You need a favor from a friend. How should you ask? You’re trying to raise funds for a favorite charity. How should you go about getting people to donate their hard-earned money? Your first job out of college is in sales. How do you get people to sign on the dotted line? These are all questions about compliance: getting people to agree to something you want. Coming from the other direction, how can you avoid being influenced by the compliance attempts of others? What techniques should you watch out for? Social psychologists have studied different strategies for eliciting compliance, and their research findings help explain how—and how effectively—these strategies work (Cialdini, 2008, 2016; Goldstein, Martin, & Cialdini, 2008).
Different requests directed at different people—and to people in different contexts—have to be tailored accordingly. The details matter. Nevertheless, there are some general principles governing compliance that you should keep in mind when deciding the best way to ask for what you want. Let’s look at three of them in some depth. First, as we saw in Chapter 6, people find inconsistency unpleasant, so appeals to consistency and past commitments can be very effective. Second, norms are powerful determinants of behavior, so any appeal can benefit from an understanding of the prevailing norms. Finally, people’s moods often drive their behavior, so appeals can succeed or fail because they come at just the right or wrong time.
All of us perform certain actions because they’re consistent with our self-image. Environmentalists take the time to recycle, even when they’re sorely tempted to toss a bottle or can into a nearby trash can, because recycling is part of what it means to be an environmentalist. Skiers rise early to tackle fresh snow, even when they really want to hit the snooze button on the alarm clock, because that’s what real skiing enthusiasts do. It’s logical, therefore, that if requests are crafted to appeal to a person’s self-image, the likelihood of compliance increases.
One way to appeal to a person’s self-image is to employ what’s known as the foot-in-the-door technique (Burger & Guadagno, 2003; Comello, Myrick, & Raphiou, 2016; Dillard, Hunter, & Burgoon, 1984; Freedman & Fraser, 1966; Souchet & Girandola, 2013). It starts with a small request to which nearly everyone complies, thereby allowing the person making the request to get a foot in the door. This person then follows up with a larger request involving the real behavior of interest. The idea is that the target person’s initial agreement to the small request will lead to a change in self-image as someone who does this sort of thing or who contributes to such causes. This person then has a reason for agreeing to the subsequent, larger request: “It’s just who I am.”
In an early test of this technique, investigators knocked on doors in a residential neighborhood and asked one group of homeowners if they would be willing to have a large billboard sign bearing the slogan “Drive Carefully” installed in their front yard for one week (Freedman & Fraser, 1966). They were shown a picture of the sign. It was large and unattractive, so not surprisingly, only 17 percent agreed to the request. Another group of residents was approached with a much smaller request—to display in a window of their home a 3-inch-square sign bearing the phrase “Be a Safe Driver.” Virtually all of them agreed with the request. Two weeks later, when this group was asked to display the billboard in their yard (receiving the very same request as those in the first group), a staggering 76 percent of the homeowners agreed to do so.
You’ve probably heard politicians oppose a piece of legislation not because there’s anything wrong with the legislation itself but because they think it might create a “slippery slope,” leading to the passage of more questionable legislation later on. Research on the foot-in-the-door technique suggests that there is merit to this concern. Human behavior, like a ball rolling down a sloping plane, is subject to momentum. Getting people started on something small often makes it easier to get them to do much bigger things down the road. We’ll see just how powerful these slippery slopes can be when we discuss the most famous research in all of social psychology later in this chapter.
Adolescent girls exposed to pregnant peers are more likely to become pregnant themselves (Akerlof, Yellen, & Katz, 1996), planning for retirement is greatly influenced by coworkers’ plans (Duflo & Saez, 2003), and student drinking is connected to students’ perceptions of how much other students drink (M. A. Lewis & Neighbors, 2004). The tendency to act as those around us do can be harnessed to achieve compliance with explicit requests or implicit suggestions. As the moral psychologist Joshua Greene (2013) put it, “The best way to get people to do something is to tell them that their neighbors are already doing it.”
