Civic Intent

A Deeper Dive on Civic Engagement Using the U.S. GivingPulse Survey

A look at trends in giving behavior and perspectives in the U.S.

Special Edition

Executive Summary

One of GivingTuesday’s core beliefs is that generosity plays a vital role in depolarization efforts by creating spaces where people can come together in community-oriented generous actions. In our work, we observe anecdotally that when individuals work together on community projects, such as organizing neighborhood cleanups, running food drives, or maintaining public spaces, they build relationships that transcend political and ideological divides. 

In this report, we contribute new data-driven insights on potential relationships between generosity-related action and political polarization, drawing on data from our GivingPulse dataset in the U.S. In parallel to our quarterly reports on generosity trends in the U.S., we took a closer look at the set of beliefs, attitudes, and behaviors we track to explore the intersection of identity and prosocial behavior in society from an academic researcher’s perspective. This report outlines our latest findings on the underlying attitudes and motivations that drive community-oriented generous action. We also report on how these measures relate to other measures of political polarization and disagreement. In this report:

We transformed our generosity data into a measure of what academic researchers call “prosociality,” which captures actions and behaviors that are intended to benefit others. In this report, we propose a new term, “civic intent,” which captures nuances relevant to discussions around polarization. This report will show some of the advantages to looking at generosity through this lens to better understand how intentions and actions interact in powerful ways.

In our conception, civic intent in the U.S. appears to depend on a mixture of factors, including an intent to help those most in need, regularly engaging in community-minded activities, and are more likely to have feelings of trust in people around them.

We find that people in the U.S. who are surrounded by prompts and reminders to participate in generosity-related activities are more likely to participate in them. They also have higher civic intent. Additionally, we find that civic intent is relatively evenly distributed across different demographic groups and political identities. Initial data indicates a willingness to give across political and ideological boundaries, in service of those who need the most help. We also find evidence that people who donate to political causes or campaigns are also more likely to give to non-political causes than the general population during the U.S. election season.

One example of a coordinated activity designed to promote generosity is GivingTuesday. In our survey, we see that people in the U.S. who are aware of—and inspired by—GivingTuesday tend to be those with the highest measured civic intent.  While survey data alone cannot distinguish between causes and effects, we hypothesize that events that raise awareness of needs and facilitate generosity within communities tend to increase a person's willingness to help in the future.

We recognize that what we are exploring in this report is from one data source, albeit a large and robust one. We welcome collaborators interested in this subject to validate, question, and help us strengthen our initial findings. We also hope that researchers use these insights to generate new research questions and opportunities to inform the field.

About the GivingPulse Survey

GivingPulse is a weekly cross-sectional survey of U.S. residents designed to investigate a broad range of giving behaviors and sentiments managed by the GivingTuesday Data Commons since 2022, supported by the Fidelity Charitable® Catalyst Fund. Over the course of each year, over 5000 people take the survey, and their responses provide insights into generosity trends, demographics, attitudes, and personal identity. We apply the broadest possible definition of generous acts in our tracking, including financial and material donations, volunteering, advocacy, and mutual aid offered to individuals or informally organized groups in addition to formally registered organizations. Because we ask people what they did in the last 7 days, our sample tends to reflect a different nearly real-time perspective of responding in the moment than other retrospective surveys of what people did in the past year. The weekly sampling approach allows this dataset to reveal how these factors are affected by crises and world events.

Introduction

GivingTuesday has tracked generosity in all forms for the past several years through its U.S. GivingPulse survey. We use an all-inclusive definition of generosity – covering monetary support, volunteering, item donation, and advocacy. This allows us to study giving patterns in the U.S. that benefit formal organizations, unstructured mutual aid groups, and individuals. In this paper, we explored repurposing our generosity data to provide signals about a person's broader actions and intentions towards helping others and improving society at large. Researchers call this prosocial behavior, and define it in various ways¹. We offer this new interpretation to prompt conversation about how to promote prosocial behavior, through a deeper understanding of what motivates a person to help others.

