Rejoinder to Gary Guttings Doubts about the Behavioral Sciences

Gary Gutting, Professor of Philosophy at Notre Dame, published an opinion piece on the New York Times Opinionator pages. In it he spoke of the “severely limited reliability of social scientific results” and asserted that “While the physical sciences produce many detailed and precise predictions, the social sciences do not.  The reason is that such predictions almost always require randomized controlled experiments, which are seldom possible when people are involved.”

If Dr. Gutting’s assessment of the behavioral sciences were accurate, it would mean that there is little hope for society.  If it is impossible, as he speculates, to rigorously investigate the factors that influence human behavior, we shall never evolve a society that has less conflict, crime, academic failure, depression, suicide, drug abuse, teenage pregnancy, marital conflict, or obesity.  Physical and biological science will continue to produce a steady stream of findings and technologies, but we will continue to cope with these evolving and sometimes harmful developments with no greater capacity to influence behavioral and cultural developments than was true three hundred years ago.

Fortunately, Dr. Gutting is mistaken. Over the past fifty years, the behavioral sciences have made enormous progress. Indeed, there has been an explosion of randomized-controlled experiments, many of them with multi-year or even several decades of follow up. Going to PsychInfo or Pubmed, one can easily find several thousand randomized controlled trials testing behavioral science policies, practices, programs and strategies.

The Institute of Medicine’s report on prevention reviews the results of the past 30 years of prevention research.  They concluded that we have the knowledge to ensure that virtually every child arrives at adulthood with the skills, interests, and health habits needed to live a productive life in caring relationships with others.  Here are some examples of what has been learned.

Effective Family Interventions

Numerous family interventions can improve children’s academic success. They do so by increasing parents’ skills, strengthening child and adolescent self-regulation, and preventing the development of myriad psychological and behavioral problems that both perturb academic success and cause substantial harm to the health and wellbeing of young people and those around them.

• The Nurse Family Partnership. Nurses provide support to mothers living in poverty during their pregnancy and the first two years of the baby’s life. Randomized trials show that the program has numerous benefits including reduced child abuse, better child cognitive development, increased parent income, and the prevention of delinquency when children reach adolescence.
• Play and Learning Strategies I and II. Susan Landry and her colleagues developed Play and Learning Strategies to strengthen parents’ support of infants’ and young children’s cognitive, language, and literacy development. In randomized trials they showed that the version for infants (PALS I) and the version for 2- and 3-year-olds (PALS II) significantly improved parents’ stimulation of children and improved children’s self-regulation, verbal development, and cooperation with their mothers–skills which are vital to children’s academic and social development.
• The Family Check-Up. Family support providers help parents clarify what they are doing well and what they can do better to ensure their children’s successful development. Multiple randomized trials have shown its value in improving children’s self-regulation, preventing problem behavior, and improving children’s academic success. A middle school version reduced delinquency as much as four years following the program. Four years after families of young children got the Family Check-Up the children were doing better in school than children who did not get the program.
• Triple P. This community-wide approach to promoting nurturing parenting produced a significant effect on child maltreatment in a randomized trial done in 18 counties in South Carolina. It cost less than $15 per child. As the Kaiser-Permanente studies on adverse childhood experiences (ACE’s) show, such reductions in child maltreatment have a large impact on lifetime health-care costs.
  • New Beginnings. This program is designed to help divorcing families prevent the psychological and behavioral problems that often arise when families break up. In carefully done randomized trials, Irwin Sandler and his colleagues have shown that the program reduces aggressive and uncooperative behavior, substance use, depression, and risky sexual behavior, and improves academic success and self-esteem.

Effective School Interventions

Numerous school interventions help students develop the self-regulation, social, and academic skills needed to succeed.

