientifically proven parenting approach based on parenting techniques scientifically shown to be better than reward-punishment approaches. Gordon also claimed that science had shown punishment and reinforcement contingencies to be harmful in child rearing.
PET’s Basic Principles and Claims
On the cover of his book, Gordon describes PET as the “Tested new way to raise responsible children.” In his book, Gordon (1970) claims that parents who are trained in PET will ‘greatly increase their effectiveness in parenthood’ (p.2). Gordon claims, “punishment can be discarded forever in disciplining children—and I mean all kinds of punishment, not just the physical kind” (p.3). Gordon criticizes the biblical support of using rewards and punishments in parenting as being an outdated and ineffective parenting approach as he notes, “Unlike almost all other institutions of civilization, the parent-child relationship seems to have remained unchanged. Parents depend on methods used two thousand years ago” (p.4).
Gordon says parents should not exercise any power and authority over children because parents are not necessarily wiser than children. He claims children should not be subject to parental supervision on issues including hair length, dress standards, music standards, smoking, or driving cars (Gordon, 1989 p. 266).
Gordon’s main parenting principles are directly adopted from Carl Rogers and include the following: unconditional love, also known as ‘communicating acceptance’ of whatever the child does; children’s choice parenting which permits the child to make their own decisions while parents surrender parental authority; active listening which encourages parents to listen to their children rather than telling them what to do; being a good parental model, assuming that children will emulate their parents if they are good models and refrain from rewarding or punishing their children; and using I-messages to influence children’s actions with the goal of changing misbehavior.
Like Rogers, Gordon carried out no substantive body of empirically based scientific research to validate his parenting ideas or support his parenting principles.
Review of Empirically Based Scientific Evidence
1. PET as a complete Parenting System
As previously mentioned, Gordon began giving parenting workshops in PET in 1962 using handouts and completed his first book in PET in 1970. In 1977, Rinn and Markle reviewed fourteen research investigations that were designed to determine the effectiveness of PET. Rinn and Markle felt the research warranted the following conclusion, “Over-all, the effectiveness of PET as a prevention or intervention strategy was not supported.”
Dembo, Sweitzer, and Lauritzen (1985) reviewed twenty investigations designed to evaluate PET in terms of three factors: (1) the ability of PET to produce behavior changes in children, (2) the ability to change attitudes of parents about parenting (some people felt PET might not cause changes in children, but might influence changes in parental attitudes), and (3) changes in children’s feelings of self-esteem (PET claimed it produces better psychosocial development in children than reinforcement-punishment based parenting, and that it would show up in changes in a child’s feelings of self-esteem).
In terms of changing children’s behavior, Dembo et al. (1985) reported, “there is little evidence that children’s behavior is affected consistently by their parents participation in PET.” Seven of the studies used the five scale Parent Attitude Survey (PAS) to measure changes in parental attitudes. According to Dembo et al. (1985), “There was little consistent change in the total score or scale scores on that instrument (The PAS).” In terms of self-esteem changes caused by PET, Dembo et al. (1985) also reported, “None of the researchers using children’s self-esteem measures reported any significant changes as a result of their parents’ PET program training.”
Since Dembo et al.’s (1985) review of the research on PET, no in depth review has produced conclusions that differ from the Dembo group review. PET as a complete parenting system has not been scientifically established as an effective parenting approach for changing children’s behavior, changing parenting attitudes or improving the psychosocial development in children as measured by self-esteem changes.
2. Research on Specific Behavior Change Procedures Advocated by PET
Our second review of PET research focused on evidence showing whether any of the specific techniques that are part of PET have been shown to be effective. For example, PET claims exhibiting unconditional love (communicating acceptance), active listening, I-messages, refraining from using rewards and punishments, reinforcement free parent modeling, and letting children make their own decisions are components of PET that affect positive behavior changes in children.