Letting people know what others are doing can be used to help combat the climate crisis. Consider a norm-based approach to energy use that was instituted in California (Schultz et al., 2007). Researchers left hangtags on people’s doors indicating their average daily residential energy use (in kilowatt-hours) alongside their neighbors’ usage. The effect of this simple intervention was clear-cut and immediate: Those who consumed more energy than average altered their habits to significantly reduce their energy use.
What about the households that used less energy than average? Did telling them that their neighbors tended to be less conscientious make them more wasteful? Yes, it did. But the investigators had a simple remedy at hand that preserved the decrease in energy use among the energy wasters while avoiding increased energy use by the energy savers. The usage information given to half the households was accompanied by a small sign of approval or disapproval: a happy face for those who had used relatively little energy and a sad face for those who had used more than average. The signal of approval was enough to maintain the superior conservation efforts of those who might otherwise have slacked off after hearing that their neighbors used more energy than they did (Figure 8.6). Used wisely, information about norms can be a powerful tool to promote energy conservation: Giving consumers information about norms reduces energy consumption by the same amount as does raising the price of energy by 10–20 percent. Normative information like this can be even more effective in helping to solve problems that require collective action—such as the climate crisis—when paired with “working together” messages that reference “joining in” or “doing it together” (Howe, Carr, & Walton, 2021).
This simple technique of informing people about social norms is likely to be most effective when the information is surprising (when people have misunderstood the norm), such as when people overestimate the popularity of destructive behavior or underestimate the popularity of constructive behavior (Tankard & Paluck, 2016). Student drinking is a case in point. On campuses across the United States, students think that binge drinking is much more common than it actually is and that “teetotaling” or moderate drinking is much less common than it really is (Perkins, Haines, & Rice, 2005). These beliefs represent examples of pluralistic ignorance, or the tendency for people to act in ways that conflict with their true beliefs or preferences because they think they are not widely shared by others. But when many people do so, it makes everyone even more convinced that the erroneous group norm exists.
In one study of pluralistic ignorance on a college campus, Deborah Prentice and Dale Miller (1993) examined the discrepancy between private attitudes and public norms about alcohol use at Princeton University. Prentice and Miller asked Princeton undergraduates how comfortable they felt about campus drinking habits as well as how comfortable with drinking they thought both their friends and the average undergraduate were. If the students were suffering from pluralistic ignorance, they would indicate that they were less at ease with drinking than they supposed most students were.
FIGURE 8.7 PLURALISTIC IGNORANCE (A) University students believe drinking alcohol is more popular among their peers than it really is. Because of this belief, they censor their own reservations about drinking, thus furthering the illusion that alcohol is so popular. (B) These results show student ratings of their own and others’ comfort with campus drinking habits at Princeton University. Source: Part B adapted from Prentice & Miller, 1993.
DATA EXPLORATION
The results, shown in Figure 8.7, indicate that this is exactly what happened. Hidden discomfort with alcohol existed side by side with perceived popular support.
Prentice and Miller attributed this discrepancy to the visibility of drinking on campus:
The alcohol situation at Princeton is exacerbated by the central role of alcohol in many of the university’s institutions and traditions. For example, at the eating clubs, the center of social life on campus, alcohol is on tap 24 hours a day, 7 days a week. Princeton reunions boast the second highest level of alcohol consumption for any event in the country after the Indianapolis 500. The social norms for drinking at the university are clear: Students must be comfortable with alcohol use to partake of Princeton social life. (Prentice & Miller, 1993, p. 244)
Efforts to stem excessive alcohol consumption by providing students with accurate information about their peers’ drinking habits have proved to be quite effective (Neighbors, Larimer, & Lewis, 2004; Perkins & Craig, 2006; Schroeder & Prentice, 1998). In one study, students attending regularly scheduled club or organizational meetings electronically submitted information about their own drinking behavior and their beliefs about the drinking habits of their peers (LaBrie et al., 2008). Their aggregate responses were immediately projected for all to see, giving everyone information about actual drinking behavior on campus—and correcting widespread misunderstandings of how much and how often other students drink. In follow-up surveys conducted one and two months later, the students who received this information reported drinking significantly less than they had previously and less than students in a control group.