Our definition of prosocial behavior differs from others used in research on the topic. Where other researchers ask about prosocial tendencies in general, we ask people what they actually did recently to help others, in conjunction with other questions that gauge people’s prosocial attitudes in a more abstract sense. Because of these differences, we decided to give our prosociality measure a different name to capture the essence of what we think we are measuring: civic intent. We use the word “civic” because we ask about recent, real actions people took to help their community or the broader world. We use the word “intent” because the intentions and motivations we capture appear to have substantive relevance when comparing against self-reported behaviors. We believe this approach shows that a combination of trying to do good, having prosocial intentions, and having an overarching trust of people in general (e.g. believing most people are good) provides a unique view of how these attitudes and behaviors interact. 

In this report, we introduce several lines of inquiry to see how they affect civic intent. We hypothesized that one factor in building community and cohesion was a willingness to cross social boundaries to help others. We examined whether particular worldviews or political identity affected how much people helped others. Current-event awareness was also examined as a factor in prosocial behavior.

Others² have noted that lack of belonging is one of the most important factors in declining trust. Through data we collected in 2024, we found that most people in the U.S. had little sense of belonging in their community. Most don't feel accepted or "fit in" with others in the places they live, and this non-belonging crisis is leading to less involvement in community events and activities. In our survey, these more isolated, excluded people say they feel less generous. With these relationships in mind, we have also started to measure levels of belonging and will explore these relationships in a future report. While survey data of this kind cannot demonstrate cause and effect, all of these lines of inquiry allow us to highlight areas of future study that will likely yield insights. 

Finally, because this is a novel way of looking at prosocial behavior and its correlates, we provide some preliminary evidence that this scale is at least internally valid³. We also compare it against other prosocial-behavior questions in our survey that were not used to construct the civic intent scale and find that these relationships also hold. People exhibiting more civic intent tend to behave in other prosocial ways.

To recap, the following guiding questions were used in this report:

  1. Could our GivingPulse survey be reframed and repurposed to track prosocial behavior and the influences that affect it?

  2. What are new ways we could define and measure prosocial behavior to capture nuances around willingness to give across various ideological and polarizing boundaries? 

  3. What examples of coordinated activity promote (or are at least correlated with) the kinds of prosocial actions we aim to promote?

¹ See https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.693174/full and https://pmc.ncbi.nlm.nih.gov/articles/PMC9956867/
² Pew, and the Belonging Barometer
³ Valid in this sense means the items on our questionnaire adequately cover the quantity we aimed to measure, and the overall measure is not heavily influenced by other factors. External validity (a future goal of our validation) refers to how well this scale works in a variety of other contexts.


How We Measure Prosocial Behavior

In most cases, researchers measure prosocial behavior by asking a group of people questions that capture how often a person thinks they support others, in what situations this happens, how it changes as people age, or to identify effective drivers of prosocial behavior. Conveniently, many of the topics and questions others have used align with questions we ask in our survey, and so we decided to include additional questions covering trust, belonging, and depolarizing attitudes. Taken together, these questions allow us to explore the extent that doing good and having good intentions depend on each other. It also allows us to examine how factors such as trust, belonging, and awareness relate to how often a person says they help others. Understanding the motivations behind the generosity behaviors we track will lead to new research questions and help us and others build more effective programs for social change and movement building.

Our survey data covers specific helping behavior such as volunteering, giving money, items, or advocating for the needs of others. Prosocial behavior researchers tend to expand the actions they survey to include sharing, consoling, empathizing, and helping others in non-monetary, non-transactional ways. In the following table we show how our questions align with the most widely-used instruments for measuring prosociality⁴:

Both our survey and other prosociality measures capture altruistic intent, willingness to help those in need, and responding during a crisis. Although our survey does not ask people to self-report their empathy, we track awareness of events and whether a person responded with generosity — a behavior closely related to empathy. Others distinguish between public displays and anonymous aid. We do not do this in our survey, though we do ask whether people feel pressured to give. We used most of these overlapping questions to define our prosociality measure, civic intent, but we added signals from our survey that we thought captured other personal qualities that leads one to intentionally work for the benefit of one's community and society at large.

