Abstract
How can schools help students build moral character? One way is to use prepackaged moral education programs, but as we report here, their effectiveness tends to be limited. What, then, can schools do? We took two steps to answer this question. First, we consulted more than 50 of the world’s leading social scientists. These scholars have spent decades studying morality, character, or behavior change but until now few had used their expertise to inform moral education practices. Second, we searched recent studies for promising behavior change techniques that apply to school-based moral education. These two lines of investigation congealed into two recommendations: Schools should place more emphasis on hidden or “stealthy” moral education practices and on a small set of “master” virtues. Throughout the article, we describe practices flowing from these recommendations that could improve both the effectiveness and efficiency of school-based moral education.
Tweet
How can schools build moral character? Research and top scientists agree: use “stealthy” approaches; concentrate on a few “master” virtues.
Key Points
Moral education typically has a weak influence on students, at best.
We consult the world’s top social scientists and recommend two changes to moral education.
Schools should rely more on “stealthy” practices and less on explicit moral education.
Schools should also focus on cultivating only a few “master” virtues.
We suggest concrete practices in line with these basic recommendations.
Introduction
Concern about the moral soul of America is on the rise (Norman, 2017), and by some metrics, our society is experiencing a time of moral decay. Though war and violence continue to decrease (Pinker, 2012), self-centeredness is increasing (Twenge, 2006; Twenge & Foster, 2008; cf. Wetzel et al., 2017), prosocial behavior is on the decline (Lever, 2014), and children lack basic moral knowledge (Smith, Christoffersen, Davidson, & Herzog, 2011). So what can be done?
A part of the solution could be intentional moral education. Some of history’s greatest minds have considered moral education a topic worthy of their focus. Plato, Aristotle, Locke, and Durkheim did not just describe what is good—they tried to understand how to bring about this good in society. As Aristotle put it, “. . . . the purpose of our examination is not to know what virtue is, but to become good, since otherwise the inquiry would be of no benefit to us” (Nicomachean Ethics, 1104a25-30).
So after thousands of years of studying moral formation, how much is known about cultivating morality? One purpose of this article is to briefly report on the current state of moral education. How effective is it overall? And what current practices—if any—seem to work well?
A second purpose is to propose improvements to moral education. To this end, we solicited the advice of leaders in behavioral, moral, and character science. We discuss our panel’s most common suggestions throughout (see their full advice in an online Supplement; https://tinyurl.com/Moral-Ed-Advice). Collectively, these experts have spent centuries on topics related to moral education, but until now, few have directly applied their knowledge to this area. We believe that their unique insight can help move moral education forward.
What is Moral Education?
Before going any further, we should first explain what we mean by moral and what we mean by moral education. Due to increased moral pluralism, a common question about moral education today is “whose ‘morals’ are these programs shaping?” (Hunter, 2000). Here, the word “moral” refers to thoughts, desires, or actions that go against or suppress self-interest for the sake of promoting the interest of others; this could be the interest of individuals, groups, or collectives (see related definitions in Blasi, 1999; Haidt, 2008; Luco, 2014).
By moral education we refer to programs intentionally designed to facilitate school-based moral development. This excludes programs targeting conduct disorder or disciplinary problems—which may impact morality but not intentionally. Nor do we discuss programs like the Girl and Boy Scouts, YMCA/YWCA (Young Men’s Christian Association/Young Women’s Christian Association), or 4-H—those that attempt to develop morality outside of schools.
How Well Does Moral Education Work?
Until recently, there had been few rigorous assessments of programs fitting our conceptualization of moral education, and the extant research was not promising. In their classic study on character development, Hartshorne and May (1930) concluded that common moral education practices did not work. Even worse, their data convinced them that morality was dictated by situations, not character; therefore, attempting to improve morality by developing children’s character might be an exercise in futility.
Today research demonstrates that meaningful moral character does exist (Meindl, Jayawickreme, Furr, & Fleeson, 2015), but do we have the tools to improve it? The largest randomized control study testing relevant programs suggests the answer is no (Social and Character Development Research Consortium, 2010). This assessment of seven promising character-building and social-development programs found no evidence of behavioral or social improvement.
