promising policy implications

SAGE Books

Self-Control and Crime Over the Life Course

Adolescence and Adulthood: Is Self-Control Stable Over Time?

Contributors: By: Carter Hay & Ryan Meldrum

Book Title: Self-Control and Crime Over the Life Course

Chapter Title: "Adolescence and Adulthood: Is Self-Control Stable Over Time?"

Pub. Date: 2016

Access Date: October 31, 2021

Publishing Company: SAGE Publications, Inc.

City: Thousand Oaks

Print ISBN: 9781483358994

Online ISBN: 9781544360058

DOI: http://dx.doi.org/10.4135/9781483397726.n5

Print pages: 111-141

© 2016 SAGE Publications, Inc. All Rights Reserved.

This PDF has been generated from SAGE Knowledge. Please note that the pagination of the online

version will vary from the pagination of the print book.

Adolescence and Adulthood: Is Self-Control Stable Over Time?

Adolescence and Adulthood: Is Self-Control Stable Over Time?

As the previous chapter emphasized, how much self-control we develop early in life depends on pivotal factors related to early social interactions between parents and children, genetics, and neurobiological development, among other things. Without question, if high self-control is to emerge, a good start is critical—acquiring self-control early (relative to similarly aged others) is like starting a race with a head start and without the clock running. Under such circumstances, impressively hitting later milestones becomes much easier.

Indeed, those who develop self-control early tend to keep it and reap its rewards across the many arenas of life. This follows in part from its self-perpetuating nature. As we will discuss, those with high self-control often are selected and sorted into social arrangements that maintain or even enhance self-control. Also, the factors that got them off to such a great start often persist. For example, those who received terrific parenting in childhood often are treated to that same advantage in adolescence. But the converse also is true—those who get a rough start often have disadvantages that persist and pile up. Putting it all together, we see that the individual self-control differences established in childhood often are preserved or amplified in the years and decades that follow—those struggling with self-control in early adolescence continue to do so, while those with the greatest self-control thrive.

And yet, as you might imagine, that is not the whole story. Many individuals will fit the pattern of self-control stability just described, but there are exceptions. Some might be late bloomers—they successfully develop self-control over time but at a slower pace than others. This has them struggling with self-control deficits around age 6, 8, or 10, but by middle adolescence, they may have caught up with most others. There also are those who coast along just fine into the teen years but are then knocked off course. These individuals initially benefited from their head start, but the complex interplay of biological, psychological, and social changes during adolescence was too much. Both types of cases—the late bloomer and the late crasher—illustrate the same fact: A person’s self-control is never set in stone. It is subject, instead, to subtle shifts, or even major transformations, as individuals biologically develop, encounter new social experiences, and drift into new habits and standards.

In this chapter and the next, we explore this idea of self-control stability and change. We begin by discussing research that addresses this question: As individuals advance into adolescence and adulthood, how much does their level of self-control change? How much does it remain stable?

We then focus in this chapter on patterns that involve stability in particular (we cover patterns of self-control change in the next chapter). Researchers have devoted significant attention to the idea of behavioral stability, much of it inspired by the adage that past behavior is the best predictor of future behavior. We describe the scholarship on this issue to offer explanations for the self-control stability often observed. In the process, we explore the idea that a person’s self-control remains the same over time in part because self-control reflects a deeply rooted trait that is resistant to change. We also consider, however, that a person’s self-control remains stable over time because the social environmental factors that shape it (including the quality of the family and community environments) often are themselves quite stable. And in joining these two ideas, we consider that a person’s deeply rooted trait level of self-control—which may be developed quite early in life—shapes the choices they make about such things as who to choose as friends, how much to cooperate with parents, and whether to try hard in school. All these things have implications for further development of self-control.

Stability and Change in Self-Control

As we age, how much do we change? One common perspective on behavior and personality is that we do not fundamentally change as we age—we are who we are, and that remains the same over time. This sentiment is captured well by William James, the noted philosopher and psychologist of the 1800s, who said that for most of us “the character has set like plaster and will never soften again” by a relatively early point

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 2 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

in life (1890, p. 124). Suspicions of this kind have persisted through the present and are captured in a more recent quote that has floated around American culture, getting passed around on Facebook and Twitter and making an appearance on HBO’s True Blood (one of the surprisingly large number of vampire dramas found on television in recent years). The quote is this: “People don’t change, they just find new ways to lie.”

Another view of human behavior, however, is that people do change over time, often in positive ways. From this perspective, we make cognitive and neurological advances as we age into adolescence and adulthood. There also are the benefits of experience—as we grow older, we are exposed to further socialization. This provides us a greater range of prior experiences from which to draw when we make important decisions. Taken together, this captures the older but wiser theory of human development.

In reality, there is truth to both perspectives, as we describe below. Before describing that research, however, an important but somewhat tedious distinction must be considered because it is critical for understanding stability and change in self-control. This distinction is between absolute and relative stability (and correspondingly, absolute and relative change). Absolute stability exists if people experience no individual changes in self-control as they age—their absolute level of self-control at one age is equal to their absolute level of self-control at another age. This can be considered in reference to the Brief Self-Control Scale (Tangney, Baumeister, & Boone, 2004) discussed in Chapter 3. This scale scores individuals between 13 and 65. If a woman scores 48 on this measure at both age 16 and age 26, she is marked by perfect absolute stability—she is staying the same in an absolute sense over time. The idea of relative stability, on the other hand, involves one’s level of self-control when compared to similarly aged others. Relative stability occurs when one’s self-control ranking in a sample remains the same over time. This occurs, for example, for a man whose score on the Brief Self-Control Scale is in the 50th percentile of a sample at both age 16 and age 26. In that 10-year span, he maintained perfect relative stability by staying in exactly the middle of the sample’s distribution. If, however, there is movement from the 50th to the 40th percentile during this stretch, relative change has occurred—his level of self-control relative to the rest of the sample has changed (it went down).

In practice, absolute and relative stability often go hand in hand—large absolute improvements in self-control, for example, often correspond to large relative improvements within a given sample. However, this is not necessarily the case—if much of a sample improves over time in an absolute sense (and this may be true of self-control during certain stretches of the life course), then absolute improvements for an individual may not translate into relative improvements in the distribution. Indeed, if individual improvements in self-control are less than a sample’s average improvement, absolute increases could correspond to relative decreases in self-control.

