methods

Erik Ingram via flickr.com.
Erik Ingram via flickr.com.

Public opinion polling in America dates back to the 1930s, and religious beliefs and behaviors have always been topics of interest. However, polls do not produce perfect measurements and their results can both shape and misrepresent the reality of public beliefs and behaviors. Religion Dispatches recently interviewed sociologist Robert Wuthnow about his new book Inventing American Religion: Polls, Surveys, and the Tenuous Quest for a Nation’s Faith in which he details the ways in which polls have shaped the religious makeup of the American population.  

Wuthnow explains that in the 1970s, the Gallup poll’s overly broad and vague questions about evangelical identity resulted in a huge overestimation of the number of evangelicals in America at the time. He explains:

The year of the evangelical was 1976, when Jimmy Carter achieved election to the White House and the role of polling was to greatly expand the number of Americans who were evangelicals… How is that possible that a poll could do that? …Gallup said there might be 50 million Americans who are evangelicals, and journalists ran with that… The way they got 50 million was basically inventing a new question that said something to the effect of, “Have you ever had a born-again religious experience, or something similar to a religious awakening?” And that was pretty much it… There was a lot of diversity among evangelicals themselves that got masked by being lumped together in the polls as if they were all the same thing.

The results of that 1976 poll influenced not only perceptions of the number of evangelicals among journalists, but also political perceptions of evangelicals as a voting bloc. If there were that many evangelicals in America, then politicians could cater to them to gain votes. Evangelicals began to be treated as one homogenous group instead of the highly heterogenous group that they really were.

Wuthnow explains how a similar phenomenon is happening with the rise of the non-religious in America—commonly known as the “nones.” Polls and surveys are reporting an increase in those who are saying they are not religious, but these general trends can mask huge variation among the non-religious. Many, for instance, still believe in a god and go to church, and only a very small percentage actually identify as atheist. Further, Wuthnow says, people often change their beliefs and opinions over time, and poll data glosses over that fact:

Polls have always assumed that whatever a person says is reliable, and that they really mean it and stick with it. So if you find any changes over time, that’s significant because polls are measuring stable opinions—if those change, that’s important.

You've got to know how your product is used. Photo by FourTwentyTwo via flickr.com.
You’ve got to know how your product is used. Photo by FourTwentyTwo via flickr.com.

The era of bothersome consumer surveys and robo-calls may be coming to a close, as these shallow techniques of data collection just don’t cut it in the information age. In a recent article in The Atlantic, Graeme Wood describes a growing trend in market research: big business hiring social scientists to do fieldwork. Corporations have long researched the quantitative aspects of their sales, but qualitative knowledge about the use of the products has been somewhat limited. Social scientists and those business researchers known as “consumer behavior” (vs. “quantitative”) economists—long since part of the business discussions within academia—are now being hired to uncover how products are used, as well as who uses them and how those users feel about the products.

After realizing that they new little about the home consumption of their product, for instance, Absolut Vodka commissioned ReD, a forerunner in what we might think of as anthropological market research, to study the home party scene and the rituals and norms of drinking. One consultant on the project, former Yale anthropology Ph.D. student Min Lieskovsky, noted some party trends that Absolut quickly applied to their marketing:

‘One after another, you see the same thing,’ Lieskovsky told me. ‘Someone comes with a bottle. She gives it to the host, then the host puts it in the freezer and listens to the story of where the bottle came from, and why it’s important.’ And then, when the bottle is served, it goes right out onto the table with all the other booze, the premium spirits and the bottom-shelf hooch mixed together.’

The quality and status of the liquor seemed to be much less important to the consumer than their personal association with it. Despite years of market research, without this use of social science, the social significance and human connection of the product might have gone overlooked—and fewer bottles of Absolut might have gotten sold.

The United States has a greater share of its population behind bars than any other nation. Yet this captive audience is almost never captured by large national surveys used to study the U.S. population. This might distort what we think we know about black progress in recent decades, the Wall Street Journal reports, because a large enough swath of the young African American male population is incarcerated and unaccounted for by these surveys.

Among the generally accepted ideas about African-American young-male progress over the last three decades that Becky Pettit, a University of Washington sociologist, questions in her book “Invisible Men“: that the high-school dropout rate has dropped precipitously; that employment rates for young high-school dropouts have stopped falling; and that the voter-turnout rate has gone up.

For example, without adjusting for prisoners, the high-school completion gap between white and black men has fallen by more than 50% since 1980, says Prof. Pettit. After adjusting, she says, the gap has barely closed and has been constant since the late 1980s. “Given the data available, I’m very confident that if we include inmates” in more surveys, “the trends are quite different than we would otherwise have known,” she says.