As with any research, this study has its limitations. It did not include concrete behavioral measures, so measurement error could have hidden real effects. Outcomes were measured over a 3-year period; perhaps later follow-ups would reveal meaningful improvement. The study involved only third to fifth graders; perhaps interventions are more effective at different ages. And though the study assessed well-respected interventions, others might work better.
Do other programs work better? To find out, we reviewed school-based programs fitting our definition of moral education. Ours is not a comprehensive assessment of all character education programs (for such reviews, see, for instance, Berkowitz, 2011; Berkowitz & Bier, 2007; McDaniel, 1998). Instead, it offers a window into the current state of moral education.
We started by synthesizing the results of morally relevant interventions reported in the Institute of Education Sciences’ What Works Clearinghouse (WWC). The WWC compiles lists of the most promising educational programs and evaluates the evidence for their effectiveness. 1 The WWC lists 54 behavioral programs. Nine fit our concept of moral education and were evaluated by WWC-approved studies. Four showed effectiveness in areas relevant to this review. 2 As we will discuss later, however, the supporting studies often had methodological limitations, so we must conclude that the WWC programs did not show strong support for the effectiveness of existing moral education programs.
The WWC does not include all existing moral education programs, though. Next, we dissected reviews of character education programs that included components fitting our criteria for moral education (e.g., Berkowitz, 2011; Berkowitz & Bier, 2007; McDaniel, 1998). We also conducted our own search for moral education programs not included in the WWC or in existing reviews. Together, these reviews revealed modest evidence of the effectiveness of moral education. For instance, Berkowitz and Bier (2007) found evidence of effectiveness in 48% of programs intending to influence personal morality (e.g., honesty, respect), 43% intending to influence prosocial behaviors (e.g., volunteerism, sharing), and 62% intending to influence relationships (e.g., friends, family).
Though promising, many of the studies testing moral education effectiveness suffer from methodological weaknesses. The practical realities of evaluating education programs make it extremely difficult, if not entirely impossible, to conduct this work using every known best practice. Nevertheless, these flaws must be taken into account when evaluating moral education’s effectiveness. Most notably, most studies used quasi-experimental methods, had small samples, used self-report measures, or did not correct for multiple comparisons. Consequently, even the modest empirical support for moral education might be inflated. School administrators should therefore implement these programs with tempered hopes.
On a more positive note, our review suggests that the most efficient practices for cultivating students’ morality might be simply developing caring relationships and fostering a strong school community. Compared with the ineffective WWC programs, the four effective programs placed more emphasis on students’ relationships—both with teachers and fellow students. However, the five WWC moral education programs that did not demonstrate effectiveness seemed to focus more on building individual character strengths, such as courage, trustworthiness, and resilience. Our review supported this pattern; practices targeting interpersonal relationships appear to be an exception to otherwise discouraging results (e.g., White & Shin, 2016).
Relationship building is also commonly at the heart of scholars’ advice for moral development (e.g., Berkowitz, 2011; Hartshorne & May, 1930). As David Brooks (2016) has noted, though social programs rarely change people, caring relationships seem to have unique transformative power. Notably, community and relationship building was a common recommendation of our expert panel members. Our recommenders also proposed practical strategies to facilitate relationship development such as teacher–student continuity across terms and teachers greeting all students by name at their door.
How Can Moral Education Improve?
It seems, then, that there are promising strategies for school-based moral formation (e.g., relationship development), but moral education as a whole has considerable room for improvement. So what specifically can be done? Two general strategies and a number of specific practices could considerably improve moral education.
Less Emphasis on Explicit Moral Education
First, schools should focus less on explicit, direct moral instruction. Traditionally, moral education programs have taught moral and virtue vocabulary and instructed children on how to wrestle with moral dilemmas. To be clear, we believe this approach has value. Today’s young people cannot articulate their moral thoughts and values (Smith et al., 2011), and this lack of explicit moral insight seems destined to thwart moral maturation. Still, we agree with John Dewey (1909), who long ago noted that simply teaching abstract moral lessons is akin to “teaching the child to swim by going through motions outside of the water” (p. 14). Instead, we think moral education would be more effective—not to mention more time efficient—if educators adopted hidden or “stealthy” practices (Lapsley & Yeager, 2012). Given the findings of our review, this should include more focus on community and relationship building but also more covert embedding of evidence-based practices into schools’ existing environments and curriculum, as we discuss in the next section.