Having said all this, what can we learn from the research? In absolute and relative senses, does self-control stay the same or improve as people age? One approach to answering this question involves looking at an entire population, with all its differently aged individuals, and considering how levels of self-control vary across the different age groups. Are there interesting differences between adolescents, young adults, the middle-aged, and the elderly? Research on the Big Five personality trait of Conscientiousness indicates that there are. Roberts, Walton, and Viechtbauer (2006) drew from multiple studies to estimate absolute differences in Conscientiousness among individuals ranging from 15 to 70 years old. They found higher Conscientiousness among older individuals, thus supporting the contention that as individuals are exposed to further socialization and greater life experiences, they develop greater self-control. This pattern starts quite slowly in adolescence—Conscientiousness is only slightly higher among 20-year-olds than among 15-year- olds, but it is substantially higher among those who are 35 or older. Improvements continue from there and are quite steady through roughly age 65. Interestingly, a similar pattern is observed for traits like Agreeableness and Emotional Stability, both of which overlap with the concept of self-control. This shows that the effects of age are largely positive—as people age, they become (on average) more socially competent and mature. In Roberts and Mroczek’s (2008, p. 33) words, “With age, people become more confident, warm, responsible, and calm.”

We must emphasize, however, that cross-sectional age comparisons like these do not directly reflect changes in self-control that individuals experience as they age. There are two limiting factors. First, recall that these age curves come from large samples of individuals who differ in age and are mostly studied at a single point in time. Self-control differences between those of different ages might reflect the effects of aging, but they also

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 3 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

might reflect the effects of unique historical experiences of the young and old individuals in the sample—the 35-year-olds just mentioned were born during a different historic and cultural era than the 20-year-olds, and their higher Conscientiousness could be explained by that. (If so, researchers refer to this as a cohort effect). Even if we set that possibility aside, however, there is a second complication: The differences we just described refer to average patterns for a sample (e.g., the average 35-year-old has higher Conscientiousness than the average 20-year-old). There is every reason to expect that these average patterns obscure a great deal of individual variation. Some individuals follow the average pattern quite closely, but others may diverge from it greatly, in both good ways and bad. Thus, we must move beyond the average pattern for a population to also consider unique individual variations.

To consider this, researchers need longitudinal panel data—we need data in which the same individuals are followed over time and measurements of self-control are collected at multiple points along the way. A number of criminological studies take this exact approach using direct measures of self-control (rather than measures of related personality constructs). They often use a method known as group-based trajectory modeling (Nagin, 1999). This method uses repeated measures of self-control across a span of time to identify the most common self-control trajectories for a sample. A major practical difference from the cross-sectional approach just described is that this method allows us to identify multiple self-control trajectories—not just the average one—for a given sample.

Most of this research has focused on children and adolescents. Hay and Forrest (2006), for example, studied stability and change in self-control from ages 7 to 15 for a national sample of about 3,500 children. Their analysis revealed eight different common trajectories, although some were much more common than others. These trajectories are shown in Figure 5.1. Each was given a hyphenated label indicating its initial level of self-control at age 7 and the nature of absolute change over time. For example, the High-Stable group started high in self-control and remained stable over time. Also, for each trajectory, the number in parentheses indicates the estimated percentage of children in that group. The different lines may appear to offer a chaotic picture, but a few themes can be readily discerned regarding both absolute and relative stability. Most notably, these trajectories point to high absolute and relative stability during this period, especially among the more than 50% of children who had high self-control by the start of the study period (i.e., those in the Very High- Stable and High-Stable groups). These groups had high self-control from the beginning, and this remained the case through age 15. Regarding absolute stability, they started with a score of 2.75 or higher (on a scale with a maximum of 3.00) and remained there through age 15. This corresponded to high relative stability—individuals in these two groups were mostly at the top of the self-control distribution throughout the entire study period.

There were another 25% (the Medium-Stable group) who had merely medium levels of self-control, but that also was marked by strong absolute stability—the members of this group were doing at least okay in self- control, and this was consistently true from age 7 to 15. On a relative basis, however, this group shows a bit of change—notice that the Medium-Stable trajectory is traversed by two smaller trajectories (at age 11 and then 13), indicating that its position in the distribution was not perfectly fixed.

Indeed, it is these smaller groups that showed interesting patterns of self-control change in both the absolute and relative sense. One group (Low-Increasing), representing about 5% of the sample, started with quite low self-control at age 7 but steadily improved over time. By age 15, they were nearly as high as the groups that started and ended with the highest self-control. On the other hand, the High-Decreasing group—estimated to be just over 1% of the sample—showed the opposite pattern. At age 7, they were excelling in self-control, but minor absolute decreases from age 7 to 11 (from about 2.75 to 2.40) were followed by dramatic absolute decreases from age 11 to 15 (from about 2.40 to 1.40). These absolute changes corresponded to relative change: This trajectory traversed the lines of almost every other one, and by age 15, this group had the lowest self-control in the sample.

Figure 5.1 Trajectories of Self-Control (N = 3,793)

Source: Hay, C., & Forrest, W. (2006). The development of self-control: Examining self-control theory’s stability thesis. Criminology, 44, 739–774.

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 4 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

Although the trajectories vary a bit from sample to sample, other trajectory studies have revealed similar results. For example, Higgins, Jennings, Tewksbury, and Gibson (2009) found that absolute and relative stability was the typical pattern among students from schools in six large to midsized U.S. cities. Change occurred as well, however, especially among one group (16% of the sample) that started high at age 12 but ended at moderate levels at age 16. Similarly, in studying a large sample of Kentucky students from age 13 to 16, Ray, Jones, Loughran, and Jennings (2013) found that about 70% started with at least medium or high self-control. Over the next three years, these adolescents maintained that level and stayed above most others in the sample, thus confirming the finding that absolute and relative stability in self-control is common, especially among those who start with reasonably high self-control by late childhood or early adolescence. The remaining 30%, however, experienced notable shifts in both an absolute and relative sense. Cases marked by better-than-expected self-control gains and unexpected self-control losses emerged in this study, just as they did in Hay and Forrest’s (2006) sample.

Several studies have examined self-control stability and change with a slightly different approach, and more specifically with fewer data points. These studies often had access to just two waves of data with a relatively short time span (two years or less) between them. In such instances, a nuanced trajectory cannot be estimated; however, interesting insights are still possible. Computing the correlation between Time 1 and Time 2 self-control is quite informative—such correlations are referred to as stability coefficients because they indicate the extent to which the relative ranking of individuals is maintained over time. If self-control is stable, the Time 1/Time 2 correlation should be quite high; in principle, such correlations can be as high as 1.00, although random measurement error makes this a practical impossibility.