Similar norm-based approaches have been used to combat harassment and bullying in schools (Paluck & Shepherd, 2012; Shepherd & Paluck, 2015). In one study, half of a group of 56 middle schools in New Jersey were randomly assigned to a social norm treatment condition in which a randomly selected group of students was asked to express opposition to the kinds of conflict and harassment that were common at their school (for example, speaking out when one student taunted or viciously teased another). The other schools served as controls. Disciplinary reports declined in the treatment schools by 30 percent relative to the control schools. As you might expect, some students were more effective than others at modeling anti-harassment norms: The more popular students had a bigger effect on their peers’ beliefs about what sorts of conflicts were common or acceptable at their school (Paluck, Shepherd, & Aronow, 2016).
What can you do when the relevant norm is counter to the behavior you would like to see? What if students at a given school are just fine with bullying, or what if students at a notorious “party school” favor quite a bit of drinking? How can norms be used to bring about desired change in environments like that? One solution is to highlight that the norms are changing—that, sure, only a third of the population favors a given action or policy right now, but that’s an increase over the one in five who favored it only recently. In one study that supports this idea, male survey respondents in a static norm condition were told that over 35 percent of men in the United States consider themselves feminists. Those in the dynamic norm condition were told that “the number of men who consider themselves feminists is increasing. . . . Now over 35% of men consider themselves feminists.” Afterward, those in the latter condition were more likely to indicate that they identified as feminists and that they supported pay equity policies (Sparkman & Walton, 2019; see also Mortensen et al., 2019). People are influenced not just by the numbers—that is, by what the norm is. They are also influenced by trends—by how the norm is changing.
In preparing norm-based compliance appeals, it’s important to be aware that there are two kinds of norms. Descriptive norms are simply descriptions of what is typically done in a given context. Prescriptive norms, often called injunctive norms, are what one is supposed to do. Descriptive norms correspond to what is; prescriptive norms correspond to what ought to be. University administrators often say that students should get 8–9 hours of sleep each night (prescriptive norm), but most students sleep much less (descriptive norm).
To increase compliance, the two norms should not be placed in conflict with each other. A common mistake is to try to strengthen the pull of the prescriptive norm by stating how infrequently it is followed. “Isn’t it a shame that so few people . . .” vote in elections, eat a healthy diet, get screened for cancer—you name it. Making such an appeal seems sensible, but note that it highlights the unfortunate reality—the descriptive norm—as much as or more than the prescriptive norm you want to promote (Sieverding, Decker, & Zimmermann, 2010; Stok et al., 2014). By saying what a shame it is that so few people vote, you’re pointing out that few people vote. Given the power of descriptive norms, such information can actually make people less likely, not more likely, to vote. Indeed, those involved in get-out-the-vote campaigns now realize, thanks to research by social psychologists, that it’s more effective to emphasize how many people vote, not how few (Gerber & Rogers, 2009).
Researchers conducted an ingenious investigation of this approach in the Petrified Forest National Park in Arizona, where visitors sometimes take samples of petrified wood home with them as souvenirs (Cialdini et al., 2006). If everyone took samples, of course, there would soon be no Petrified Forest to visit. To figure out the most effective way to deal with this problem, the investigators rotated different warning signs at various locations in the park. One sign included the usual emphasis on the severity of the problem, stating, “Many past visitors have removed petrified wood from the park, changing the state of the Petrified Forest,” accompanied by photographs of visitors taking wood. An alternative sign was framed positively: “The vast majority of past visitors have left the petrified wood in the park, preserving the natural state of the Petrified Forest,” with accompanying pictures of visitors admiring and photographing a piece of petrified wood. The investigators placed specially marked pieces of wood along trails near these signs and monitored how many of them were stolen over the course of the experiment. In a remarkable demonstration of the importance of aligning prescriptive and descriptive norms, the theft rate was over four times lower when the signs emphasized how few people take wood from the park.