There are at least four reasons why it would be useful for moral educators to shift toward implicit and stealthy practices. First, conscious thoughts rarely drive moral judgments (Haidt, 2001; Valdesolo & DeSteno, 2006) or behavior (Blasi, 1980). Perhaps this is why Aristotle observed that “arguments seem . . . unable to turn the many toward being fine and good” (Nicomachean Ethics, 1179b5-10). Fortunately, the psychological processes that account for the lion’s share of morality—emotions, intuitions, inspiration, motivation—can develop through practices beyond explicit instruction (for instance, through reading literature, as we will discuss).
Second, explicit moral instruction tends to do little to tap into one of the most powerful moral forces: social influence (e.g., Haidt, 2001; Milgram, 1963). In contrast, many stealthy practices harness the formative potential of social influence. For instance, schools can encourage good behavior by conveying information about descriptive norms (i.e., typical behavior). Descriptive norms can exert their effects simply through low-cost messages on bulletin boards or posters (e.g., “87% of students participated in our last food drive”).
Third, stealthy practices buffer against psychological reactance (the tendency to reject persuasive messages to act in a certain way; Brehm, 1966; Nucci, Killen, & Smetana, 1996). This “don’t you tell me what to do” impulse is especially strong in young people (Yeager, Fong, Lee, & Espelage, 2015). Psychological reactance reduces the effectiveness of behavioral interventions, and it can even cause interventions to backfire by emboldening people to do the opposite of what they are being pushed to do. Whereas explicit moral instruction is highly vulnerable to psychological reactance, stealthy moral education, when done right, should be nearly immune.
Fourth, moral educators should rely more on stealthy moral curriculum (and less on explicit moral instruction) simply because educators’ interest in a stealthy curriculum approach is more likely to persist over time. Character education is currently popular in American schools, but how long will this popularity last? Less expensive programs stand a better chance of experiencing sustained attention. Compared with more traditional moral education, stealthy practices (described next) require considerably fewer resources.
Noninvasive behavioral interventions
This section suggests three stealthy strategies.
Wise interventions
First, moral educators should use existing, rigorously tested, state-of-the-art behavioral interventions that (a) are known to affect important behaviors and (b) require little class time. So-called “wise” interventions are one such category of intervention (Walton, 2014). Wise interventions are “wise to” the psychological seeds of social or personal problems (Walton, 2014). They also tend to target System 1 processes (e.g., well-being interventions target emotions like gratitude; Seligman, Steen, Park, & Peterson, 2005) and/or harness the power of social influence. In addition, they avoid psychological reactance by being stealthy: People rarely realize they are taking part in these interventions.
Schools should rely more on these brief interventions for many reasons. Foremost, they work: Wise interventions consistently improve prosocial and civic behavior (Bryan, Master, & Walton, 2014; Paluck, 2011), relationships (Finkel, Slotter, Luchies, Walton, & Gross, 2013; Page-Gould, Mendoza-Denton, & Tropp, 2008), and grades (Blackwell, Trzesniewski, & Dweck, 2007; Hulleman & Harackiewicz, 2009). They also require little time—from a few seconds to a couple hours. And they often entail students simply completing preprogrammed material on a computer, obviating the extensive teacher training that many moral education practices require.
Many existing wise interventions should influence morality. For instance, schools can use linguistic cues that invoke identity-based moral motivation. When making behavioral injunctions, people can use verbs (“please give to charity”) or nouns (“please be a giver”). The difference is subtle, but when people are asked to be a particular type of person rather than act in a certain way, subsequent behavior has implications for identity. As a consequence, noun-based directives are more powerful (Bryan, Adams, & Monin, 2013; Bryan et al., 2014). This insight has increased children’s helpfulness (Bryan et al., 2014) and reduced cheating (Bryan et al., 2013).