The studies on this question typically reveal correlations in the .40-to-.50 range over relatively short time periods (four years or less; Burt, Simons, & Simons, 2006; Raffaelli, Crockett, & Shen, 2005; Turner & Piquero, 2002; Winfree, Taylor, He, & Esbensen, 2006). This points to a strong but imperfect level of relative stability. Burt and colleagues (2006), for example, found a correlation of .48 between self-control measures taken two years apart in early adolescence. Underlying this correlation was a pattern in which about 65% of subjects remained in the same half of the distribution across the two time periods (e.g., those in the top half of the self-control distribution remained there two years later). Similar correlations have also been observed in personality research, and they become higher than .50 as individuals age deeper into adulthood (Roberts & DelVecchio, 2000).1

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 5 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

In taking this research as a whole, we offer this direct conclusion: Shifts in self-control do occur, but stability in self-control is quite common. Thus, people can change, and a nontrivial portion do just that, but in practice, stability is the more common pattern. With that conclusion in mind, any framework for understanding self- control over the life course must consider both stability and change, because there is plenty of both. In the remainder of this chapter, we focus on stability to consider why it occurs and how it shapes life as individuals navigate their way through childhood, adolescence, and early adulthood.

Why Does Self-Control Often Remain Stable?

The adage that past behavior is the best predictor of future behavior rings true, and as noted above, this often is the case for self-control in particular. In this section, we describe explanations for why self-control remains stable. We consider three central factors contributing to self-control stability. The first is that self- control remains stable because it is a manifestation of a persisting individual trait—because this trait persists within the individual over time, so too does a given level of self-control. The second explanation involves a comparable argument about persisting social environments—stability in self-control occurs because the early social environments that affect self-control tend to be stable themselves, therefore producing continuity in self-control. We consider this especially in connection to stability in parenting and peer associations during childhood and adolescence, but also in reference to the stability of poverty and economic disadvantage. And then our third explanation involves a life course theory dynamic known as state dependence that in some ways combines the first two ideas. These different explanations are sometimes seen as contradictory to one another, but as we discuss, each is almost certainly part of the puzzle for explaining the significant self-control stability that is observed.

In Focus 5.1

The “Buddha Boy”

Few long-term displays of self-control are as remarkable, and as controversial, as the months-long meditation of the Nepalese teenager Ram Bahadur Bomjon, also known as the “Buddha Boy.” According to various accounts, Bomjon set out to meditate undisturbed for a period of six years without food or water. Shortly after beginning his period of meditation in 2005, large crowds of followers, scientists, and journalists flocked to the site. Fences had to be erected and tight security enforced. While skeptics claimed that food and water was provided to Bomjon during the middle of the night, several film crews, including one from National Geographic, observed and taped him for lengthy periods, noting that the typical physical effects of dehydration and starvation were not happening to Bomjon. The frenzy surrounding Bomjon ultimately led him to leave the site to continue his meditations elsewhere, but the mystery surrounding his prolonged periods of meditation continues to stir debate as to the legitimacy of his feat.

Bomjon’s story is intriguing for discussions of persistence and self-control, as it speaks to the human capacity to override basic impulses to eat, drink, or simply leave a seated position. So, the next time you find yourself fidgeting and antsy after having to sit in the same seat for an hour—maybe during a class, or, for those of us who are professors, during a tedious faculty meeting—consider the self-control it must have taken Bomjon to remain in the same position for days, weeks, and, if you believe the claims, perhaps months. For more information on the incredible story of Bomjon (who is now referred to by some followers as His Holiness Dharma Sangha), visit www.dharma-sangha.com.

Persistent Individual Traits as Contributors to Self-Control Stability

The presence of persistent individual traits is the most straightforward explanation for self-control stability. In criminology, this often is referred to as a latent trait perspective because of its emphasis on stable traits that are difficult to directly observe but that operate consistently over time. From this perspective, people who display low self-control at one point will continue to do so in the future simply because that is who they are. They possess deeply rooted, inherent characteristics that prevent them from regulating their emotions, thoughts, and actions. Nagin and Paternoster (2000, p. 119) refer to this characteristic as “an initial propensity

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 6 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

or proneness . . . that has reverberations over time” and that “affects the probability of antisocial conduct early in life and at all subsequent points.” Similarly, Wright and his colleagues (2008, p. 35) speak of a latent tendency that “resides within individuals”—something they refer to as antisocial potential. From their perspective, there is variation over time in how antisocial potential manifests itself, but the latent trait is always there, and it encourages behavioral stability over time and across different situations and realms of life, including the home, school, and work. Moffitt (1993, p. 679) speaks of this in reference to antisocial behavior in general (rather than self-control in particular), but her point on “cross-situational consistency” is the same: Those with the latent trait “lie at home, steal from shops, cheat at school, fight in bars, and embezzle at work.”

The latent trait may be a function of genetic or biological forces from the earliest stages of life. Theorists like Wilson and Herrnstein (1985), Moffitt (1993), and Wright and his colleagues (2008) emphasize biological and genetic causes of the latent trait, including the neuropsychological deficits discussed in Chapter 4 that arise from such things as exposure to abuse, neglect, and harmful toxins in critical stages of development. Moffitt (1993), for example, emphasizes how neuropsychological deficits early in childhood interact with criminogenic social environments to produce a latent trait that persists over the life course. A latent trait could also follow from early social environmental experiences in which self-control habits and preferences become fixed. Gottfredson and Hirschi are the most prominent advocates of a self-control conceptualization in which the latent trait follows from social environmental experiences, especially those involving the quality of parenting. From their perspective, differences in parenting produce persistent individual differences in self-control that are unlikely to change:

The differences observed at age 8 or 10 tend to persist . . . Good children remain good. Not so good children [those with the latent trait of low self-control] remain a source of concern to their parents, teachers, and eventually the criminal justice system. (Hirschi & Gottfredson, 2001, p. 90)

Regardless of how the latent trait emerges, the argument remains the same: Differences in self-control between individuals persist because of varying possession of a latent quality that becomes an inherent part of who people are. For those possessing the latent trait of low self-control, the idea of overriding—whereby an urge to behave impulsively is replaced with a line of action that better contributes to long-term well-being—is not a major part of their behavioral repertoire.

There is another important argument of this perspective, one that we emphasized in Chapter 2 when discussing Gottfredson and Hirschi’s theory in particular: Because the latent trait explains stability in observed self-control over time, other events and relationships that might appear to affect self-control in fact do not. For example, we might observe that those with delinquent friends are especially likely to have low self-control. Perhaps delinquent friends lower one’s self-control by encouraging impulsive behavior that prioritizes short- term thrills over long-term well-being. From a latent trait perspective, however, this argument is incorrect. The latent trait is responsible for both outcomes (having low self-control and having delinquent friends). Those with self-control deficits in childhood go on to have them in adolescence as well (because that is just who they are), and not surprisingly, they self-select into friendships with others who are similarly low in self-control. Indeed, they likely self-select into all sorts of antisocial arrangements, including problems at school during adolescence and rocky, unsuccessful romantic relationships in early adulthood. None of these circumstances, however, is critical to explaining their behavior. Over these different time periods and contexts, the latent trait is guiding it all.