When someone does something for us, we usually feel compelled to do something in return. Indeed, all societies that have ever been studied possess a powerful norm of reciprocity, according to which people are expected to provide benefits for those who have provided benefits for them (Fiske, 1991; Gouldner, 1960). This norm also exists in many bird and mammal species. When one monkey removes parasites from another’s back, the latter typically returns the favor, thus helping cement the social bond between them.
This powerful norm can be a highly effective tool to elicit compliance. If you do something nice for someone, that person will be more likely agree to a reasonable request you subsequently make. By failing to respond favorably, the person would violate a powerful social norm and run the risk of being viewed negatively (Cotterell, Eisenberger, & Speicher, 1992). Indeed, the English language is rich in derogatory terms for those who don’t uphold their end of the bargain: sponge, freeloader, moocher, bum, deadbeat, ingrate, parasite, bloodsucker, leech. Who wants to be seen in any of these ways? This may be why restaurant customers often leave larger tips when the server gives them a piece of candy with the check (Strohmetz et al., 2002).
The influence of the norm of reciprocity in getting others to comply was demonstrated in a simple experiment in which two people were asked to rate a number of paintings, supposedly as part of a study of aesthetics (D. T. Regan, 1971; see also Burger et al., 2009; Whatley et al., 1999). One was a real participant, and the other was a confederate of the experimenter. In one condition, the confederate returned from a break with two sodas and offered one to the participant: “I asked (the experimenter) if I could get myself a Coke, and he said it was OK, so I bought one for you, too.” In another condition, the confederate returned empty-handed. Later, the confederate asked the participant for a favor. He explained that he was selling raffle tickets; the prize was a new car, and he’d win $50 if he sold the most tickets. He then proceeded to ask if the participant was willing to buy any tickets for 25 cents apiece: “Any would help, the more the better.” (To make sure all participants had the means to purchase some tickets, they had already been paid—in quarters—for participating in the study.)
In a testament to the power of the norm of reciprocity, participants who earlier had been given a soda by the confederate bought twice as many raffle tickets as those who had not (or those who had been given a soda by the experimenter, to control for the possibility that simply receiving a soda, and perhaps being in a good mood as a result, is what had increased compliance).
Thus, doing a favor for someone creates an uninvited debt that the recipient is obligated to repay. Businesses and other organizations often try to take advantage of this pressure by preceding a request with a small gift. Insurance agents give out calendars or return-address labels. Marketers who want people to complete a survey send it out along with a dollar. Cult members offer a pamphlet, book, or flower before giving their pitch. Sometimes our hearts sink when we see these gifts coming, and we often go to great lengths to avoid them because we recognize the obligations they bring.
Robert Cialdini, social psychology’s most innovative contributor to the literature on compliance, has explored a novel application of the norm of reciprocity. The inspiration for his research on the subject is best introduced in his own words:
I was walking down the street when I was approached by an eleven- or twelve-year-old boy. He introduced himself and said that he was selling tickets to the annual Boy Scouts circus to be held on the upcoming Saturday night. He asked if I wished to buy any at five dollars apiece. Since one of the last places I wanted to spend Saturday evening was with the Boy Scouts, I declined. “Well,” he said, “if you don’t want to buy any tickets, how about buying some of our big chocolate bars? They’re only a dollar each.” I bought a couple and, right away, realized that something noteworthy had happened. I knew that to be the case because: (a) I do not like chocolate bars; (b) I do like dollars; (c) I was standing there with two of his chocolate bars; and (d) he was walking away with two of my dollars. (Cialdini, 1984, p. 47 )
Cialdini’s experience with the Boy Scout led him to articulate a general compliance technique whereby people feel compelled to respond to a concession by making a concession themselves (Cialdini et al., 1975; Feeley, Anker, & Aloe, 2012; Genschow et al., 2021; O’Keefe & Hale, 1998, 2001; Reeves et al., 1991). First, you ask someone for a very large favor that will certainly be refused, and then you follow up with a request for a more modest favor that you are really interested in receiving. The idea is that the drop in the size of your request will be seen as a concession; the person being asked will feel compelled to match your concession in order to honor the norm of reciprocity. The most available concession the person can make is to comply with your second request.
