Survey report June 8, 2021

The State of American Friendship: Change, Challenges, and Loss

Findings from the May 2021 American Perspectives Survey

Daniel A. Cox

3 friends having coffee time on a terrace

Coming out of a once-in-a-generation global pandemic, Americans appear more attuned than ever to the importance of friendship. However, despite renewed interest in the topic of friendship in popular culture and the news media, signs suggest that the role of friends in American social life is experiencing a pronounced decline. The May 2021 American Perspectives Survey finds that Americans report having fewer close friendships than they once did, talking to their friends less often, and relying less on their friends for personal support. 

The COVID-19 pandemic is the most obvious culprit in the national friendship decline, but broader structural forces may be playing a more important role. First, Americans are marrying later than ever and are more geographically mobile than in the past—two trends that are strongly associated with increasing rates of self-reported social isolation and feelings of loneliness.[1] Second, American parents are spending twice as much time with their children compared to previous generations, crowding out other types of relationships, including friendships.[2] Finally, Americans are working longer hours and traveling more for work, which may come at the cost of maintaining and developing friendships.[3] In fact, perhaps reflecting its central place in the hierarchy of American social life, Americans are now more likely to make friends at work than any other way—including at school, in their neighborhood, at their place of worship, or even through existing friends. 

But the news is not universally negative. The survey paints a more complex and perhaps more fluid picture of American friendships. Roughly half of Americans report having lost touch with at least one friend during the pandemic. But, surprisingly, nearly as many Americans report having made a new friend over this same period. Many Americans report having activity friendships or situational friends—people they see at certain times or places—and most Americans have a best friend, even if it’s fewer than in the past.

Personal Pandemic Challenges

Most Americans report having faced significant personal challenges over the past 12 months. More than one in five (22 percent) Americans say the past 12 months have been much more difficult for them than usual. Nearly half (46 percent) say it was somewhat more difficult. About only one in three (32 percent) Americans say the past 12 months were not especially difficult for them—at least no more than usual. 

Women report having had a more challenging time than men did. More than one in four (27 percent) women say the past 12 months have been much more difficult than usual, compared to 17 percent of men. Middle-aged men (age 50–64), in particular, report having a significantly easier time than other Americans did. Only 14 percent of middle-aged men report that the past 12 months have been much more difficult than a typical year has, while 43 percent say it was not any different than usual.

The financial devastation wrought by the COVID-19 pandemic has been well-documented, but less widely reported is the emotional toll many Americans faced as a result of quarantine requirements and self-imposed social isolation. Nearly half (47 percent) of Americans report having lost touch with at least a few friends over the past 12 months. Nearly one in 10 (9 percent) Americans report having lost touch with most of their friends.

Young women appear to have been more affected than most were. Nearly six in 10 (59 percent) report having lost touch with at least a few friends, and 16 percent say they are no longer in regular contact with most of their friends.

Pandemic Friendship

Despite prolonged periods of social isolation and quarantine that characterized much of American life over the past year, nearly half (46 percent) of Americans report having made a new friend within the past 12 months. Nearly one-third (30 percent) of the public say they have made a new friend in the past one to four years. Notably, 22 percent of Americans say it has been at least five years since they last made a new friend.

There are stark and predictable generational divisions in how frequently Americans are developing new friendships. A majority (56 percent) of young adults—including roughly similar numbers of young men (54 percent) and young women (58 percent)—say they have made a new friend in the past 12 months. Seniors are considerably less likely to have made a new friend during that period, with only 41 percent reporting they had done so. Nearly one-third (31 percent) of seniors say it has been at least five years since they developed a new friendship.

Of the many ways Americans make friends and the many places friendships develop, the workplace is the most common. A majority (54 percent) of Americans with close friends say they met a close friend at their or their spouse’s workplace. Less than half (47 percent) report making a close friend at school—either one they currently attend or one they previously attended. Forty percent of Americans say they made a close friend through their existing network of friends. About one in three (35 percent) Americans have made a close friend in their neighborhood, and about one in five made a close friend at their place of worship (21 percent) or a club or organization they belong to (19 percent). Ten percent of Americans—and 16 percent of adults with children living at home—say they developed a close friend through their child’s school. Only 8 percent of Americans report having made a close friend online.