Persistent Environmental Characteristics: Parenting and Peers

The argument of persistent environmental characteristics mirrors the latent trait approach. The main difference is its focus on the stability of social environments (rather than individual traits) as the source of self- control stability. Research on this possibility focuses especially on the stability of parenting. Parents who raise their children well during the early stages of life are expected to be similarly effective during adolescence. If so, stability in self-control may follow from stability in such things as parental attachment, involvement, monitoring, and discipline across the first two decades of an individual’s life. To be clear, specific parenting techniques likely will vary across different stages of early life. For example, direct controls based on parents’ power—including strict supervision or coercive punishments—that are common in early childhood often are phased out over time and replaced with parental appeals to principles of fairness, reciprocity, and specific

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 7 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

values (Collins, Madsen, & Susman-Stillman, 2002; Steinberg & Silk, 2002). Nevertheless, the parents who are most effective during the early years should make the best adjustments during adolescence. If so, there will be significant relative stability in the quality of socialization.

Much research does in fact reveal notable relative stability in parenting. Stability coefficients for measures taken at different times can be computed for parenting variables in the same way they are for self-control. Meta-analyses often reveal correlations in the range of .40 to .50 (Holden & Miller, 1999)—a level of stability quite similar to what is observed for self-control. Kandel and Wu (1995), for example, in a study of children who aged from childhood to early and middle adolescence, found correlations of .45 and .46 for variables measuring the consistency of discipline and the closeness of the parent–child relationship (see also Loeber et al., 2000).

The stability of peer associations has also received attention. Of particular interest is the idea of “sticky” delinquent friends—when an adolescent begins hanging out with delinquents, those friendships have a way of sticking, much to the disappointment of parents. Warr (1993) looked at this phenomenon with five waves of data from adolescents who were initially 13 years old and 17 by the end of the study period. Warr (1993) found that for various types of delinquency, many adolescents reported having no friends who committed the act during any year of the study; this was especially true for law violations involving marijuana and theft. This obviously is good news for parents who want their children to stay away from troublemaking friends, but Warr (1993, p. 31) also observed the sticky friends phenomenon in which “delinquent friends, once acquired, are not lost in subsequent years.” Specifically, among those who ever reported having delinquent friends, the typical sequence involved delinquent friends emerging at one age and then being reported again in every subsequent year of the study. This stability in delinquent peer association has been replicated by others studying different samples (e.g., Jennings, Higgins, Akers, Khey, & Dobrow, 2013), and many studies indicate that having delinquent peers affects later self-control (Meldrum, Young, & Weerman, 2012). This suggests that stable self-control during adolescence follows in part from stability in the type of friends—delinquent or nondelinquent—with whom an adolescent spends time.

In Focus 5.2

Is Self-Control Contagious?

Many studies point to how a person’s self-control is influenced by his or her social networks. This raises an interesting possibility: On some level, is self-control contagious? Can it be transmitted from one person to another, paralleling what is seen with a virus? In an interesting series of experiments, vanDellen and Hoyle (2010) found that the metaphor of contagion is more relevant to the study of self-control than we previously thought. Across five separate experiments, they found that watching or even thinking about someone with good self-control made people more likely to exert self-control themselves. For example, in one study, subjects were randomly assigned to think about a friend with either good or bad self-control. Those assigned to think about the friend with strong self-control persisted for a longer time on a handgrip task used to measure self-control in laboratory studies. In a second study, participants were recruited to take part in a “taste test,” but half of the subjects were given the role of observer. These observers were assigned to one of two conditions that involved small plates of chocolate chip cookies and carrots. In the experimental condition, observers witnessed an individual avoid the chocolate chip cookies and eat only the carrots (thus showing strong self-control), while in the control condition, this was reversed. VanDellen and Hoyle (2010) found that the observers in the experimental condition showed substantially greater self-control in a self-regulation task administered just a few minutes later.

This same basic pattern held in three other experiments—proximity to acts of self-control led to greater self- control among the subjects. One interesting aspect of this is that the experiments never involved the subject being socially pressured or coerced to use self-control. Instead, they simply were exposed—either by thought or in a face-to-face situation—to an act of self-control, and this by itself primed the participants to exercise self-control themselves. VanDellen and Hoyle emphasized that this pattern strongly goes against the notion that self-control is purely an individual struggle. How much we exercise self-control depends in part on how much self-control we are exposed to in our social networks, and, in that sense, self-control operates with an

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 8 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

element of contagion.

Persistent Environmental Characteristics: The Stability of Poverty

As others have impressively documented, poverty is highly consequential for child development, often because of its effects on such things as the emotional well-being of parents and the health and physical well-being of family members (Maholmes & King, 2012). Research indicates, for example, that family poverty increases a child’s problem behavior in part by increasing his or her parents’ stress, depression, and willingness to use harsh or physical forms of discipline (Rijlaarsdam et al., 2013). There are also important biological risks associated with chronic poverty during childhood. The neuropsychological deficits discussed in Chapter 4 that follow from such things as exposure to abuse and neglect, toxins like lead, and severe stress are all more likely under circumstances of intense and chronic poverty (Evans, Chen, Miller, & Seeman, 2012).

What makes this especially relevant to self-control stability is that for some children, exposure to poverty is a quite stable feature of their social environment—they were born into a poor family that remains poor over the course of their childhood and adolescence. This likelihood points to the fairly chronic nature of poverty in the United States. Indeed, for children who face poverty early in life, it often persists not simply during childhood but also into adulthood—the majority of children raised in households in the bottom 20% of income earners will be near the bottom of the income distribution as adults also (Pew Charitable Trusts, 2012).

Moreover, in adolescence—when individuals begin spending more time away from the household—poverty carries special risks associated with the neighborhood context. Chronically poor families often live in neighborhoods with high concentrated poverty, and this is especially true among racial and ethnic minorities—poor blacks and Hispanics are more likely than similarly poor whites to live in high-poverty neighborhoods (Massey & Denton, 1993). These neighborhoods have heightened problems of all kinds, including high unemployment, high population turnover, and weak social ties among neighbors. These factors in turn combine to produce greater levels of crime, drug use, and gang involvement (Sampson, 2013).

Ethnographic neighborhood research also documents cultural adaptations to these circumstances. Most notably, isolation from conventional institutions and opportunities can produce an oppositional street subculture that rejects mainstream values, instead prioritizing toughness and the maintenance of respect from others (Anderson, 1999). In these social contexts, individuals often must respond violently to any challenges and provocation. As Stewart and Simons (2006, p. 6) note, “the street code and the respect it demands is so entrenched among the hard-core, street-oriented individuals that they are willing to risk dying violently rather than being ‘dissed’ or victimized by another.”