⁴ Examples are taken from the Prosociality Scale and Prosocial Tendencies Measure (PTM-Revised). See https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9956867/ and https://scales.arabpsychology.com/s/prosocial-tendencies-measure-revised-ptm-r/ and https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8339254/


Civic Intent

Our measure of what we call “civic intent” consolidates the prosocial generosity behaviors we track with depolarizing beliefs and actions, trust in people and/or nonprofits, and one's good intentions to participate in and strengthen one's community. By combining these features in a single index, we can compare a single metric against all other beliefs, attitudes, worldviews, awareness, and feelings of belonging in our survey. 

Civic intent score components⁵

Each item is weighted equally and adds one point to the overall score, or a 0-1 scaled value as noted. Scores are then normalized to a 0-100 point scale for ease of comparison.

Attitudes

Attitude: everyone has a responsibility to give and help those in need (0-1)

Attitude: it is important to me personally to help those in need (0-1)

Attitude: giving is somewhat or very important to me (0-1)

Attitude: intend to give as much or more in the future (0-1)

Trust

Trusting of people in general (0-1)

Trusting of nonprofits (0-1)

Depolarizing actions

I help others, even people whose beliefs, politics, or lifestyle I don’t agree with (0-1)

I strive to help those most in need, even if that means helping those from my community less (0-1)

I am helping to make my community a better, more civil place (0-1)

I have performed a random act of kindness for someone before (0-1)

Generosity behaviors

Gave in any form (1 point, or -1 point if did not give in any form)

Giving in multiple forms to multiple recipient types (0-1)

Recently started an effort to help others or an organization in one's community (0-1)

Choose any form of giving as one's most significant form of giving in the past year – and not "none of these" (0-1)

Recently gave in some form (0-1 based on recency)

Recently helped a local community person or organization (0-1)

The way we define and quantify prosocial behavior (apart from beliefs and attitudes) represents a diversion from other measurement rubrics⁶. We ask about recent action, factoring in the recency of that action, and also add a scaling measure to represent how many different types of prosocial actions a person took in the past week. Someone who did nothing for any person or group recently will necessarily have a low civic intent score, as will a person who could not find meaning in any of these types of actions they might have performed in the past year.

Our goal is to measure how prosocial behavior relates to feelings of community belonging, political identity, worldview, and attitudes about nonprofits. These serve as dependent variables and help us to assess how well the index captures the real-world phenomena of intentional prosociality. 

The histogram in Figure 2.1 shows that the majority of those surveyed engage in multiple forms of prosocial behavior. For our Q3 2024 sample, less than a quarter (23%) fell in the 0-50 score range (low civic intent). More than half (51%) scored between 51 and 75 (moderately high), and the remaining 26% scored in the highest civic intent range (76-100). This smooth distribution of civil intent scores gives us greater confidence in the representativity of people’s civic intent. We further illustrate how various behaviors and attitudes shape this score in Appendix Figure A.6.

Many of our prosocial measures ask about recent actions. In creating this measure, we explored what happens to the overall distribution of scores if a person took no generous action recently. Non-recent givers have essentially the same score distribution as shown above, except that all scores are shifted lower (the peak moves from 78 to 58). This implies that the spread of people with good intentions, trust, and depolarizing beliefs is somewhat similar between those who gave or helped recently and those who did not. 54% of non-givers fall in the bottom half (scoring 0-50), versus 23% in the total sample. Among those who claim that giving is just "not important to me," we see a more dramatic downward shift in civic intent. 82% of those people are in the bottom half of civic intent scores. 