Who Do Americans Rely on for Personal Support?

Overall, more than half (53 percent) of Americans say that the first person they talk to when they have a personal problem is their spouse or partner. Sixteen percent of the public say they go to a friend first when confronting a personal issue, and 10 percent say they rely on their parents. Roughly the same amount (9 percent) say they turn to a sibling or other family member. Five percent of Americans say they first reach out to their children when dealing with a personal issue. 

Fewer Americans appear to be relying on friends for personal support than they have in the past. A survey conducted by Gallup in 1990 found that more than one-quarter (26 percent) of Americans said their friend was the first person they would turn to when they had a personal problem.[4]

Young adults and singles are unique to the extent that they rely on friends for emotional and personal support. One in four (25 percent) young people (age 18–29) say they typically talk to a friend first when they have a personal problem. Young women are slightly more likely than young men are to rely on friends for support (29 percent vs. 22 percent). 

With the average age of first marriages rising steadily and fewer Americans turning to friends for support, parents have stepped in to fill the gap. Young men are now more likely to rely on their parents for personal support than other people in their life. Thirty-six percent of young men say their parents are the first people they reach out to when facing a personal problem. Roughly one in four (24 percent) young women say their parents are their first call. 

The number of young men relying on their parents for personal support has more than doubled over the past several decades. In 1990, only 17 percent of young men and an identical number of young women reported that their parents were the first people they talked to when confronting a personal problem. Close to half (45 percent) of young men said they turned to their friends first.

One reason young men are more likely to seek out their parents for support may have to do with their living situation. More than half (52 percent) of young men are currently living with their parents, compared to 44 percent of young women. Young adults who live with their parents are twice as likely to rely on their parents for support than those with other living arrangements.

There is an even larger gender gap between unmarried men and women. Nearly four in 10 (39 percent) single unmarried women say they usually talk to a friend when facing a personal problem, compared to 30 percent of unmarried men[5]. Again, unmarried men are significantly more likely than women are to say they rely on their parents for help when dealing with a personal issue (39 percent vs. 26 percent).

Gender differences are also apparent among married Americans. Married men are significantly more likely than married women are to say the first person they talk to when they have a problem is their spouse. Eighty-five percent of married men, compared to 72 percent of married women, say they turn to their spouse when they have a personal problem.

How Many Close Friends Do Americans Have?

Many Americans do not have a large number of close friends. Close to half (49 percent) of Americans report having three or fewer. More than one-third (36 percent) of Americans report having several close friends—between four and nine. Thirteen percent of Americans say they have 10 or more close friends, which is roughly the same proportion of the public that has no close friends (12 percent).[6]

The number of close friendships Americans have appears to have declined considerably over the past several decades. In 1990, less than one-third (27 percent) said they had three or fewer close friends, while about as many (33 percent) reported having 10 or more close friends.[7] Only 3 percent said they did not have any close friends.

Many Americans are not overly satisfied about the size of their friendship group. About half of Americans (51 percent) report they are very satisfied or completely satisfied with the number of friends they have. Thirty percent say they are only somewhat satisfied, and 17 percent say they are not too or not at all satisfied with the number of friends they have.

Women are slightly more likely than men are to report being satisfied with their number of friends. A majority (54 percent) of women say they are completely or very satisfied, compared to less than half of men (48 percent). 

There are notable racial and ethnic differences in feelings of satisfaction about the number of friends Americans have. Black and Hispanic Americans express greater feelings of satisfaction than White Americans do. Close to six in 10 Black (58 percent) and Hispanic (56 percent) Americans report they are very or completely satisfied with how many friends they have. About half (49 percent) of White Americans say the same.

Not surprisingly, Americans who have more friends report higher levels of satisfaction with the number of friends they have. In fact, levels of satisfaction move linearly: As Americans accumulate additional friends, their level of satisfaction grows. Among Americans without any close friends, only 29 percent report being completely or very satisfied with their number of friends. Less than four in 10 (39 percent) Americans with only one close friend and 43 percent of those with two or three close friends report being completely or very satisfied. A majority (56 percent) of Americans with four or five friends say they are completely or very satisfied. Two-thirds (67 percent) of Americans with between six and nine friends are completely or very satisfied, and three-quarters (75 percent) of those with 10 or more close friends express this level of satisfaction.