To be clear, the power or prevalence of these cultural beliefs should not be overstated—in even the poorest neighborhoods, street-code values are interwoven with conventional values emphasizing hard work, self- reliance, and staying out of trouble (Anderson, 1999; Benoit, Randolph, Dunlap, & Johnson, 2003). That being said, it is easy to imagine how self-control development would be undermined by residence in a high- poverty neighborhood in which street-code values are common. Self-control involves overriding dangerous, harmful impulses, whereas the street code embraces such impulses, no matter how much they detract from future well-being. Indeed, a common theme in this literature is that hard-core street youth do not envision a bright future worthy of protecting (Drummond, Bolland, & Harris, 2011; Piquero, 2014). Thus, regulating one’s impulses in the present perhaps carries little payoff for the future. In support of this, several studies find that residence in poor, socially disorganized neighborhoods reduces self-control even after statistically controlling for the quality of parenting (Pratt, Turner, & Piquero, 2004; Teasdale & Silver, 2009). These patterns indicate that stable problems with self-control could follow in part from enduring poverty and consistent exposure to environments that discourage long-term considerations.

State Dependence as a Contributor to Self-Control Stability

Two explanations for self-control stability have been considered thus far. First, a latent individual trait may persist and manifest itself similarly over time. Second, features of the social environment may persist over

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 9 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

time, steadily pushing one in the same behavioral direction (independent of any individual trait). Importantly, however, these two ideas are not incompatible with one another—they may be true at the same time, and there may be an interesting connection between them. This possibility is central to the idea of state dependence, a commonly invoked concept in life course and developmental approaches to behavior (Nagin & Paternoster, 1991; Sampson & Laub, 1993).

A state dependence perspective begins with the recognition that some children will have low self-control and behave poorly from the earliest point in which such patterns can be detected. Such children may be especially restless and difficult to soothe as toddlers and inattentive and aggressive as young children. Perhaps this involves a latent trait, or it may simply reflect powerful effects of an existing social environment. In either case, some children stand out by virtue of their impulsive antisocial behavior. The defining quality of a state dependence view is that this bad behavior is expected to have a causal effect on the likelihood of bad behavior in later years—bad behavior at one point actually produces bad behavior at a later point.

How might that occur? The basic idea is that through patterns of social exclusion and self-selection, an individual’s behavior fundamentally shapes the social environment to which he or she is exposed. Those who behave poorly are treated differently by the people and institutions in their lives—their bad behavior provokes stigmatizing reactions and exclusion from prosocial opportunities. Moreover, poorly behaving children may actively exclude themselves from prosocial relationships and experiences.

This can be seen in reference to a hypothetical 10-year-old boy who shows all the signs of low self-control: he cannot stay quiet or sit still when he is supposed to, he loses his temper and throws tantrums when not getting his way, and he pushes other children to secure an object or get to the front of a line. Such a child will be treated differently by others. Parents and teachers will be frustrated by him and will more commonly use coercive or harsh forms of discipline that are counterproductive over the long term. Moreover, they will be less likely to become emotionally attached to him and therefore will invest less in his future. Other children—especially those who know how to regulate their own emotions and behavior—will be turned off as well. They will avoid playing with him, will not become friends with him, and may even ostracize him. Parents, teachers, and children alike may socially label him—they may define him as the problem kid or the stupid kid and treat him accordingly, and he may accept this self-identity and behave in ways that match it (Jussim & Harber, 2005; Rosenthal & Jacobson, 1968). All this contributes to a vicious cycle in which problems with self-control and social functioning persist, perhaps even growing worse. Moffitt (1993, p. 682) describes this process aptly: “It may well be that early behavioral difficulties contribute to the development of persistent antisocial behavior by evoking responses from the interpersonal social environment, responses that exacerbate the child’s tendencies.” Caspi, Elder, and Bem (1987, p. 308) comment similarly: “The child acts; the environment reacts; and the child reacts back in mutually interlocking evocative interaction.”

This pattern is likely to persist into adolescence and adulthood, with this child’s low self-control (and the behavior it produces) affecting much along the way. He is unlikely to develop strong commitments to school—his grades will be low, he may not graduate from high school, and he will forego college. The people he becomes friends with will be similarly impulsive and uninterested in school. Together, they may get involved in drugs, alcohol, and some degree of property and violent offending, behaviors that could give rise to contacts with the criminal justice system and further undermine his future. His low self-control and resulting behavior will also have implications for his relationship with his parents. That relationship may not have been strong to begin with—problems in the family environment likely contributed to his initial self- control deficits—but it likely will grow worse as the possibilities for parent–child conflict multiply during the teenage years. And in early adulthood, he likely will bypass two things that might have pushed him to a better path: steady participation in the labor force and the development of a long-term romantic relationship with a prosocial partner.

In this sense, his problems with low self-control are like a snowball rolling down a hill, gaining speed but also surface area as it accumulates more snow with each rotation. Consistent with this analogy, Sampson and Laub (1997, p. 21) describe state dependence as a process of cumulative disadvantage in which “deficits and disadvantages pile up” to produce “environmental traps” that restrict future options and encourage involvement in deviant subcultures. Similarly, Nagin and Paternoster (1991; 2000, p. 118), who have done much to clarify the implications of state dependence for criminal behavior, describe it this way:

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 10 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

It is a process of contagion in which an offender’s current activities make their life circumstances worse, accelerating the probability of future crime. For example, committing crimes can weaken or destroy one’s involvement in a network of conventional relationships (spouse, children, relatives, neighbors) that could have provided even partial restraint on criminal tendencies. Criminal acts committed now can also increase one’s risk of future crime by leading one into closer affiliation with other offenders.

In Nagin and Paternoster’s quote, each reference to crime and offending could easily be replaced with a reference to low self-control. The state dependence logic is the same regardless of which problematic outcome is emphasized—involvement in problem behavior is often stable because such behavior tends to increase exposure to antisocial roles and relationships that encourage further problems.

There is, however, one other interesting and important twist to the state dependence idea: In the same way that antisocial roles and relationships encourage low self-control, the unexpected emergence of prosocial roles and relationships can jump an individual to a different self-control trajectory. Thus, just as things can get worse, they also can get better. Such things as new friendships with prosocial others, entry into college or the military, the start of a career, or a good marriage can “increase social capital and investment in [pro]social relations and institutions” (Sampson & Laub, 1993, p. 21). These beneficial social experiences are less likely among those low in self-control, but they still may occur (Hay, Meldrum, & Piquero, 2013), and when they do, positive changes may ensue. We pursue this idea further in the next chapter, which focuses on self-control changes that have often been observed.