Another way of looking at this reciprocal concessions technique is that the first favor is so large and unreasonable that the target inevitably refuses, slamming the door in the face of that request but keeping it open just a crack for the subsequent, smaller request to get through. Accordingly, it’s also known as the door-in-the-face technique.
Cialdini demonstrated the power of this technique in a field study in which members of his research team posed as representatives of the “County Youth Counseling Program” and approached students around campus. They asked individual students if they would be willing to chaperone a group of juvenile delinquents on a trip to the zoo. Not surprisingly, the overwhelming majority, 83 percent, refused. But the response rate was much different for a second group of students who had first encountered a much larger request. They were first asked whether they would be willing to counsel juvenile delinquents for 2 hours a week for the next two years! Not surprisingly, all of them refused, at which point they were asked about chaperoning the trip to the zoo. Fifty percent of these students agreed to chaperone—triple the rate of the other group (Cialdini et al., 1975). A series of carefully crafted follow-up studies revealed that the pressure participants felt to comply with what was perceived as a concession (chaperone one trip to the zoo rather than provide counsel for the next two years) was responsible for their dramatic increase in compliance. Accordingly, this technique doesn’t work when the two requests are made by different individuals. In that case, the second, smaller request is seen not as a concession but as an entirely separate request by a different person, so the person being asked doesn’t feel the same obligation.
Suppose you want to ask your dad for a new computer, a new amplifier for your guitar, or simply to borrow the family car for a road trip. When would you ask? When he’s just come home from work in a foul mood, cursing his boss and his suffocating job? Or after he’s just landed a promotion and a big raise? It doesn’t take an advanced degree in psychology to know that it’s better to request a favor when a person is in a good mood (Andrade & Ho, 2007). A positive mood makes people feel expansive and charitable, so they’re more likely to agree to reasonable requests. Even little children know to wait until mom or dad is in a good mood before asking for a favor of some kind.
The wisdom of this approach has been verified in countless experiments. In one study, participants received a telephone call from someone who claimed to have spent her last dime on this very (“misdialed”) call. (This study was run before cell phones, when people often went to phone booths and paid a dime to make a local call. You may have seen someone do this in the movies!) The caller asked if the participant would dial a specified number and relay a message (Isen, Clark, & Schwartz, 1976). In one condition, shortly before receiving the call, participants were given a free sample of stationery to put them in a positive mood. In another condition, participants did not receive a free sample before the call. When the request was made of those who had not received the free sample, only 10 percent complied. But the compliance rate shot up dramatically among participants who received the request a few minutes after receiving the gift. The compliance rate then declined gradually as the delay between receiving the gift and hearing the request increased (Figure 8.8).
A positive mood tends to increase compliance for two main reasons. First, our mood colors how we interpret events. We’re more likely to view requests for favors as less intrusive and less threatening when we’re in a good mood, and we’re more inclined to give others the benefit of the doubt. For instance, when you’re in a good mood, you’re more likely to consider someone who asks to borrow your class notes to be a victim of circumstance who could get back on track with a little help rather than an irresponsible or lazy person who doesn’t deserve to be bailed out (Carlson, Charlin, & Miller, 1988; Forgas, 1998a, 1998b; Forgas & Bower, 1987).
The second reason a positive mood tends to increase compliance involves what’s known as mood maintenance. Pardon the tautology, but it feels good to feel good, and we typically want that feeling to last as long as possible (Clark & Isen, 1982; Wegener & Petty, 1994). One way to sustain a good mood is to do something for another person (Dunn, Aknin, & Norton, 2008). Stated differently, one way to wreck a good mood is to turn down a request for a favor and invite all sorts of self-recrimination: “What kind of heartless person am I?”