Friendship Types

Friends come in many shapes and sizes. Some friends Americans see infrequently or maybe only in certain places. Other friends may have been close confidants since childhood. Most Americans report having several different types of friendships.

Most Americans have situational friends or place-based friendships—people they mostly see in certain places such as work, school, the gym, or the park. Nearly seven in 10 (69 percent) report having a friend they see only in certain places or at certain times. Activity friendships are also quite common. About half (51 percent) the public say they have activity friends, people with whom they participate in social activities, such as sports, hobbies, or community service. Nearly four in 10 (39 percent) Americans have online-only friendships or friends they interact with solely via the internet.

Childhood friendships are ubiquitous among the public. Two-thirds (67 percent) of Americans say they have a friend whom they have known since childhood. Yet despite the prevalence of these types of friendships, they have become less common as Americans age. More than three-quarters (76 percent) of young adults have a childhood friend, compared to 60 percent of seniors.

Childhood friendships are particularly prevalent among Black Americans. Nearly eight in 10 (78 percent) Black Americans report having a friend whom they have known since childhood. These types of friendships are significantly less common among White (66 percent) and Hispanic Americans (64 percent). 

Having a friend of a different gender is fairly common in American society, but the incidence varies. Fifty-eight percent of Americans who have a close friend report having a close friend of the opposite gender. Interestingly, men are more likely than women are to report having a close friend of a different gender (63 percent vs. 53 percent).

Among those with close friends, having an opposite-gender friend is much less common among married people, particularly among married women. Married women are far less likely than other women are to report having a male friend. Only 43 percent of married women—and 54 percent of married men—say they have a close friend who is a different gender. In contrast, nearly two-thirds (65 percent) of unmarried, single women say they have a close male friend.

Friendship Closeness and Communication

The frequency with which Americans talk to their friends and their preferred mode of communication vary widely. A majority (54 percent) of Americans say they texted a friend within the past 24 hours. Thirty percent say they talked with a friend on the phone in the past 24 hours, while fewer Americans—only 16 percent—report sending an email to a friend over this period.

There are stark generational differences in the mode of communication Americans prefer to use to get in touch with their friends. More than six in 10 (61 percent) young adults say they have texted a friend within the past 24 hours, compared to only 41 percent of seniors. Conversely, older Americans are far more likely to have sent an email to a friend. Nearly three in 10 (29 percent) seniors say they emailed a friend within the past 24 hours, compared to 10 percent of young adults. There are no significant generational differences in the frequency with which Americans talk to their friends on the phone.

Emotional Support

There are massive differences in the degree to which men and women rely on friends for emotional support and are willing to share their personal feelings. Nearly half of women (48 percent) and less than one-third of men (30 percent) say they have had a private conversation with a friend during which they shared their personal feelings in the past week. 


Men are also far less likely than women are to have received emotional support from a friend. Four in 10 (41 percent) women report having received emotional support from a friend within the past week, compared to 21 percent of men. 

Finally, compared to men, women more regularly tell their friends they love them. About half (49 percent) of women say they have told a friend they loved them within the past week. Only one-quarter (25 percent) of men say they have done this.

There are no generational differences, meaning younger men are no more likely than older men are to have shared their personal feelings with a friend. However, men who have female friends are significantly more likely to express their feelings and receive emotional support than are those without. Twenty-eight percent of men with female friends report that they received emotional support from a friend within the past week, compared to 16 percent of men who do not have female friends. Compared to men who have only male friends, men with female friends are also more likely to have shared personal feelings (38 percent vs. 25 percent) and to have told a friend they loved them (35 percent vs. 15 percent) in the past week.

Best Friends

Most Americans report having a best friend. Nearly six in 10 (59 percent) Americans say they have one person they consider their best friend. Forty percent say they do not. Notably, these types of friendships are common, and their prevalence hardly varies at all among the public. Across the lines of race and ethnicity, age, religion, and politics, best friends are a common feature of American social life. 

However, fewer Americans have a best friend today than they once did. In 1990, three-quarters (75 percent) of Americans reported having a best friend, a dramatic decline over the past three decades.