Voter turnout is another example, especially in light of this year’s presidential election.

…commonly accepted numbers show that the turnout rate among black male high-school dropouts age 20 to 34 surged between 1980 and 2008, to the point where about one in three were voting in presidential races. Prof. Pettit says her research indicates that instead the rate was flat, at around one in five, even after the surge in interest in voting among many young black Americans with Barack Obama in the 2008 race.

“I think that’s kind of stunning,” Prof. Pettit said.

Experts debate the feasibility of including prisoners in such surveys, as well as how to make the best use of available data. Even Pettit admits, “These are really, really tricky things.”

 

In a recent Boston Globe story, diversity consultants and social scientists debated the effects of workplace diversity training.  While diversity training programs are a common job requirement these days, they may look very different from one company to the next:

The courses vary widely, in content and duration and method and philosophy: Some are short videos followed by structured discussions, some are multiday retreats, some are informational, teaching participants about their “diversity circle” and the difference between a generalization and a stereotype, others focus on role-playing. But they all promise to help people better navigate the fault lines of race, gender, culture, class, and sexual orientation that can divide co-workers and unsettle offices.

Opinions about the programs are also varied, and good social science evidence for either side of the debate has been scarce:

Such programs have always been controversial, with critics arguing that they’re unnecessary and needlessly politicize the workplace. But despite the growth and prevalence of diversity training, there have been few attempts to systematically study it.

Now a few social scientists are taking a hard look at these programs, and, so far, what they’re finding is that there’s little evidence that diversity training works.

Research by a team of sociologists on more than 800 companies over three decades has found that the best diversity training programs make little difference in who gets hired and promoted, and many programs actually decrease the number of women and minorities in management.

“Even with best practices, you’re not going to get much of an effect,” says Frank Dobbin, a Harvard University sociology professor on the research team. “It doesn’t change what happens at work.”

Diversity consultants are confident in their programs, claiming social science research in this area can’t accurately measure the impact of the training they deliver, generalizes unfairly, and rarely offers solutions to the problems it identifies:

Practitioners and some scholars disagree, arguing that, while there have been some unsubstantiated claims and overhyped “innovations” in diversity training, the field as a whole has begun to figure out what works. The changes that training triggers can often be subtle, defenders argue, and, in a setting as dynamic and stubbornly multivariate as the workplace, it’s all but impossible to come up with the clear, falsifiable evidence social science demands. The poor results that do show up in broad-based studies, they say, are due to companies whose commitment to diversity training programs is merely pro forma, and who see training as just a way to protect themselves from lawsuits.

“My experience is that a lot of these studies make good points, but they tend to fall into one particular trap,” says Howard Ross, a leading diversity consultant. “When we talk about diversity training as a megalith, it’s similar to saying, ‘Are restaurants good places to eat?’ The answer is ‘yes’ and ‘no,’ depending on the restaurant.”

Critics, on the other hand, argue that today’s practitioners are unlikely to be converging on a set of best practices, since the field is characterized by divergent, even contradictory approaches to the same set of problems. To critics, the proponents are simply mistaking the fact that people feel better about themselves after training for real results. Just because people think they’re less prejudiced doesn’t mean they are. Indeed, with something as subtle and reflexive as bias, we’re often our own worst judges

Dobbin and his colleagues have designed their research to address the potential alternatives to conventional diversity training programs practitioners often call for:

“We were increasingly frustrated by the fact that we know a lot about what kinds of disparities there are in organizations, and what kind of disadvantages women and minorities faced, but we know almost nothing about how to how to reduce them,” says Alexandra Kalev, a sociologist at the University of Arizona.

Several years ago Kalev, along with Dobbin and Erin Kelly of the University of Minnesota, set out to see what works. As a measure of program success, they looked at the number of women and minorities in a company’s managerial ranks – a much more concrete metric than the surveys of employee attitudes that many other studies relied on. The researchers drew on 31 years of Equal Employment Opportunity Commission data, specifically the annual reports that companies file detailing their racial and gender makeup. The sociologists then surveyed 829 of those companies on what diversity programs they had and when they instituted them. The results were described in a 2006 study, and in another paper that Kalev and Dobbin are currently writing.

The researchers found that while diversity training was by far the most popular approach, it was also the least effective at getting companies to hire and promote women and minorities. Some training programs were more effective than others: Voluntary programs were better than mandatory ones, and those that focused on the threat of bias and harassment lawsuits were worse than those that did not. But even the better programs led only to marginal changes. And those that were mandatory or discussed lawsuits – the vast majority of the programs the researchers examined – slightly reduced the number of women and minorities in management. Required training and legalistic training both make people resentful, the authors suggest, and likely to rebel against what they’ve heard.

