#p4e

Hypothesis: The authors are examining at what age “adult behaviors” (here defined as drinking alcohol, having sex, dating, driving, working for pay, and going out without parents) occur in different generations and through different economic systems. They argued that people who grew up with economic and social instability would likely take a “fast path to adulthood”: having children and engaging in adult behaviors sooner. On the other hand, they argued that people with more stability were more likely to delay gratification, taking a “slow path” to adulthood where they engage in these behaviors later, with more focus on extended education.

Methods: For the purpose of this research, the authors examined three surveys examining children and young adults back as far as 1966. The surveys used were Monitoring the Future (MtF); the Youth Risk Behavior Surveillance System (YRBSS), distributed by the Center for Disease Control; and the American Freshman (AF) survey. All together, these surveys examined over eight million children and young adults.

MtF examines people in 8th, 10th, and 12th grade, as well as young adults. It has data on 12th graders through 1971 and on the other groups through 1991. MtF randomly selects middle and high school populations to sample and asks questions surrounding drug and alcohol use, as well as questions surrounding paid employment, going out with friends, dating behavior, driving, and extracurricular activities. In addition, it records information on race, gender, and parental education level—the authors chose to divide children based on their father’s education level, with children whose father had a high school degree or less defined as low SES [a], and those with fathers possessing a higher degree defined as high SES. The research also breaks students down into urban, suburban, and rural regions.

YRBSS is a government-funded survey, sampling 9th-12th graders every other year to ask questions about risky and sexual behaviors. This research has been ongoing since 1991. For the purposes of the study at hand, the authors exclusively looked at the questions surrounding sexual behavior. This data did not report on SES, exclusively reporting race and gender.

The AF survey reports on incoming college freshman, with data going back since 1996. This survey examines freshman work experience, alcohol use, and extracurricular activities. It exclusively breaks students down by gender.

In addition to these surveys, the authors examined basic cultural information: number of children born per woman, median household income, unemployment rate, life expectancy, teen birth rate, mean age of first birth, college enrollment, prevalence of disease, and crime rates.

Results: As anyone who has read the multitudes of headlines can infer, younger people engaged in fewer “adult activities” than people in past generations. They were significantly less likely to go out, with 12th graders in the 2010s going out less than 8th graders in the 1990s. They were less likely to go on dates or have sex, with 54% of high schoolers having sex in 1991 to only 41% in 2015. In addition, the number of adolescents who drank alcohol declined massively, with a 59% decrease in the number of 8th graders who tried alcohol and a 26% decrease in 12th graders. There was also a substantial decrease in the number of children who had worked for pay, from around 75% in 1976  to around 55% in 2014.

While these declines hit certain groups harder (e.g. sexual activity for black high school students declining from 82% to 49%), it was fairly stable among SES, race, and location. Interestingly, the authors did not find that these decreases coincided with an increase in hours spent studying, and saw only a slight increase in hours of community service among 10th and 12th graders.

Finally, the authors looked at how cultural changes correlated with these decreases. They found that adolescents engaged in fewer adult behaviors when, during their childhood, the average family size was smaller, there was a lower life expectancy, there were more diseases, and the household income was lower. They found this correlation to be higher versus the correlation between these rates when the adolescent was engaging in the behaviors, supporting the authors’ argument that childhood experiences influence a person’s desire to reproduce earlier because of instability and insecurity.

Concerns and Important Issues: Like I stated above, I don’t like this study. The minute you reduce all human behaviors to reproduction, you’re walking a very dangerous line—what about the increase in same-sex sexual relationships? Is a teetotalling  asexual who lives in New York City (and therefore has no reason to learn to drive) never an adult? The one thing this study seems to ignore is the growing decrease in emphasis on reproduction, especially among younger people. As women grew more liberated and the focus shifted from a nuclear, heterosexual family to a fulfilling life with or without children, these sort of evolutionary arguments seem to be trying to force old theories into a culture that doesn’t have the same values or pressures.

Indeed, this focus on older values and ideas is made even more obvious when you consider that the authors defined a family’s SES exclusively by the father’s education level. Women are working more and more outside the home, and their education level was presumably listed as well—the authors just chose not to look at it. I’ve worked with data analysis to know it wouldn’t be that hard to combine the two, so refusing to do so was certainly a choice.

This paper was also an excellent example of something that tends to get to me a fair bit: misleading figures. As an example, look at this:

On first glance, it looks like there’s been a massive decline in, say, work for pay: almost half the figure! But if you pay attention to scale, it’s actually only a change of 20 percentage points. The authors chose to cut the scale between 95% and 50%, making numbers look much lower and like there was a much larger gap between them. Always check scales-- it’s important.

In the end, the arguments in this paper seem better explained by a changing culture, not by theories assuming all human behavior can be traced back to evolution. Of course, none of this is what the media cared about—they were more interested in the statement that millennials are engaging in fewer adult behaviors.

 Jargon Definitions:

a. SES= Socioeconomic status, aka “class”

Reference: