In my eternal quest to understand the causes and consequences of low self-control, I have focused my gaze as of late on childhood adversity (abuse, neglect, and household dysfunction). While there is a large literature linking adverse childhood experiences (ACEs) with a myriad of negative outcomes related to health and behavior, relatively little attention has been directed at considering the possibility that greater exposure to ACEs could be detrimental to self-control. Drawing on both the social science and neuroscience literature, I am working on a study that assesses this association across two large independent samples of adolescents. The pattern of results is clear: greater exposure to ACEs is positively associated with low self-control during adolescence. Given these findings, this area of research appears to be ripe for expansion.
Given recent dialogue, both in the popular press and in the academic literature, surrounding police misconduct, an interesting empirical question is whether police officers differ in their levels of self-control relative to offenders. As protectors of the public who stand on the opposite side of the law from offenders, it seems reasonable to expect, for a number of reasons, that officers should have much higher self-control than offenders. Considering this issue directly, however, remains illusive, in part because a study in which the self-control of police officers and offenders has not been conducted. Yet, there have been separate studies assessing the self-control of officers and offenders.
Drawing on separate samples of male officers and male offenders (who served time in prison) from prior studies in which self-control was measured using the Grasmick et al. (1993) scale, I have created a single data set which includes information on the age, race, education, and self-control (including the sub-scale scores) of the officers and offenders. The results thus far support the expectation that officers, on average, score much higher than offenders on global self-control. Yet, there is overlap in the distribution of scores between officers and offenders (i.e., some offenders score higher in self-control than officers). Further, some interesting patterns emerged when comparing the six self-control sub-scale scores between officers and offenders, with possible implications for hiring and screening practices of prospective officers. I hope to turn this fun exercise into a manuscript for publication with the individuals who originally collected the officer and offender data as co-authors.
A large number of studies show that spending time with friends, with no particular agenda, without parents around (i.e., unstructured socializing), increases risk of substance use and delinquency among teens. To my knowledge, however, no one has considered where the risk begins. In other words, how many hours per week spent engaging in unstructured socializing does it take to put teens at risk of antisocial behavior? One? Five? Ten?
This has some obvious practical implications. If it takes 10 hours, then only a small percent of the adolescent population would be at risk of substance use and delinquency resulting from unstructured socializing, as less than 20% of adolescents report spending more than 10 hours per week engaged in unstructured socializing. If, on the other hand, risk starts at only a couple hours per week engaging in unstructured socializing, then a much larger percent of the adolescent population could be at risk, as 75% of teens report spending at least one or two hours per week engaging in unstructured socializing.
I recently began analyzing data on about 8,000 Florida middle school and high school students concerning time spent in unstructured socializing and substance use. What I found is that for some substances, particularly marijuana use, spending a mere 1-2 hours per week with friends with no adults present doubles the risk of using pot compared to teens who report spending 0 hours per week in unstructured socializing. For other substances (tobacco and alcohol), it takes only 3-5 hours per week to pose a risk. For all substances examined, risk increases incrementally after the minimum hourly threshold is met. For example, a teen who spends 20 or more hours per week with friends with no adults around is 5x (500%) more likely to report recent marijuana use relative to a teen who spends 0 hours with friends unsupervised.
It is worth noting that these associations were found even after accounting for age, race, sex, self-control, parental monitoring, neighborhood disorder, peer substance use, truancy, peer attitudes, parental attitudes, and personal attitudes about substance use. While preliminary, these results reinforce the immediate and opportunistic nature of substance use when in a setting with friends where there are no authority figures to monitor behavior. I will be presenting these results at the ASC Conference in November.
I am excited to say that I have partnered with a couple of psychologists at FIU to study whether self-control alters the association between drinking alcohol and violent behavior. We are using an experimental design in which some participants will be randomly assigned to consume what amounts to about four alcoholic drinks (vodka mix), while others will only be assigned to drink juice where the rim of the glass is coated in vodka (placebo). All participants will then respond to scenarios in which they have an opportunity to be physically violent toward someone else in response to being provoked. The two key questions are: 1) Will the participants who consumed the alcoholic drink be more likely to state intentions to be aggressive? 2) Does this association depend on the level of self-control in the participants? Self-control is being measured using the Tangney et al. (2004) scale before participants are randomly assigned to their drinks. Data for this study will be collected over the next year. Findings may help to explain why getting drunk is more likely to lead to violence in some people relative to others.