An Implicit Idea: Human Agency

The state dependence perspective includes an idea that is implicit in the discussion above but that merits explicit focus. This is the idea of human agency, which Laub (2006, p. 244) has defined as “the purposeful execution of choice and individual will.” Simply stated, human agency involves the idea that people are not merely passive entities governed by powerful forces beyond their control (such as a deeply rooted latent trait or adverse social environments). Instead, there is a degree of choice whereby we actively help create our own future. Indeed, individuals play a big role in creating their own environments. This idea is very much seen in the preceding paragraphs—those with low self-control make choices about such things as who to spend time with (e.g., delinquents or nondelinquents), whether to commit to education, and whether to cooperate with parents, and these choices in turn shape the types of experiences and relationships that characterize their lives.

Human agency comes into play in interesting ways for self-control, although we must speculate to some degree—quantifiable assessments of an abstract notion like human agency will always be difficult. That said, we believe that human agency often—but certainly not always—enhances the absolute and relative stability in self-control that is observed over time. This happens when individuals make choices and develop identities and commitments that reinforce their existing self-control tendencies. As a result, those with low self-control often remain low, those with high self-control often remain high, and the differences between the two groups are reinforced over time.

This role of human agency first arises when individuals become cognitively capable of purposefully and consciously embracing certain lifestyles. Under such circumstances, some individuals may actively adopt the principles of self-regulation. Being a thoughtful and deliberative decision maker (i.e., having high self-control) becomes a personal ethos that guides their self-identity and daily choices. These are individuals that “look before they leap,” so to speak, and they do so with conscious self-awareness. Should they prepare for an exam or blow it off? Should they exercise today or skip their workout? Should they drive home after drinking too much? Each question seems like a separate discrete choice, but through this process of human agency, there is something that ties them all together: a personal commitment to self-control, whereby a person sees himself or herself as the type of person who controls desires for immediate gratification, assesses actions in terms of the benefits they offer and the costs they impose, and then behaves in ways that protect long-term well-being.

Those with low self-control would be drawn to a different self-identity and approach to decision-making. Instead of seeing themselves as conscientious deliberators, they may actively appreciate the excitement

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 11 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

and thrills of taking unnecessary risks, rebelling against convention, and testing the boundaries of behavior. Especially for adolescents, such actions may be seen as a “liberating departure from the typical status of youths in which their behavior is constrained by adults” (Meldrum & Hay, 2012, p. 694). Rebellon and Manasse (2004) make this point in noting that adolescent peer subcultures may reinforce this identity by rewarding those who are willing to take risks, rebel against authority, and test the boundaries of allowable behavior. Thus, in the same way that some may embrace a personal ethos of high self-control, others may do the opposite. In so doing, they prioritize thrills over prudence, courage over caution, and quick action over careful deliberation. In this sense, human agency is exerted—the peers they select, the ways they spend their time, and the goals they pursue (or ignore) all come to reflect their own personal will and the identity they have embraced.

Two relevant lines of thought suggest that human agency comes into play in this way. The first involves Tittle, Ward, and Grasmick’s (2004) insightful distinction between one’s self-control ability and interest. The concept of self-control ability captures the traditional approach in which self-control is seen as a traitlike capacity that is fairly stable over time. Self-control interest, on the other hand, is more a matter of preference that can vary over time and across situations as individuals decide whether they are interested in using self-control. As part of this, some individuals will not exercise the self-control ability they possess simply because they lack an interest in doing so. Others may be the opposite—their strong interest in self-control leads them to make full use of any self-control ability they have. The analysis from Tittle and his colleagues (2004) supported this contention: Measures of self-control interest independently predicted involvement in crime (net of any correlations with self-control ability). This finding has key relevance for understanding the links between human agency and self-control: One’s observed level of self-control at least partially reflects the self-control interest that guides daily choices about whether to avoid costly temptations and whether to place oneself in social contexts in which temptations abound.

Silver and Ulmer (2012) also emphasize human agency in their discussion of future selves. They argue that research on self-control has not properly highlighted the ways in which self-regulation decisions are informed by people’s “conceptions of themselves in the future” (p. 701). From this perspective, a given course of action is appealing if it helps one achieve a desired future self. If that desired future self is oriented around a respectable and prosocial identity, then behaving impulsively is unappealing and perhaps even unthinkable. On the other hand, for those embracing a future self that is built around thrills, risk-seeking, and immediate gratification, it is the consistent exercise of self-control that is unappealing and unthinkable.

Moreover, and consistent with the emphasis on human agency, these differences are not just episodic in nature—they guide how people organize their lives. Silver and Ulmer (2012) emphasize that conceptions of future selves encourage the development of “commitment portfolios” by which people become committed to (a) specific courses of action, (b) moral rules that justify those actions, and (c) social relationships that reinforce those actions. This is reflected, for example, in a person whose future self rejects a personal ethos of self-control—as part of this, he or she may become committed to recreational drug use and may embrace moral values that justify this use (perhaps relating to the importance of personal liberties), and their friendships and romantic relationships may encourage further recreational drug use in the future.

Once again, the resonating idea behind all of this is human agency—humans are not merely puppets that are manipulated and determined by internal traits or external social environments. Instead, there is the possibility of personal reflection and choice as humans critically evaluate and construct the conditions of their lives (Emirbayer & Mische, 1998; Hitlin & Elder, 2007). As argued above, this role of human agency will often encourage self-control stability, as individuals develop identities, interests, habits, values, and relationships that reinforce their self-control tendencies. We do not suggest, however, that this is always the case. Quite the opposite, human agency may often be a catalyst for change—reflections on one’s life will sometimes encourage a fundamentally different course of action. In this sense, human agency is not just about continuing a given trajectory—it also sometimes involves constructing a new one.

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 12 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

In Focus 5.3

Human Agency, Self-Control, and the Four-Minute Mile

Athletics provide a great forum for seeing self-control in action. A fascinating example of the intersection between human agency and self-control comes from the human quest to run a mile in less than four minutes.2 Few activities require greater self-control than pushing your body to maintain a fast running pace after the body has seemingly edged up against its maximum capacity for exertion. In the mid-1900s, that maximum capacity in the mile rested right around the 4:00-minute mark. In 1942, the record stood at 4:06.4, but it was broken six times over the next few years, and by 1945, the record was 4:01.4.

Fully nine years later in 1954, however, mile runners worldwide had gotten no closer to a sub-4:00 mile, despite a zealous pursuit of a breakthrough. A trio of runners were most visible in this pursuit: Australia’s John Landy, England’s Roger Bannister, and America’s Wes Santee. The case of Landy is particularly interesting. Between 1952 and 1954, he ran between 4:02 and 4:03 fully six times, but he could never go lower. In one race in December 1953, Landy was indeed on pace for a sub-4:00 mile, but he seized up a bit in the final 200 meters and finished at 4:02 again. After the race, he famously declared to reporters, “I feel I could go on for 10 years, but I don’t think it’s worth it. Frankly, I think the four-minute mile is beyond my capabilities. Two seconds may not sound [like] much, but to me it’s like trying to break through a brick wall.” This was an astounding declaration—one of the most accomplished middle-distance runners in human history was essentially giving up, conceding that he could not imagine running two seconds faster.