What worked much better than even the best training, the researchers found, were more structural measures: minority mentoring programs, or designating an executive or a task force with specific responsibility to change promotion practices.

“You can imagine, if you’re in a meeting for two hours once a year to refresh your diversity awareness, what’s the effect of that going to be compared to being a mentor to someone?” says Dobbin.

At least some diversity consultants seem willing to accept that research finding, while still defending the role of training programs in an overall diversity policy:

Diversity trainers concede that there are poorly designed programs out there. There are also, they point out, companies that implement diversity training without much concern for whether it works, which is not a recipe for success. That doesn’t mean that well designed, conscientiously applied programs don’t work.

And diversity consultants bristle at the suggestion that they believe diversity training programs are a panacea. Properly instituting a diversity training program, many of them insist, means combining it with other, more systemic changes, including measures like those that the Kalev, Dobbin, and Kelly research found were more effective.

“If you look at just the efficacy of diversity training programs, that’s not how we look at it as a practitioner,” says Rohini Anand, global chief diversity officer at the food services giant Sodexo. “To me diversity training is one small but very necessary piece of what I need to do.”

Lethal Injection ChamberThe Houston Chronicle recenly reported on the efforts of social scientists to understand whether the death penalty deters potential murderers.  According to the article, research on the issue has historically produced mixed results:

In 1967, sociologist Thorsten Sellin found no significant impact when he studied murder rates in adjacent states with differing approaches to capital punishment.

The next year, Nobel Prize economist Gary Becker developed a theory supporting the deterrent value of the death penalty, and eight years later one of his students published a study based on national statistics purporting to show that each execution saved eight lives.

The controversy led to a study commissioned by the National Academy of Sciences that found evidence of deterrence to be unconvincing.

More recent studies have reached conclusions all over the map. A national study in 2005 found “profound uncertainty” on the question and even suggested that executions might slightly increase the murder rate (possibly through a cultural “brutalization”). Another study that year suggested that each execution saves 150 lives.

The article discusses a new study, forthcoming in Criminology, by Duke University sociologists Kenneth C. Land and Hui Zheng and Sam Houston State University criminologist Raymond Teske Jr.:

After reviewing earlier studies, these authors came to the conclusion that the death penalty is used too sporadically and inconsistently around the nation for studies on national data to accurately measure its effect on crime.

They decided to focus their study by taking advantage of Texas’ gift to social science, what they call “an orgy of executions in Texas beginning in 1994,” during which time the state provided more than a third of the nation’s executions.

The authors compared this period to an era in which Texas carried out fewer executions from 1980 to 1993, attempting to isolate the effect of the increased use of the death penalty:

They found that many earlier studies had vastly overestimated the effect, but the number of murders did go down in the short-term aftermath of executions.

Based on two different statistical models, they found the effect in the months after each execution to be a reduction of between 0.5 to 2.5 homicides.

That may not sound like much, but as the authors note, “even the estimated .5 deterrent per execution yields an estimated reduction in the expected numbers of monthly homicides of 5 to 10 during the subsequent 12 months, which is substantial.”

Perhaps more interesting are the difficult issues that remain unresolved:

Here’s the mystery:

This study and previous ones show no correlation between the amount of publicity executions receive and their deterrent effect.

“We have no theory on that,” Teske said on Friday. After a few more questions, he said, “I hear your frustration. If I wasn’t working with one of the top guys in the nation, my confidence would be shaken.”

One other mystery: The study shows, as other studies have, more impact on the kinds of murders that don’t qualify for the death penalty than on those that do.

View of Philadelphia from the Museum of Art #1The Philadelphia Inquirer reports this week on a new study from the Pew Charitable Trusts regarding census preparations in a number of large US cities.

The study compared Philadelphia’s census preparation efforts with those of five larger cities – New York, Los Angeles, Chicago, Houston and Phoenix – and five cities similar to Philadelphia and its history with the census: Pittsburgh, Baltimore, Boston, Detroit and Atlanta.

Although the census is done by the federal government, ensuring the local count is complete and accurate has become a municipal self-promotion campaign.

“Census preparation really matters,” said [Thomas] Ginsberg, [the project manager of Pew’s Philadelphia Research Initiative,] and the census has become a hard sell.

Why is it a ‘hard sell?’