• Improving the quality of preschool and childcare settings. A child living in poverty is much more likely than other children to lack the self-regulation, vocabulary, and social skills needed to succeed in elementary school. However, high-quality preschools have a clear benefit in reducing the risk that children living in poverty will fail when they get to elementary school. James Heckman, a Nobel Laureate in economics, has estimated that the greatest return on investment from programs to support young people’s development will come from investing in early childhood education. Susan Landry and colleagues have shown that their eCIRCLE system for training preschool providers significantly improves teachers’ skills and children’s language, literacy, and social development.
• Kids in Transition to School (KITS). This brief, intensive program helps children who are about to start kindergarten learn the readiness skills they will need to succeed. A randomized controlled trial of KITS showed that it improved children’s self-regulatory skills and social competence, both key to success in school.
• Effective Reading Instruction. Children who cannot read by third grade are unlikely ever to read well. There are a number of well-validated reading instruction programs and good systems for helping schools implement them. Every high-poverty neighborhood should ensure that such a system is in place in its elementary schools and that the progress of children in learning to read is closely monitored.
• The Good Behavior Game (GBG). Teachers reward student teams for on-task and cooperative behaviors. A randomized trial of GBG showed that playing the game in just in first or second grade prevented development of smoking and delinquency by sixth grade and prevented suicidality, drug abuse, and criminal behavior when students reached young adulthood. For every dollar invested in the game, there was a return of $95. If applied as a behavioral vaccine like we do with medical vaccines, this proven strategy could save America $57 billon per first-grade cohort when they reach their 21^st birthday.
• Positive Action. This program provides a comprehensive, integrated approach for preventing problem behavior, promoting character development, and improving academic achievement among youth. Randomized trials have shown that the program prevents tobacco, alcohol, and other drug use and violent behavior, and increases students’ academic performance.
• Positive Behavioral Intervention and Support. More than 15,000 U.S. schools now use this schoolwide system for teaching and rewarding positive social and academic behavior. Randomized trials have shown that it reduces discipline problems and increases student academic performance.

Evidence-Based Kernels

In addition to comprehensive programs, some simple behavior influence techniques can help parents and teachers nurture the skills and interests of children and adolescents. Embry and Biglan (2008) have identified more than 50 such kernels.

• Mystery Motivator or Prize Bowl. This is a great way to motivate kids, individually or in groups. Create or decorate a bowl or coffee can. Select and/or invent prizes for the classroom or home. Write a description of each prize on a small card or slip of paper and put it in the bowl. The power of the prize increases when students have a bowl filled with prizes to draw from. If the child or youth does something that you want to encourage, then they get to draw a prize for the whole class or their group.
• Beat the Timer. A teacher or parent sets a timer and challenges the child to complete transitions, chores, tasks, quizzes, or seatwork before the timer rings. If the child completes the activity before the timer goes off, they get a reward. Beat the Timer dramatically reduces angry faces and voices that can trigger negative behavior or withdrawal among those who have been exposed to trauma and trouble.
• Meaningful roles. Creating meaningful roles for students creates opportunities to provide reinforcement and recognition of students’ competence and reduce downtime in the classroom.

Evidence-Based Policies

There are many policies that have empirically established effects on public health. Kelli Komro and her colleagues have identified 46 policies that met the highest standard for empirical support. Specifically, they had consistent, positive outcomes from 2+ high-quality experimental or quasi-experimental trials using a comparison group or interrupted time series design. Here are three examples:

• Alcohol Beverage Excise Tax. Increasing the tax on alcoholic beverages has numerous benefits. Evidence indicates that doubling the tax would produce.
  • 35% reduction in alcohol-related morbidity and mortality
  • 11% reduction in traffic crash deaths
  • 6% reduction in sexually transmitted disease
  • 2% reduction in violence
  • 1% reduction in crime
  • 6% reduction in general alcohol consumption
  • 17% reduction in beer consumption
  • 25% reduction in wine consumption
  • 25% reduction in distilled spirits consumption
  • 1% reduction in consumption among heavy drinkers
  • Tobacco-Free Policies (Smoke-free workplaces, restaurants, and other public places).  On average, smoke-free workplace policies can achieve:
    • 29% reduction in total cigarette consumption per employee (95% CI: 11%, 53%).
    • 3.8% reduction in smoking prevalence (95% CI: 2.8%, 4.7%).
    • Decrease in consumption of 3.1 cigarettes per day per smoker (95% CI: 2.4, 3.8).
    • Teenagers who worked in totally smoke-free worksites were 32% less likely to ever smoke than those who worked in less restricted work sites.
    • Enforcement of Minimum Legal Drinking Age Laws On average, higher legal drinking ages can achieve:
      • Significant reductions in alcohol consumption among underage youth
      • Significant reductions in traffic crashes
      • Lower rates of alcohol-related health and social problems (e.g., unintentional injuries, crime); although evidence is inconsistent across some outcomes).