In 1983, Educational Testing Service (ETS) put together a task force of research psychologists to review all the empirical research on specific behavior change techniques being advocated by different psychological theories and parenting approaches. They also aimed to determine how much empirical support each technique has. The task force (Weber, Crawford, Roff, and Robinson, 1983) reported that the behavior change techniques of rewards and punishments, which PET says parents should refrain from using, actually received their highest rating in terms of having substantial research showing them to be effective behavior change procedures. Weber et al.’s (1983) review indicated PET’s behavior change techniques of communicating acceptance, using active listening and I-messages had no empirical evidence showing they were effective. For example, when referring to communicating acceptance, Weber et al. noted, “Although (communicating) acceptance is a widely advocated strategy, an examination of the literature yields little support for its desirability or effectiveness.” No research reviews published since Weber’s review has presented substantive evidence that contradicts Weber et al.’s findings.
Gordon’s contention that parental modeling without the use of rewards and punishments not only fails to have any supportive empirical evidence, but also is condemned by the leading modeling researcher of the past 50 years, Albert Bandura. Bandura’s extensive research points out that rewards (a child seeing the model being reinforced for his behavior and the child receiving rewards when he emits the behavior) are important factors in getting children to exhibit behaviors being demonstrated by models.
Bandura (1977) was concerned about parenting systems like PET advocating parenting modeling without reinforcement and made the following comment:
“Some child rearing authorities have popularized the view that healthy personality development is built on ‘unconditional love.’ If this principle were, in fact unfailingly applied, parents would respond affectionately regardless of how their children behaved whether or not they mistreated others, stole whatever they wanted, disregarded the wishes and rights of others, or demanded instant gratification. Unconditional love, were it possible, would make children directionless and quite unlovable” (Bandura, 1977, p. 102).
Gordon (1970,1991) makes some very strong claims about punishment not only being an ineffective behavior change procedure, but also a detrimental behavior change procedure. Therefore, we focused special attention toward looking at the empirical research on the issue of the role and value of punishment (any stimulus that decreases probability that a behavior will be repeated (Chance, 1999)) as a behavior change procedure. After 1962, when Gordon began his parenting workshops that condemned the use of parental punishment, three major reviews of the role and value of punishment came out over the next twenty years. Azrin and Holz (1966) authored a large chapter in a book that compared the empirical support for punishment to the three other most commonly advocated approaches for changing misbehavior. Walters and Grusec (1977) authored a book that reviewed all the empirical research published on the use of punishment on both human and nonhuman animals. Axelrod and Apsche (1983) edited a book that focused on empirical research for the effects of punishment on human behavior. Walters and Grusec (1977) and Axelrod and Apsche’s (1983) punishment reviews looked at both primary effects and secondary or side effects of punishment. All three of the reviews agreed that punishment is an effective behavior change procedure that has received a great deal of unsupported condemnation. The fact that a number of mental health professionals have authored articles making unsupported claims against punishment is exemplified by Axelrod and Apsche’s (1983) review of punishment research where they talk about addressing claims about punishment effects:
“In at least one account (Maurer, 1974), punishment is credited with causing juvenile delinquency, hyperactivity, anti-social aggression, vandalism, minimal brain damage, and homicide. Evidence for such claims, except in the case of aggression, is nonexistent… But evidence that physical punishment is a significant variable in the determination of antisocial lifestyles of criminal proportions has not been obtained” (Axelrod & Apsche, 1983, p. 291-292).
In 1950, Glueck and Glueck published Unraveling Juvenile Delinquency. Their data showed that 56.8% of boys in detention centers came from homes with lax and permissive mothers, while only 4.4% of boys in detention came from homes where mothers were overly strict. This empirical data challenges the unsupported claims opposing the use of punishment.