Immigrants, especially those here illegally, are suspicious of the count despite the promise of anonymity. They fear the count could lead to discovery and deportation of them or relatives.

Anti-government campaigns involving the Obama Administration’s proposal for universal health insurance coverage also appear to be affecting the census.

And the economic recession has dramatically reduced local tax revenue, meaning there is much less to spend on municipal census promotion efforts.

And the sociological commentary…

“Nobody is expecting a good census in 2010. I’m not optimistic,” said Joseph J. Salvo, New York City’s population division chief and a sociologist, quoted in the Pew study.

Salvo, who advises and critiques the Census Bureau, noted that “since the last census we had 9/11, privacy issues, trust of government issues. And there’s been no public declaration that we’re going to suspend immigration raids like in 2000.”

Read more.

Anna KournikovaThe Vancouver Sun ran a story yesterday about a new study by sociologist Laurel Davis-Delano of Springfield College, which suggests that “female athletes are still apologizing for smashing stereotypes while they pursue their sports.”

The Sun reports:

A newly published study that included college basketball, soccer and softball players found nearly three-quarters of them engage in “apologetic behaviours” — stereotypically feminine conduct such as cultivating a girlie appearance, apologizing for being aggressive and hanging out with men to emphasize their heterosexuality — to deflect prejudice.

“If you break a norm, you apologize. If I burp out loud, I know this offended other people, so I apologize,” says Laurel Davis-Delano, a professor of sociology at Springfield College in Massachusetts, explaining why researchers label these behaviours apologetic. “If you are offending people’s sense of gender ideals . . . people don’t necessarily realize they’re apologizing, but you are catering to other people’s sense of what’s proper.”

Most sports are still associated with masculinity in Western cultures, so female athletes are challenging gender expectations by their very participation, she says.

So what makes this different from female athletes looking pretty just because they want to?

Apologetic behaviours are different from female athletes having long hair or wearing makeup simply because they like to, Davis-Delano says, because they’re performed specifically in response to this gender tension.

While 73 per cent of the study participants said they engaged in at least one apologetic behaviour, from criticizing unfeminine athletes to being seen with a boyfriend, no one shied away from aggression or competing hard against male athletes.

On one hand, apologetic behaviours may help female athletes gain acceptance and be rewarded in their sport, Davis-Delano says. But they do little to challenge gender stereotypes, she says, and Russian tennis player Anna Kournikova is a “classic example” of the result: a female athlete of lesser talent who gets attention and endorsements for her ultrafeminine looks.

Read more.

amy's birthday cake!A recent article in USA Today, based on new data from the Pew Research Center, indicates that few people see themselves as ‘old,’ regardless of their age.  USA Today reports, “No matter what their chronological age, most people say that they aren’t yet “old” — and that they feel younger than their birthday count, according to a new nationally representative survey of almost 3,000 adults by the Pew Research Center.”

The findings:

The average age considered “old” by respondents was 68 — but there were real differences in perception driven by the respondents’ own ages:

•More than half of those under 30 say the average person becomes old before 60.

•Middle-aged respondents say it’s closer to 70.

•Those ages 65 and older say “old” is not until 75.

“What you find is the older people are, the more people push back the age that is old,” says Russell Ward, a sociologist who focuses on aging at the University at Albany, State University of New York, and who was not involved in the survey. “It’s more in your future. You’re not there yet.”

A researcher at the Pew Research Center notes the following…

“We are becoming an older society, as are most advanced societies around the world, and we are about to hit a big new wave of adults entering older age,” says Paul Taylor, who directs Pew’s Social and Demographic Trends project.

The study notes that about 39 million Americans, or 13% of the U.S. population, are 65 and older — a figure that has tripled from 4% in 1900. In two years, the oldest of the nation’s 76 million Baby Boomers will turn 65. And by 2050, according to Pew Research projections, about one in five Americans will be over 65, and about 5% will be ages 85 and older, up from 2% now.

Expectations and realities about aging in the survey also differ. Among those age 65 and older, the perceived downsides of aging (such as memory loss, illness, inability to drive or an end to sexual activity) aren’t experienced as much as younger people think they’ll be.

Also, the perceived benefits of growing older (more time with family, more leisure travel, having more time for hobbies or volunteer work) are less than either age group thought they would be. Experts say the recession has reduced the “fun” part of retirement.

Read more.

With the murder of a physician who was a regular target of anti-abortion activists this past Sunday, news outlets have returned to covering the schism in our country on the abortion issue, this time focusing on a new study linking the likelihood of having an abortion to religiosity.