Clinical Interventions

There has also been tremendous progress made in our ability to treat mental, emotional, and behavioral disorders. Starting in the 1960’s the behavior therapy movement began to identify effective interventions for psychological and behavioral problems of children and adults. In the past 15 years even more effective approaches have emerged. For example, there are now more than 50 randomized trials of Acceptance and Commitment Therapy (ACT) showing its value for a surprisingly broad range of problems. By helping people to accept unpleasant thoughts and feelings without judging them and focus on pursuing valued outcomes, ACT has helped people lose weight, stop smoking, decrease drug abuse, reduce depression and anxiety, control diabetes, cope more effectively with schizophrenia, reduce epileptic seizures, and prevent job burnout.

The Consequences of Misinformation

Stereotypic views of the various sciences and the prestige or lack of respect that they confer have real consequences. It is doubtful that many members of Congress know much about the physics underlying the research they are helping to fund at the $ 7.5 billion Hadron Collider. But we know that physicists are reliable scientists.  They blew up Japanese cities.

Little recognized, but just as reliable and far more beneficial, is the behavioral science knowledge that can ensure that millions of children develop successfully. In light of this knowledge, the nation should be mobilizing.  The National Institutes of Health, the Centers for Disease Control and Prevention, the Substance Use and Mental Health Services Administration, and a large network of foundations and non-governmental organizations should be pursuing a coordinated strategy to ensure that every family, school, workplace, and community nurtures the development of young people. They could be developing and testing strategies to disseminate the programs, policies, and practices that have proven benefit in helping children develop. As these practices spread, hundreds of thousands of children will be saved from dropping out of school, developing drug abuse, delinquency, or depression.

To some extent these things are happening, but at a much slower pace in the United States than is justified by the randomized controlled trials conducted in the U.S.  There are, of course, a variety of reasons.  But as long as it is widely believed that the social sciences are unreliable there will be no reason for policymakers to study the facts.

I don’t want to oversell the power of the behavioral sciences to bring about improvements in people’s health, happiness, and productivity. In every one of the programs, practices, and policies that I enumerated, plenty of people do not benefit. But in every case, a significantly greater proportion benefit than would have been true if usual practices prevailed. For example, the Nurse Family Partnership’s 13 year follow-up showed that 35% of children whose mothers didn’t get the NFP were arrested by age 15, while fewer than 15% of the NFP children were.  Consider the benefits in the lives of these young people and the reduction in crime victim and criminal justice costs.

If the fruits of behavioral science progress are not as obvious as the fruits of the past 100 years of physical and biological sciences it is largely because the programs, policies, and practices that can so improve human wellbeing have not yet been widely disseminated. One reason is that the most powerful and well-documented interventions have only emerged in the last twenty years. But another is that a broad swath of the nation’s intellectual leaders and policymakers mistakenly believe, as Dr. Gutting does, that real progress in the social sciences is not possible.  What hope do we have of achieving the nurturing families, schools, and communities that our research shows are vital and possible, if those who make the decisions about public policy are not only unaware of our work, but do not believe that effective interventions are even possible?

The problem is not limited to philosophers and policymakers.  The notion that real science can be done in biology, but not in the “social sciences” is alive and well at the National Institutes of Health.  Tremendous progress is being made in genetics and neuroscience. With good reason, the leadership of the National Institutes of Health is excitedly pursuing these areas, with reasonable hope that major breakthroughs will occur in our ability to treat and prevent major illnesses.  Dr. Nora Volkow, the Director of the National Institutes on Drug Abuse, believes that we are getting close to creating vaccines that will block the ability of substances such as cocaine or nicotine to addict people.

Most of the research I have described was funded by the National Institutes of Health.  Yet this work has been a tiny fraction of the NIH research budget and support for it is dwindling.  One reason is that the budget of NIH is declining.  This harms all science and thereby society. But in this context, when the leadership of NIH is far more aware of advances in neuroscience and genetics than in the social and behavioral sciences, it is the latter areas that are getting squeezed the most.

We have reached a stage in the scientific understanding human behavior where very significant improvements in human wellbeing can be achieved. It will happen only if we can get the leadership of society to put resources into getting these well-researched programs, policies, and practices, widely and effectively implemented.  That will never happen so long as misinformation about the effectiveness of the social and behavioral sciences continues to be spread.

Dr. Tony Biglan