Relative contribution of each component question to civic intent

Figure 2.2 summarizes the increase in civic intent for those that affirmed each of the questions used to calculate civic intent. Nearly everyone (90%) affirmed the statement, "I have performed a random act of kindness for someone before" (+27 points), and 81% affirmed that "it is important that I personally help others" (+27 points). Note that we ask people how important it is to them to help others in two slightly different ways on the survey, which is why it appears twice in the civic intent score and in Figure 2.2.

Actions, trust, and good intentions all appear to contribute to civic intent, as the magnitude of the shift from affirming each of these types of questions contributes a similar amount to overall civic intent. One notable exception – the weakest contributing question – was about one's intention to give the same amount or more over the next year. This may be because it is a relatively easy statement to agree with, making it a less predictive question overall. It is the only "general tendencies" question we included, and demonstrates a frequent observation that survey questions that ask about a person's future intentions tend to be less predictive than those that ask people what they have actually done recently. 

What core questions within civic intent matter most?

If researchers wanted to capture most of what matters from our 16 questions, we find six questions⁷ that may provide a good approximation of civic intent (r=0.93 correlation with the overall civic intent scoring rubric). The questions that are most relevant are as follows:

Attitude: It is moderately to very important for me to help those in need

Depolarization: I am helping to make my community a better, more civil place

Depolarization: I strive to help those most in need, even if that means helping those from my community less

Giving: Yes, I performed some act of generosity in the last 7 days

Giving: Yes, I recently gave in some form, weighted for recency

Trust: In general, most people can be trusted

What environments promote civic intent?

Our GivingPulse survey tracks the context and structure surrounding moments of generosity, and we find that people who have more opportunities or prompting to give tend to have higher civic intent. When giving money is made easier via workplace giving, automatic payroll deduction, or recurring credit card payments, scores increase by 5-6 points. When an employer hosts a giving event and people participate, their civic intent increases by 12 points. When people feel it is easy to support nonprofits and are reminded and solicited to do so, their civic intent scores increase by 18 and 25 points, respectively. In contrast, those who said donating money puts too much financial strain on them had nearly the same civic intent as those who felt no financial strain. It is important to note that none of these questions were used to calculate civic intent.


Civic Intent and GivingTuesday awareness and inspiration

In this report, we use GivingTuesday as another example of an enabler of generous action and prosocial engagement. Each week we ask people if they are aware of the GivingTuesday annual event. If they are, we ask whether it inspired them to give. 34% of the public was aware of GivingTuesday in Q3 of 2024, and those who were aware had higher civic intent. The more a person was inspired by GivingTuesday to be generous (a little, a lot, or not at all), the higher their civic intent scores⁸. The group most inspired by GivingTuesday had a civic intent score of 79, 16 points higher than the average of those who were not aware or inspired. Appendix Figure A.5 further explores the correlation between civic intent and GivingTuesday awareness and/or inspiration.

While survey data cannot ascribe cause or effect, we note that this effect is large, similar to joining a workplace or employer-led giving effort, and something that we intend to study in more detail in the future. We recognize that it could be the case that people who already have high civic intent tend to be more aware of and more inspired by GivingTuesday, or perhaps those who are aware of and inspired by GivingTuesday act in ways that increase their civic engagement, broadly speaking. 

Taken together, our measure of prosociality tends to give plausible results when compared with other questions that relate to prosocial behavior, but does not tend to shift when people are grouped on unrelated variables, including most demographic differences, giving contexts, worldview, or political leanings. One insight here is that people who perform larger or more frequent prosocial behaviors tend to affirm more of the statements we used to define civic intent (Figure A.2). In contrast, some key demographic features that we and others⁹ have reported to be associated with greater generosity in past research have a somewhat weaker relationship with civic intent than did the attitudes and behaviors we tracked – namely wealth, religiosity, and being part of a younger generation. We will delve into the larger thematic components of civic intent (trust and depolarization) next, before looking at how civic intent relates to worldview political identity. We will cover the relationship between civic intent and belonging in our next quarterly report.