For most Americans, best friends include people beyond their immediate or extended family. About only one in five (21 percent) Americans say their best friend is a family member, while the vast majority (78 percent) say they are not. Among those who say their best friend is a family member, they are most frequently identified as their spouse or partner. 

There are few differences among the public in the identity of best friends, but White evangelical Protestants are unique to the extent that their best friends are family members. More than one in four (26 percent) White evangelical Protestants who have a best friend say this person is a family member. 

The survey included a question that asked respondents to share, in their own words, what it was that made someone their best friend. While there is not any one characteristic or experience that Americans identify, for many, longevity is a crucial element. A 25-year-old Hispanic woman described her best friend as someone whom she has known for most of her life. "She and I have been best friends since we were in the third grade. She and I have a unique friendship, where she and I can reach out to each other after any amount of time and no matter how long it’s been there’s only good vibes and love."

A White 54-year-old man similarly leads with how long he has known his best friend. “I’ve known this person for over 30 years. We have seen the best and worst in each other and have seen each other through some difficult times.” For many Americans, being friends with someone for a long time is an important, and perhaps key, ingredient.

These open-ended results are consistent with other findings. Compared to Americans who do not have a childhood friend, those who do are much more likely to have a best friend. Nearly two-thirds (65 percent) of Americans who have a childhood friend also have a best friend. In contrast, less than half (47 percent) of Americans who do not have a friend from their childhood say they have a best friend. A majority (53 percent) do not have either. 

Friendship and Politics

Few Americans report that they regularly discuss politics or government with their friends. About only one in five (21 percent) Americans say they discuss political issues at least a few times a week. About one in four (24 percent) say they talk with friends about politics a few times a month. More than half (55 percent) the public report that they talk about politics with their friends less often. Notably, Democrats and Republicans[8] are not any more likely to discuss politics with their friends than the public overall is.

For most Americans, political affiliation is probably not a prerequisite for forming a friendship, but both Democrats and Republicans are far more likely to have friends who belong to their preferred party. About eight in 10 (82 percent) Democrats and Republicans (80 percent) say they have at least some friends who share the same political identity. Importantly, Republicans have more bipartisan friendships than Democrats do. A majority (53 percent) of Republicans say they have at least some friends who are Democrats. In contrast, less than one-third (32 percent) of Democrats say they have at least some Republican friends.

Losing a Friend over Politics

Although political disagreements are common, few Americans report having stopped talking to or being friends with someone because of their views about government or politics. Only 15 percent of the public have ended a friendship over politics.

Ending friendships over political disagreements occurs more among liberal and Democratic-leaning Americans. Democrats are twice as likely as Republicans are to report having ended a friendship over a political disagreement (20 percent vs. 10 percent). Political liberals are also far more likely than conservatives are to say they are no longer friends with someone due to political differences (28 percent vs. 10 percent, respectively). No group is more likely to end a friendship over politics than liberal women are; 33 percent say they stopped being friends with someone because of their politics. 

The reasons Americans give for dropping a friendship are varied, but President Donald Trump loomed large. A White college-educated man said: “I can’t deal with crazy people who worship Donald Trump.” A Hispanic woman offers a similarly blunt explanation: “If they were a fan of DJT, I wanted nothing to do with them.” But many Trump supporters were equally willing to walk away from friends whose views of the former president did not align with their own. “I have unfriended people online and stopped talking to people who didn’t respect our great President Trump.” In all, 22 percent of Americans who have ended a friendship cited Trump as the reason.


Survey Methodology

The survey was designed and conducted by the Survey Center on American Life. Interviews were conducted among a random sample of 2,019 adults (age 18 and up) living in the United States, including all 50 states and the District of Columbia. All interviews were conducted among participants of the the Ipsos KnowledgePanel, a probability-based panel designed to be representative of the US general population, not just the online population. Interviewing was conducted between May 14 and May 23 2021. Interviews were conducted in Spanish and English.