Five months after Landy’s sobering declaration, Roger Bannister stepped on to a track at Iffley Road in Oxford, England, and ran the mile in 3:59.4. The human quest for a sub-4:00 mile had been achieved, and Bannister earned himself a permanent place as one of the great athletes in history. But where did this leave Landy? His pursuit of the four-minute mile was decidedly not yet over, despite his earlier claims. Bannister’s landmark performance had a powerful motivating effect on Landy, one that we see as comparable to a surge in Landy’s self-control interest. Bannister had shown Landy that humans were capable of a sub-4:00 mile. Bannister had also struggled to get over the 4:00 mark, but he had broken through. Why couldn’t Landy as well? The “brick wall” had been broken—Landy needed to persevere for four laps in a way he never had before and bound over the wall himself.

Little more than a month later, at a race in Finland, Landy smashed Bannister’s record with a 3:57.9 mile. After an unsuccessful three-year quest to shave two seconds off his time, Landy improved by four seconds in the span of just six months. Landy’s breakthrough almost certainly cannot be explained in purely physiological terms (Tucker & Dugas, 2009). Instead, it followed in part by the removal of a mental barrier, a barrier that we see as a self-imposed limit on his self-control. That limit was built upon Landy’s belief that a sub-four mile was not accomplishable. Once the limit was removed, a record-setting performance followed.

Empirical Evidence on Explanations for Stability

The three explanations for behavioral stability offered above—a persisting latent trait, persisting social environmental factors, and the integrative idea of state dependence—have been assessed empirically. The idea of persisting social environmental factors (regarding such things as parenting, peers, and poverty) is widely accepted (some of this research was described above). Much research has focused instead on comparing the accuracy of the latent trait and state dependence perspectives. These two have been pitted against one another in large part because of a key contrast on the question of whether social roles and relationships are potent causes of self-control and behavior over the life course. The latent trait perspective is interpreted as saying they are not—once the latent trait emerges (as indicated perhaps by early self-control deficits), it drives all subsequent outcomes. Any statistical relationships between social roles and relationships (like peer relationships or problems at school) and later self-control and behavior are spurious—these things are correlated because they all follow from the same cause (the latent trait/early self-control deficits). The state dependence perspective, on the other hand, predicts that these social roles and relationships do have causal significance—they help causally explain the relationship between early self-control deficits and later self-control and behavior. Specifically, early problems with self-control lead individuals to self-select into

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 13 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

antisocial roles and relationships, and once they do, those antisocial influences have causal significance of their own—they encourage further problems with self-control and behavior.

So, what does the research indicate? On balance, the research supports a mixed version of a state dependence perspective—one that acknowledges the importance of a latent trait but also allows for an important effect of adolescent and adult social roles and relationships. This general conclusion is supported across a wide variety of studies. An especially influential study from Wright, Caspi, Moffitt, and Silva (1999) used data from a birth cohort in New Zealand. They had comprehensive, multimethod measures of self- control for childhood (ages 3 to 11) and adolescence (ages 15 to 18); they also had measures of social roles and relationships and criminal behavior during adolescence and young adulthood. They found an unmistakable pattern: Low self-control in childhood led to antisocial roles and relationships in adolescence (e.g., poor relationships with family, weak commitment to school, and association with delinquent peers). These things, in turn, led to criminal behavior in young adulthood, even after statistically controlling for the self-control deficits in childhood that got everything going to begin with. This pattern is consistent with a state dependence view on the importance of social roles and relationships.

Other findings from Wright and his colleagues (1999), however, show the value of a latent trait perspective that emphasizes the enduring importance of early levels of self-control. Most notably, there was relative stability in self-control that existed independently of the effects of social roles and relationships—self-control deficits in childhood were followed by self-control deficits in adolescence, and this was followed by crime in young adulthood. This life course continuity in self-control and resulting behavior from very early in life points to the powerful influence of a latent trait (even if it is not so powerful as to crowd out the effects of social roles and relationships later on).

Similar findings have been reached in other studies with independent measures of self-control, social roles and relationships, and criminal behavior (Hay et al., 2013; Longshore, Chang, & Messina, 2005; Nagin & Paternoster, 1994). Low self-control increases the chances that adolescents associate with delinquent peers, are weakly committed to school, and get along poorly with parents; these things, in turn, have an independent effect on delinquency even after accounting for initial levels of self-control. Self-control typically has effects of its own, however, and self-control often is quite stable over the study period, even after accounting for the social roles and relationships noted above.3

Taken together, this line of theorizing and the related research give a clear picture of why individual self- control remains relatively stable from childhood into adolescence and then young adulthood. Part of the explanation is that children’s initial level of self-control is linked to a stable latent trait. Because of a complicated mix of genetic, biological, and social environmental forces, having low (or high) self-control becomes a fairly enduring part of who that individual is, and it sticks with that person as he or she advances through the different stages of life. Another important contributor, however, is the process of state dependence, whereby self-control remains stable because early levels of self-control lead to social roles and relationships that reinforce the earlier patterns. For example, those with low self-control become friends with delinquents, dismiss the value of educational success, and develop stormy relationships with parents as they enter adolescence—all of these are likely to further erode their self-control. Those with initially high self- control, on the other hand, will do the opposite in these key arenas of life. The net effect is that initial levels of self-control will often lead to consistency in the antisocial or prosocial nature of the experiences, roles, and relationships to which an individual is exposed. This in turn encourages a high level of self-control stability in which the differences between individuals are maintained over time.

Policy Implications and Possibilities

This chapter has focused on the significant stability in self-control observed from childhood into adolescence and then adulthood. On the face of it, this seems to offer few prospects for policy—children with high self- control remain that way, those with low self-control do the same, and policy efforts will make little difference. Quite the opposite is true, however; we see three notable policy implications that follow from the research described here.

The first is that we must remember that although stability in self-control is quite high, it is far from

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 14 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

perfect—self-control changes do occur, and major shifts over the life course may occur among as much as 15% to 20% of the population. These self-control reversals are the focus of the next chapter, and we will discuss the policies and programs that can bring about reversals among children who enter adolescence with less-than-optimal levels of self-control.