MSNBC reports:

Unwed pregnant teens and 20-somethings who attend or have graduated from private religious schools are more likely to obtain abortions than their peers from public schools, according to research in the June issue of the Journal of Health and Social Behavior.

“This research suggests that young, unmarried women are confronted with a number of social, financial and health-related factors that can make it difficult for them to act according to religious values when deciding whether to keep or abort a pregnancy,” said the study’s author, sociologist Amy Adamczyk of John Jay College of Criminal Justice and the Graduate Center, City University of New York.

More from Adamczyk…

Adamczyk examined how personal religious involvement, schoolmate religious involvement and school type influenced the pregnancy decisions of a sample of 1,504 unmarried and never-divorced women age 26 and younger from 125 different schools. The women ranged in age from 14 to 26 at the time they discovered they were pregnant. Twenty-five percent of women in the sample reported having an abortion, a likely underestimate, Adamczyk said.

Results revealed no significant link between a young woman’s reported decision to have an abortion and her personal religiosity, as defined by her religious involvement, frequency of prayer and perception of religion’s importance. Adamczyk said that this may be partially explained by the evidence that personal religiosity delays the timing of first sex, thereby shortening the period of time in which religious women are sexually active outside of marriage.

Despite the absence of a link between personal religious devotionand abortion, religious affiliation did have some important influence. Adamczyk found that conservative Protestants (which includes evangelicals and fundamentalist Christians) were the least likely to report having an abortion, less likely than mainline Protestants, Catholics and women with non-Christian religious affiliations.

Regarding the impact of the religious involvement of a woman’s peers, Adamczyk found no significant influence. However, Adamczyk did find that women who attended school with conservative Protestants were more likely to decide to have an extramarital baby in their 20s than in their teenage years.

“The values of conservative Protestant classmates seem to have an abortion limiting effect on women in their 20s, but not in their teens, presumably because the educational and economic costs of motherhood are reduced as young women grow older,” Adamczyk said.

The LA Times also picked up on the story in their Health section this week. They report: 

In a study published in the June issue of the Journal of Health and Social Behavior, a sociologist at the City University of New York analyzed the abortion decisions of unmarried teenagers and young twentysomethings. Specifically, she was looking at how those decisions were affected by personal religious devotion, schoolmates’ religious devotion and the type of school (public or religious).

Come decision-making time, religiosity —  the importance attributed to religion and the involvement in it — didn’t make much difference.

Maybe that’s surprising to you, maybe not.

But of note, she writes: “Conservative Protestants appear less likely to obtain abortions than mainline Protestants, Catholics, and women of non-Christian faiths. Regardless of personal religious affiliation, having attended a school with a high proportion of conservative Protestants appears to discourage abortion as women enter their twenties. Conversely, women from private religious high schools appear
more likely to report obtaining an abortion than women from public schools.”

Read more from MSNBC.

Read more from the LA Times. 

Introductions Science Daily (a press release service) brought a new study by sociologist Gerald Mollenhorst to the attention of the Crawler today, which examines how “the context in which we meet people influences our social network. One of his conclusions: you lose about half of your close network members every seven years.”

His research offers several important findings…

Limited in your choices

Mollenhorst investigated, for example, whether the social context in which contacts are made influences the degree of similarity between partners, friends and acquaintances. It was expected that the influence of social contexts on similarity in relationships would be stronger for weak relationships than for strong ones. After all, you are less fussy about your choice of acquaintances than your choice of partner. In relationships with partners, Mollenhorst indeed found more similarity than in relationships with friends. Yet interestingly, the influence of the social context on similarity did not differ between partners, friends and acquaintances. This reveals how strongly opportunities to meet influence the social composition of personal networks.

With his research Mollenhorst has confirmed that personal networks are not formed solely on the basis of personal choices. These choices are limited by opportunities to meet. Another strong indication for this came from the fact that people often choose friends from a context in which they have previously chosen a friend. Moreover, the extent to which our friends know each other strongly depends on the context in which people meet each other.

Individualism

Many sociologists assume that our society is becoming increasingly individualistic. For example, it is held that we strictly separate work, clubs and friends. Mollenhorst established, however, that public contexts such as work or the neighbourhood and private contexts frequently overlap each other.

Furthermore, Mollenhorst’s research reveals that networks are not shrinking, whereas American research reveals such a decline. Over a period of seven years the average size of personal networks was found to be strikingly stable. However, during the course of seven years we replace many members of our network with other people. Only thirty percent of the discussion partners and practical helpers still held the same position seven years later. Only 48 percent were still part of the network. Therefore value the friends you have. As long as you have them that is.

Read more.