Other “wise” interventions target growth mind-sets: lay beliefs that personal characteristics and abilities can change. These interventions are brief yet powerful; they regularly improve school-related motivation and behavior (e.g., Aronson, Fried, & Good, 2002; Blackwell et al., 2007). Similarly, self-affirmative statements consistently improve academic achievement, especially among minority students (Cohen, Garcia, Apfel, & Master, 2006; Cohen, Garcia, Purdie-Vaughns, Apfel, & Brzustoski, 2009). Though growth mind-set and self-affirmation interventions have not yet been used to influence morality, they are robustly effective and should play some role in morality (Lindsay & Creswell, 2014; Porter, 2017).
That said, wise interventions have drawbacks. First, they do require some class time (up to 2 hr). Also, these interventions might not work in combination (Paunesku et al., 2015), perhaps because of reactance, so they probably work best when spread across the school year.
Nudging morality
Moral educators could avoid both of these drawbacks by using behavioral “nudges.” A nudge is a subtle environmental alteration that influences behavior by exploiting biases in human cognition, often System 1 processes and social influence. They also avoid psychological reactance by encouraging behavior that is in the interest of the actor without using coercion (Epley & Tannenbaum, in press; Sunstein & Thaler, 2008).
No-cost nudges affect many prosocial behaviors, nearly doubling organ donation (Johnson & Goldstein, 2003) and more than doubling charitable giving (Small, Loewenstein, & Slovic, 2007). One nudge relevant to moral education is placing images of eyes in hallways and classrooms, which could encourage students to act more morally (likely because they make people feel like they are being watched; Haley & Fessler, 2005; see a list of nudges at https://tinyurl.com/Moral-Ed-Interventions).
Sometimes wise interventions and nudges may seem too good to be true, and recent replication failures suggest that they occasionally are. But the evidence for their effectiveness is at least as strong as for traditional moral education programs. Furthermore, the stealthiness of these interventions means that they barely cut into instruction time, require little financial investment, and often require zero staff development.
Moral education through storytelling
Schools can also create effective indirect moral curriculum by making small changes to their nonmoral curricula, especially subjects involving storytelling (e.g., language and history). The third most common recommendation by our expert panel was to stimulate moral emotions, and stories can do this (Kidd & Castano, 2013; cf. Panero et al., 2016). Stories can also inspire children to perform more prosocial behavior (e.g., Larsen, Lee, & Ganea, 2017; cf. Hakemulder, 2000).
Some moral education programs already teach morality through literature, but not all stories equally develop morality (Narvaez, 2002). For instance, schools should be careful about how they use moral heroes to inspire morality. Children can learn new behaviors from heroes, and their extraordinary morality can stimulate moral sentiments, motivation, and behavior (Aquino, McFerran, & Laven, 2011). But some figures (accessible ones) are more inspiring than others (exceptional, unreachable heroes—famous [e.g., Gandhi] or not [e.g., Giraffe Heroes Projects’ everyday exemplars; www.giraffe.org]; see also Lockwood & Kunda, 1997). Redemptive heroes—exemplars who did not always act in exemplary ways—may be more inspiring (Klein & O’Brien, 2017). Teachers could find ways to remind children about more personally relevant and attainable moral figures, such as family members or even peers (Han, Kim, Jeong, & Cohen, 2017). For instance, students could learn about moral exemplars in their school, their local community, or their family. They could also bring to school pictures of these personal exemplars to keep at their desks or on classroom walls.
Stories with moral themes should stress the positive consequences of acting morally, rather than the negative consequences of acting immorally (Lee et al., 2014). In the traditional story of George Washington and the cherry tree, George admits to his father that he cut down the tree, and his father praises him for his honesty. Teaching this story improves honesty. However, when children read an adapted story in which George is punished for lying to his father about cutting down the tree, their honesty is unaffected. In the same way, children’s behavior is unresponsive to stories like Pinocchio and The Boy Who Cried Wolf, in which the protagonists experience negative consequences for their dishonesty.
Last, moral learning can occur when young children read picture books—provided they do not include animal protagonists. Although more than half of children’s books include anthropomorphized animals (Marriott, 2002), only children’s books with human characters improve moral behavior (Larsen et al., 2017). This of course does not mean that schools should throw out books with anthropomorphized animals, but when the goal is moral formation, teachers should emphasize books with human characters.
Focus on a Few Virtues
Our second recommendation is more speculative but related to our panel’s most common advice: Schools should focus on cultivating only a few “master” moral virtues (Blasi, 2005), such as self-control and humility.