A second major implication of self-control stability is that it underscores our earlier emphasis on early childhood prevention. Without question, the best policy approach involves preventing self-control deficits from ever emerging in the first place. If this can be done, self-control stability over the life course will be working to society’s benefit rather than its detriment. This directs our attention back to the programs that establish good outcomes in the early stages of life, such as home visitation programs like the Nurse–Family Partnership, family training programs like Triple P or Incredible Years, and preschool and school enrichment programs like the Abecedarian Project or the PATHS curriculum.

A third major policy implication involves those adolescents who enter adolescence seemingly doing just fine. We might be tempted to think that these individuals can be left alone—through processes of state dependence and human agency, they will naturally gravitate toward prosocial experiences that reinforce existing self-control levels. Thus, these are the cases in which self-control stability will naturally work in our favor. Although this may often happen, the threat of change is always there, and this follows in part from the “storm and stress” possibilities of adolescence that we discuss in the next chapter. The key message is this: If we want self-control stability to work to society’s benefit, we must take concrete steps to encourage self- control in adolescence, even among those high in self-control to begin with.

The promising programs in this regard are often school-based life skills programs that are universal in scope; rather than targeting only those adolescents already showing signs of behavior problems, they target the broader adolescent population. Thus, while they may promote reversals among those entering adolescence with self-control deficits, they can also encourage self-control stability among those relatively high in self- control. Two programs in particular have generated impressive evaluation results and focus on skills relevant to self-control. The first of these is Life Skills Training (Botvin, Griffin, & Nichols, 2006), a classroom-based middle-school program that uses 30 sessions taught over a span of three years. The program is designed to reduce substance use and risky behavior by promoting social skills and self-regulation habits (e.g., setting personal goals and self-monitoring one’s progress, assessing problem situations, considering the consequences of different actions) that can lead to better decision-making. The Positive Action program is quite similar (Lewis et al., 2013). It is implemented on a schoolwide basis to encourage a school culture that reinforces prosocial values and behavior. In the middle school years, it includes roughly 80 lessons that last about 15 to 20 minutes each and that provide guidance on a wide variety of skills relevant to self-control, including problem-solving in difficult situations, using good self-management skills (setting standards and self- monitoring), and making decisions based on a future orientation.

Conclusion

Do people stay the same over time, or do they change? The major theme of this chapter has been that when it comes to self-control, there is a good amount of both continuity and change. Studies using a wide variety of samples and research designs indicate that a person’s absolute and relative levels of self-control remain fairly stable through adolescence and adulthood, but changes occur also. After describing these results, we presented the major explanations for the stability that is often seen: (a) a persistent individual trait explanation in which self-control remains stable because it reflects a deeply rooted quality that is an enduring part of who that person is; (b) a persisting social environments explanation in which the social environments that influence self-control are themselves quite stable over time, and (c) a state dependence explanation that combines the first two in key respects. Research generally supports a state dependence perspective in which persisting individual traits affect later outcomes, in part by influencing the types of social environments and relationships a person experiences.

For many individuals, this will involve a reinforcing cycle of behavior, one in which self-control tendencies from the first decade of life give rise to social experiences, roles, and relationships that reinforce those initial tendencies. For those coming out of childhood with high self-control, this cycle increases the chances of smooth and prosocial transitions to adolescence and adulthood. On the other hand, for those with initially

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 15 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

low self-control, this will be a vicious cycle fraught with problem behavior, life hassles, and unexpected complications. We should not, however, overstate the predictability of these patterns. As we discussed, there are policy tools at our disposal—effective programs can intervene to encourage the use of self-control skills. Moreover, as we convey in the next chapter, opportunities for self-control change still abound, even for those who get off to a rough start in childhood and adolescence—a person’s self-control is never set in stone.

Discussion Questions

• 1. What is meant by the phrase past behavior is the best predictor of future behavior? How can this be extended to a focus on self-control?

• 2. Consider three friends: Jim, Jake, and Josh. In the third grade, Jim has a self-control score of 15, Jake has a score of 20, and Josh has a score of 25. By the time the three boys reach eighth grade, Jim has a score of 30, Jake has a score of score of 25, and Josh has a score of 20. How many of these boys have experienced absolute increases in self-control? What about absolute decreases? Has each of the three boys experienced relative stability in self-control?

• 3. Describe the idea of state dependence and how it encourages self-control stability.

• 4. What is human agency, and how might it contribute to self-control stability?

• 5. What is the difference between self-control ability and interest? Describe some hypothetical situations in which someone who is high in self-control may nonetheless be uninterested in exercising it.

Notes

1. Some researchers argue that stability is even more common than these figures suggest. According to this view, measurement complications make it difficult to observe the full extent of stability. When subjects complete self-control surveys, they may at times not pay attention to the questions as much as they should; moreover, they may sometimes answer in biased ways, depending on random circumstances. Also, a parent’s assessment of a child’s self-control could be especially influenced by a single recent event that has exaggerated importance in their minds. With these things occurring, observed shifts in self-control over time could follow from measurement error. Another complicating issue is that the manifestations of low self-control can vary across different ages. For example, whining and throwing tantrums may be common among 6-year- olds, whereas substance use and risky sexual activity are impulsive behaviors that become possible only in adolescence. If measures of self-control are not adapted over time, children and adolescents may appear to show changes in self-control that instead reflect heterotypic stability, which occurs when traits express themselves in different forms over time (Wright, Tibbetts, & Daigle, 2008). We concur that these factors come into play, but as we emphasize in the next chapter, we also believe that shifts in self-control do occur—studies with impressive methodological rigor indicate as much.

2. Our discussion draws in large part from Tucker and Dugas’s (2009) excellent retelling of this history in The Runner’s Body.

3. Many other studies have reached similar conclusions with different datasets, often with quite sophisticated analyses that try to account for the effects of an initial latent trait that is not expected to be directly observable. This is quite the analytical challenge—if we cannot observe and measure the latent trait, how can we statistically account for its effects? A common approach tracks individuals over time to see if the changes in social roles and relationships that they experience correspond to shifts in their own behavior (Bushway, Brame, & Paternoster, 1999). For example, if they experience an increase in association with delinquent peers, does this lead to corresponding decreases in self-control and increases in crime? In analyses like this, each individual serves as his or her own control—examining individual changes over time controls for time- stable latent traits that affected their initial circumstances and behavior (Bushway et al., 1999). Studies of this

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 16 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?

kind often indicate that shifts in social roles and relationships in adolescence and adulthood are associated with shifts in observed behavior even after accounting for observed and unobserved individual differences (i.e., differential exposure to a latent trait; Horney, Osgood, & Marshall, 1995; Nagin & Paternoster, 1991, 2000).

http://dx.doi.org/10.4135/9781483397726.n5

SAGE © 2016 by SAGE Publications, Inc.

SAGE Books

Page 17 of 17 SAGE Books - Adolescence and Adulthood: Is Self-Control Stable Over

Time?