Several studies have shown that wanting to maintain a good mood is an important component of a positive mood’s effect on compliance. In one experiment, some participants were first given cookies, which put them in a good mood, while other participants weren’t given cookies. All of them were then asked (by someone other than the person who provided the cookies) if they’d be willing to assist with an experiment by serving as a confederate. Half the participants were told that the job of confederate would involve helping the “true” participant in the experiment, and the other half were told that it would involve hindering the participant. Having received a cookie (and being in a good mood) increased the compliance rate when the task involved helping the participant—but not when it involved hindering the participant. Helping another person promotes feeling good; hurting someone doesn’t. Thus, while being in a good mood increases compliance, it does not do so when the act of compliance would undermine that good mood.
If a good mood increases compliance, does a bad mood decrease it? It surely can (your dad is less likely to lend you the family car when he’s mad at his boss), but even the slightest introspection reveals that certain types of bad moods are actually likely to increase compliance, not decrease it. Some people know this and use it to their advantage. Suppose, for example, that you catch the person you’re dating flirting with a classmate of yours, and you point out the offense. Would that be a good time to ask for something? You bet it would! When people feel guilty, they’re often motivated to do whatever they can to get rid of that awful feeling. And as we have seen, doing something for someone else makes us feel good and elevates our mood. So at least one type of bad mood, centered around guilt, should increase compliance.
Social psychologists have demonstrated a strong positive association between guilt and compliance in many experiments. Participants have been made to feel guilty by being induced to lie, tricked into thinking they’ve broken a camera, or maneuvered into knocking over stacks of carefully arranged index cards—all of which led to increased compliance (Carlsmith & Gross, 1969; Darlington & Macker, 1966; O’Keefe & Figgé, 1997; D. T. Regan, Williams, & Sparling, 1972).
Other types of bad moods, not just those produced by guilt, can also increase compliance. In one study, watching an adorable lab rat get “accidentally” jolted with an intense shock led participants to donate more money to charity than did those who hadn’t seen the unfortunate event (J. W. Regan, 1971). And, in general, it seems that bad moods sometimes increase compliance in part because we simply don’t want to feel bad. Helping others makes us feel better, so we jump at the chance to brighten our mood. This is the negative state relief hypothesis in action. According to this hypothesis, taking an action to benefit someone else, especially when it’s for a good cause, is one way to make ourselves feel better (Cialdini, Darby, & Vincent, 1973; Cialdini & Fultz, 1990; Cialdini et al., 1987). We often help others, in other words, to help ourselves.
A final word about the impact of moods, good and bad, on compliance: Investigators in Israel found that if parole judges had just finished a meal before hearing a prisoner’s plea for release from prison, there was a two-thirds chance they would vote for parole (Danziger, Levav, & Avnaim-Pesso, 2011). Cases that came up just before lunch, however, when the judges were hungry and presumably crankier, had precisely a zero chance for parole. A full stomach makes a difference, so hit your dad up for the car keys after dinner, not before.
LOOKING BACKA variety of techniques can be used to increase the chances that a person will agree to a request. The foot-in-the-door technique, whereby people who comply with a small initial request are more likely to agree to a larger request later on, takes advantage of people’s desire for consistency and their commitment to a personal identity. Norm-based approaches capitalize on people’s tendencies to look to others for guidance. People are responsive to both descriptive and prescriptive norms, but it is important that norm-based appeals do not pit the two against each other. An especially powerful norm is the norm of reciprocity, which compels people to benefit those who have benefited them. In the reciprocal concessions (door-in-the-face) technique, people who have refused a large request are then induced to agree to a smaller request. People who are in a positive mood are more likely to comply with a request in order to maintain their good mood. In contrast, according to the negative state relief hypothesis, people who feel guilty or sad are also likely to comply with a request in order to feel better.