Diana Baumrind (1978) identified a permissive parenting style as one requiring parents to “behave in an affirmative, acceptant, and benign manner towards the child’s impulses and actions…and the parent sees him- or herself as a resource for the child to use as he wishes, but not as an active agent responsible for shaping and altering the child’s ongoing and future behavior…Some permissive parents are very protective and loving, while others are self-involved and offer freedom as a way of evading responsibility for the child’s development” (p. 244). PET falls under this permissive category of parenting with its underlying principles of children’s choice parenting, unconditional love, active listening, and I-messages. While Gordon strongly denies PET is a permissive parenting system, it is difficult to define it any other way.
When looking at punishment research where punishment was experimentally applied to human subjects (rather than ex post facto punishment research having a host of uncontrolled secondary variables) the data indicate punishment was very effective in changing behavior without producing undesirable side effects, including aggression and antisocial behavior as concluded by Axelrod & Apsche (1983):
“The undesirable side effects reviewed here come from a relatively small proportion of all the studies on the therapeutic use of punishment. Even when allowance is made for the probable under-reporting of negative side effects due to editorial sanctions, it is interesting to note how few studies in the literature contain observations that would suggest clinical or ethical problems. In considering the studies where undesirable side effects were observed, the overall impression that results is one of mild surprise that serious side effects are seen so infrequently…Most of the undesirable side effects described lasted only for a few minutes or days, were quickly responsive to treatment if they did not disappear spontaneously, and constituted a relatively small and ethically justified price to pay in return for the elimination of much more detrimental behaviors” (Axelrod & Apsche, 1983, p. 300-301).
At the American Psychological Association (APA) 2001 annual convention, Baumrind and Owen announced findings of their longitudinal study of 100 families on the effects of spanking used as a parental punishment method. “The results showed no negative effects on cognitive, social, or behavioral skills of those youngsters and found no difference between them and the 4 percent of children who were not physically disciplined” (Baumrind & Owens, 2001).
In 1975, twelve years after Gordon began telling society punishment was an ineffective and detrimental behavior change procedure the number one selling college textbook on the Psychology of Learning, authored by Hulse, Deese, and Egeth, cited over 600 empirical research studies supporting the effectiveness of reinforcement and punishment. In response to the parenting books and literature condemning punishment, Hulse, Deese, and Egeth (1975) stated, “From this literature, one can gain the strong impression that punishment is a maladaptive, ineffective, and wasteful technique to use in the establishment and guidance of behavior. Such is patently not the case…” (p. 201).
In 1975, college textbooks on the psychology of learning were citing hundreds of studies showing reinforcement and punishment as effective behavior change procedures. Contrary to claims of PET advocates at that time, the learning texts did not cite any research showing PET techniques (e.g. unconditional acceptance, active listening, modeling without reinforcement, I-messages) to be effective.
In 1989, Gordon authored a book entitled Discipline that Works and wrote, “I searched the psychological literature for all the reports of research studies I could find on discipline, punishment, power, styles of parenting, and so on…I became determined to share what I learned with others who are seeking an understanding of discipline and its effects on children” (p. xxvviii). Although Gordon claimed he was presenting all the research evidence on discipline and punishment, his book includes no references to any of the scientifically sound research previously presented in this article that indicates punishment is effective. Gordon referenced 103 books and articles in his review. Most of his citations reference opinion papers and anecdotal information against punishment (e.g. Spock, 1957; Gilmartin, 1979; Gordon, 1970, 1974, 1975, 1977; Dreikurs, 1948; Rogers, 1961) and some good research articles he misrepresented (e.g. Kadushin & Martin, 1981; McCord & McCord, 1959). For example, he suggests McCord and McCord’s (1959) in-depth study of 650 youth over two decades shows delinquency and aggression are caused by punitive parenting by selectively citing some of McCord & McCord’s statistics which report 70% of the boys with punitive and rejecting mothers and fathers were convicted of crimes. Gordon failed to point out that the overall findings of the study was (1) permissive parenting (like PET) and not punitive parenting was the main type of parenting situation from which juvenile delinquents come, and (2) intense professional Freudian and Rogerian (PET type) counseling not only did not reduce juvenile delinquency, but in fact resulted in more delinquency (McCord, 1975). What Gordon (1989) described as an in-depth review of punishment research literature did not represent the overall findings of the empirically based punishment research.