Initially, participants are chosen scientifically by a random selection of telephone numbers and residential addresses. Persons in selected households are then invited by telephone or mail to participate in the Ipsos KnowledgePanel. For those who agree to participate but do not already have internet access, Ipsos provides at no cost a laptop and internet service provider connection. People who already have computers and internet service are permitted to participate using their own equipment. Panelists then receive unique log-in information for accessing surveys online and then are sent emails throughout each month inviting them to participate in research.

The data were weighted to adjust for gender by age, race, education, Census region by metropolitan status, and household income. The sample weighting was accomplished using an iterative proportional fitting (IFP) process that simultaneously balances the distributions of all variables.

The use of survey weights in statistical analyses ensures that the demographic characteristics of the sample closely approximate the demographic characteristics of the target population. The margin of error for the qualified survey sample is +/– 2.4 percentage points at the 95 percent level of confidence. The design effect for the survey is 1.2.


Notes

[1] Daniel A. Cox, Ryan Streeter, and David Wilde, “A Loneliness Epidemic? How Marriage, Religion, and Mobility Explain the Generation Gap in Loneliness,” American Enterprise Institute, September 26, 2019, https://www.aei.org/research-products/report/loneliness-epidemic-how-marriage-religion-and-mobility-explain-the-generation-gap-in-loneliness/.

[2] Economist, “Parents Now Spend Twice as Much Time with Their Children as 50 Years Ago,” November 27, 2017, https://www.economist.com/graphic-detail/2017/11/27/parents-now-spend-twice-as-much-time-with-their-children-as-50-years-ago.

[3] Steven John, “7 Ways American Work Habits Have Changed in the Past 10 Years,” Business Insider, March 27, 2019, https://www.businessinsider.com/american-work-habits-culture-change-2019-3.

[4] Although the wording was identical in both questions, there was slight variation in the response options. Gallup asked about only “husband or wife,” while the May American Perspectives Survey asked respondents about their “spouse or partner.” 

[5] The single unmarried married category includes respondents who have never been married, and who are not currently in a committed romantic relationship.

[6] The definition of “close friend” is subjective and open to interpretation. The results are in keeping with previous work on the social networks. The American National Social Network Survey conducted in 2020 found that 17 percent of Americans had no close social ties, defined as someone the respondent had talked to in the past six months about an important personal matter. 

[7] The survey conducted by Gallup in 1990 was administered over the telephone, while the current study was conducted completely online. Both surveys are based on random samples and designed to represent the adult population. However, differences in survey administration may affect how respondents answer questions. Telephone survey respondents talking to a live interviewer might have been less willing to report having no friends or inflate the number of close friends they had out of concerns about how their responses would be perceived by the interviewer, a phenomenon known as social desirability bias. Gallup Organization. “Gallup News Service Poll: January 1990, Wave 2,” Roper Center for Public Opinion Research, 1990, https://doi.org/10.25940/ROPER-31088676.

[8] Note: In this analysis, Democrats include people who identify as Democrat and Independents who lean towards the Democratic party. Republicans include people who identify as Republican and Independents who lean towards the Republican party.

Survey Reports

gender divide banner

Daniel A. Cox, Kelsey Eyre Hammond
September 24, 2024

The Politics of Progress and Privilege: How America’s Gender Gap Is Reshaping the 2024 Election

Americans are increasingly divided on gendered issues. A new report by the Survey Center on American Life provides context for how these divisions might impact the results of the 2024 Presidential election.

Daniel A. Cox, Sam Pressler
August 22, 2024

Disconnected: The Growing Class Divide in American Civic Life

Disconnected: Places and Spaces presents new survey findings that suggest Americans are less connected than ever before.

Daniel A. Cox, Kyle Gray, Kelsey Eyre Hammond
May 28, 2024

An Unsettled Electorate: How Uncertainty and Apathy Are Shaping the 2024 Election

A survey of more than 6,500 US adults focused on the 2024 presidential election reveals a pessimistic and unsettled American electorate fractured by education, ideology, class, and gender.

Generation Z and the Transformation of American Adolescence Cover Image

Daniel A. Cox, Kelsey Eyre Hammond, Kyle Gray
November 9, 2023

Generation Z and the Transformation of American Adolescence: How Gen Z’s Formative Experiences Shape Its Politics, Priorities, and Future

This report explores the foundational differences between American generations through their formative adolescent experiences.