I continue my travels down the exciting rabbit hole that is biosocial criminology. Most recently, I have collaborated with researchers from the University of Michigan to investigate the association between anterior cingulate cortex functioning (ACC) and both self-control and delinquency. Using fMRI data to assess ACC functioning, we find an indirect effect of ACC functioning on later delinquency that is mediated by self-control. I am looking forward to sharing these exciting results with students taking my biosocial criminology course this fall.
With some extra time at the ASC conference, I conducted a crude content analysis of the ASC Program index entries and created a word cloud. If you are not familiar with word clouds, they are a way to visualize how frequently words are used in some type of document. I tabulated the frequency of index entries to create the cloud. I am thinking of maybe conducting an across-time trend analysis and writing a piece for The Criminologist. More broadly, using word clouds is a great way to introduce students to qualitative data analysis.
Having prepared and taught a biosocial criminology course here at FIU this spring, I became aware of a vast array of studies linking biomarkers (e.g., heart rate, skin conductance) to antisocial and violent behavior. One of these biomarkerrs is the r2D:4D ratio, which is a measure of the relative length of a person's right pointer finger (2D) to their right ring finger (4D). This ratio provides an indirect indicator of the amount of testosterone someone was exposed to while in the womb. Recently, criminologists have started to take an interest in the association between the r2D:4D ratio and antisocial behavior, with studies finding that adults who have a smaller r2D:4D ratio are more likely to report involvement in violent behavior (Ellis and Hoskin, 2013; Hoskin and Ellis, 2014). Given the infancy of this line of inquiry, I have sought out Anthony Hoskin and proposed that he and I collect new data to investigate unexplored questions. He happily agreed and we will be collecting new data in the fall to examine, among other things, the following questions: To what extent does the r2D:4D ratio correlate with violent victimization? Are various measures of self-control correlated with the r2D:4D ratio? Does the r2D:4D ratio moderate the effect of holding street code values on violent offending? We anticipate obtaining data from at least 400 individuals using a combination of survey methods and by taking high resolution scans of the right hands of participants to get a very precise measure of the r2D:4D ratio. We will present our preliminary findings at the ASC conference in November in Washington, D.C.
A Retrospective Study on Parental Self-Control, Family Environments, Young Adult Self-Control, and Young Adult Offending
I recently had an IRB protocol approved to collect data on a sample of about 500 college students. I will be examining how parental self-control is related to current young adult offending behavior, and how family environments during high school and current young adult self-control mediate this relationship. What is unique about this particular study is that young adults will be responding to items about the self-control of their parents (e.g., how often their parents would get drunk, use foul language, fail to wear seatbelts, etc.) in addition to their own self-control, which has not been done before. In piloting the 8 items this last spring, I found they had good internal consistency (alpha = .77) and a 1-week test-retest reliability of .85. In that same pilot study, parental self-control correlated with young adult self-control at r = .42. If anyone would like to see the survey instrument being used in the upcoming study, just let me know. Data will be collected during the last week of August and I hope to have it all compiled and will start running analyses in early September.
This semester I added a new assignment to my graduate research methods course - I had students conduct in-person survey interviews with 20 strangers using a questionnaire I had developed focused upon self-control and driving behavior. With 14 students in my class we ended up with data on 280 adults from Dade and Broward counties here in south Florida. Though only a simple exercise in introducing students to the experience of collecting data (and having to do so face-to-face with participants), some cool results emerged:
1. Participants who rated themselves as lower in self-control were more likely to report that they text while they drive.
2. Participants who rated themselves as higher in self-control were more likely to report a) wearing seat belts, b) using a turn signal while changing lanes, and c) using a turn signal while making turns.
3. While participants rated the driving ability of the average driver to be 5.1 on a scale from 1-10, participants rated their own driving ability to be 7.8.
Today I came across the phrase "The Dark Triad" when browsing through studies on personality. Catchy, isn't it? Turns out, the phrase refers to three overlapping personality traits: narcissism, Machiavellianism, and psychopathy. According to Paulhus and Williams (2002), "To varying degrees, all three entail a socially malevolent character with behavior tendencies toward self-promotion, emotional coldness, duplicity, and aggressiveness." Recently, Jonason and Webster (2010) developed a 12-item measure of the Dark Triad they called "The Dirty Dozen." By all appearances the scale has solid psychometric properties. I think I may have to test it out on a sample of students and see how scores on the scale might be differentially related to different crime typologies (property vs. violent offending). Given that higher scores on the scale would represent a person who is self-interested, lacks consideration of the feelings of others, and pursues goals through means of aggressive action, an argument could be made that a person possessing such a constellation of traits may be likely to engage in fraudulent activity and other forms of criminal behavior. Time to do a little branching out.