A thorough look at the punishment research appears to strongly support the idea that punishment is an effective behavior change technique. Gordon’s (1970) claim that PET’s argument about punishment being ineffective is not scientifically supported.
In conclusion, Rogers and Gordon never carried out nor identified any substantive body of empirical scientific evidence to support their punishment free, children’s choice parenting ideas and approach. Gordon’s claims of the PET parenting system as a whole being effective in changing children’s behavior, changing parental attitudes, or producing more healthy psychosocial development in children as measured by changes in children’s self-esteem are not scientifically supported. Empirical evidence does not support Gordon’s claim that the specific behavior change techniques of unconditional acceptance, refraining from using rewards and punishments, active listening, and I-messages are effective. Gordon’s claim that parents can effectively employ reinforcement free modeling to change children’s behavior lacks empirical support and goes against current modeling research findings. Gordon’s claims about punishment being an ineffective and detrimental behavior change procedure are not supported by scientific evidence either.
The substantive body of empirical scientific research in fact identifies punishment as one of the most effective behavior change procedures.
References
Axelrod, S., & Apsche, H., (1983). Effects of Punishment on Human Behavior. New York: Academic Press.
Azrin, N. H., & Holz, W. C., (1966). Punishment. In W.K. Honig’s (Ed.). Operant Behavior: Areas of Research and Application. New York: Appleton-Century-
Crofts Bandura, Albert (1977). Social Learning Theory. Englewood Cliffs: Prentice-Hall.
Baumrind, Diana (March 1978). Parental Disciplinary Patterns and Social Competence in Children Youth and Society, 9 (3), 239-273. Baumrind, Diana & Owens, Elizabeth. Press release, August 2001.
Bergin, A., & Strupp, M., (1972). Changing Frontiers in the Science of Psychotherapy. Chicago: Adline.
Chance, P., (1999). Learning and Behavior 4th ed. New York: Brooks/ Cole Publishing Company.
Dembo, M. H., Sweitzar, M., & Lauritzen, P., (1985). An Evaluation of Group Parent Education: Behavioral, PET, and Adlerian programs. Review of Educational Research ,Summer, 55 (2), 155-200.
Dreikurs, Rudolph (1948). The Challenge of Parenthood. New York: Hawthorn Books.
Glueck, S., & Glueck, E., (1950). Unraveling Juvenile Delinquency. New York: The Commonwealth Fund.
Gordon, T., (1970). Parent Effectiveness Training. New York: Peter H. Wyden.
Gordon, T., (1989). Discipline That Works. New York: Plume Books.
Hilgard, E. R., Marquis, D.G. (1940). Psychoanalysis as Science. New York: Basic Books.
Hulse, S.H., Deese, J., & Egeth, H. (1975). The Psychology of Learning. New York: McGraw-Hill.
Kadushin, A. & Martin, J.A. (1981). Child Abuse: An Interaction Event. New York: McGraw-Hill Co.
Kirschenbaum, H., (1979). On Becoming Carl Rogers. New York: Delacorte Press.
Maurer, A. (1974). Corporal Punishment. American Psychologist, 29: 614-626.
McCord, W. & McCord, J. (1958) Origins of Crime: A New Evaluation of Cambridge-Somerville Study. New York: Harper Row.
Rinn, R. C., & Markle, A., (1977). Parent Effectiveness Training: A Review. Psychological Reports, 41, 95-109.
Rogers, C., (1961). On Becoming a Person. Boston: Houghton Mifflin.
Walters, G.C., & Grusec, J.E., (1977). Punishment. San Francisco: Freeman.
Weber, W.A., Crawford, J., Roff, L.A., & Robinson, C. (1983). Classroom Management: Reviews of the Teacher Education and Research Literature. Princeton: Educational Testing Service.