Self-control
Self-control refers to the “ability to override or change one’s inner responses, as well as to interrupt behavioral tendencies (such as impulses) and refrain from acting on them” (Tangney, Baumeister, & Boone, 2004, p. 274). Self-control resolves inner motivational conflicts in the service of higher order goals (Hofmann, Baumeister, Förster, & Vohs, 2012) and leads to reduced cheating (Gino, Schweitzer, Mead, & Ariely, 2011; Mead, Baumeister, Gino, Schweitzer, & Ariely, 2009) and increased prosocial behavior (DeWall, Baumeister, Gailliot, & Maner, 2008; Martinsson, Myrseth, & Wollbrant, 2012).
At least two strategies to improve students’ self-control require neither substantial teacher training nor considerable time investment by students. First, because, like a muscle, self-control seems to build with use (Muraven, Baumeister, & Tice, 1999; Oaten & Cheng, 2007), students can perform small tasks that require willpower. Just practicing good posture can help students build their general capacity for self-control (Muraven et al., 1999).
Second, students can learn to avoid temptations that overpower self-control. To demonstrate self-control, people must sometimes exert self-restraint, but usually avoiding temptations in the first place is much more effective, simpler, and less odious (Duckworth, Gendler, & Gross, 2016; Greene & Paxton, 2009). Schools can teach students to think of ways to avoid temptations that keep them from acting morally. For instance, perhaps a student does not want to cheat on their exams, but they always have a clear view of their neighbor’s tests. This student might ask their neighbor to be sure to cover their answers, or they could ask their teacher to be moved to a different desk. Of course, a prerequisite for the effective use of situational strategies such as these is moral motivation; children need to have the desire to avoid temptations to attempt to do so. But provided students do have the desire to act morally, teaching them when and why to use situational strategies could quickly and substantially improve their behavior.
Humility
Here, humility refers to the accurate perception of one’s own weaknesses and strengths as well as greater concern for others relative to oneself (Chancellor & Lyubomirsky, 2013). In essence, humility is having a “small self.” Humility has long been viewed as a master virtue, just as its obverse, pride, has long been viewed as a core vice (Edwards & Wesley, 1745; Augustine, Letter 118). Recent research suggests that humility relates to many other virtues, including gratitude (Kruse, Chancellor, Ruberton, & Lyubomirsky, 2014), compassion (Krause & Hayward, 2015), generosity (Exline & Hill, 2012), and self-control (Tong et al., 2016).
Humility was at the core of a plurality of the recommendations we received (https://tinyurl.com/Moral-Ed-Advice). For instance, one psychologist proposed that the main reason people have trouble acting morally is that they delude themselves into believing that they are more moral than they really are. Increasing one’s moral self-accuracy (humility) could therefore benefit moral development.
Schools can increase humility through self-affirmation, gratitude, and awe (Ruberton, Kruse, & Lyubomirsky, 2016). We mentioned self-affirmation interventions earlier (Cohen et al., 2009). To increase gratitude, students could deliver thank you letters to people who have been kind to them (Seligman et al., 2005), or they could simply list what makes them grateful (Emmons & McCullough, 2003). A quick, cheap way to boost awe is contact with nature (Zhang, Piff, Iyer, Koleva, & Keltner, 2014); simply looking up at a tall tree might work (Piff, Dietze, Feinberg, Stancato, & Keltner, 2015). Of course, many students do not have the luxury to take leisurely walks in the woods, but fortunately, pictures and videos of beautiful nature scenes induce awe as well (Valdesolo & Graham, 2014). All these techniques can improve humility (Ruberton et al., 2016). Given humility’s relationship to so many other virtues, incorporating these practices into a moral education program could be highly efficient.
Conclusion
Today’s moral education programs show promise, but if educators are serious about using school as a conduit for moral formation, progress still needs to be made. Here two general strategies appeared: more emphasis on “stealthy” practices and more emphasis on master virtues like self-control and humility. We believe that these strategies will increase the effectiveness and efficiency of moral education.
Footnotes
Authors’ Note
The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agency.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article was made possible by the Walton Family Foundation (Grant number: 2015-350).