⁵ Our Civil Intent scale ranges from 0-16 but is then normalized. It can carry information to 3 decimal places (with over 500 distinct values observed between 0 and 1 in our dataset). Everywhere in this report, we display it on a 0-100 point scale for ease of comparison.
⁶ Both the Prosociality Scale (https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2021.693174/full) and Prosocial Tendencies Measure (https://pmc.ncbi.nlm.nih.gov/articles/PMC9956867/) ask people to self-report how likely or how often they think they take certain prosocial actions, but don't ask people to report about any specific recent action they took.
⁷ Scoring: Four of these questions combine scaled scores in a 0-1 range (where strongly agree = 1; agree=0.75, disagree = 0.25, strongly disagree = 0), and the other two ("important to help those in need" and "acted generously in the last 7 days") are weighted 1 for yes and 0 for no.
⁸ About half of those aware (18% of the overall sample) were either somewhat or very inspired by GivingTuesday.
⁹ Trends found in https://www.philanthropyroundtable.org/almanac/who-gives-most-to-charity/ and https://www.givingtuesday.org/wp-content/uploads/2023/09/GT_FNA_2023_Final.pdf


Depolarizing attitudes

Over the last few years, we have heard others¹⁰ in the social sector express concerns that an increasingly polarized populace would be less willing to support causes and organizations doing work across political divides, and would restrict their support to particular identity-reinforcing issues or causes. This led us to ask people these questions in a 7-country survey (Brazil, Canada, India, Kenya, Mexico, Great Britain, and the USA), as part of our 2024 Lookback report on the sector. We found that in all represented countries, the largest groups of people are those who are willing to help those they disagree with and those in need outside their community.

In Q3 2024, we added these depolarization questions to the U.S. GivingPulse survey as part of a deeper and more detailed look at the intersection of prosocial behavior and identity. These questions gauge a person's willingness to support others outside of their "kin community” (that is people who are connected outside of ties of family, ancestry, or community) crossing comfort boundaries to strengthen and unify communities. As a starting point, 9 out of 10 people affirmed that they had performed a random act of kindness for someone at some point. For the most challenging depolarization question, only 54% of people affirmed that they were striving to help those most in need, even if that meant helping those in their own community less.

The highly trusting are most committed to depolarizing behaviors

Robust trust in others is very strongly correlated with support for depolarizing actions. Figure 3.2 examines the depolarizing attitudes of only highly-trusting respondents from the overall sample. The size of the red, orange, yellow, and green bars represents the percentage of highly trusting people who agree or disagree with the corresponding statements shown, and the numbers on the boxes represent the percentage point change seen for highly trusting people with that answer compared to the overall sample shown in Figure 3.1. Highly trusting individuals are 21% to 30% more likely to strongly agree with each of these statements. Just about every high-trust individual has performed a random act of kindness for someone at some point, and 75% to 85% believe they are working to help others (even those they do not agree with), making their community better, and helping those in need even if that means helping their own community less.

Likewise, we see modest increases in these depolarizing beliefs and behaviors among very religious people, people with children, people who expected to give more in the coming year, and those who said personally helping those in need was "very important" to them. 

Other attitudes associated with depolarizing views

We averaged the four depolarization questions into a normalized score (0-100) so that we could compare a person's willingness to help across polarizing boundaries against all other attitudes tracked in the survey. For each of the statements shown in Figure 3.4, the bars represent the magnitude of increase we observed for people who affirmed each statement, compared to those who rejected it. For each of the top four statements, a much larger proportion of people both agreed with these and agreed with the depolarizing statements - about 35% to 40% more. The four depolarizing viewpoint questions and these top three statements about giving and trusting nonprofits constitute seven of the questions we used to define civic intent.

We have heard that the perception among many nonprofit leaders is that negative news about one nonprofit harms the reputation of nonprofits in general. Many survey respondents will say that this is true if asked directly, but the evidence here – that both having recently heard something positive or having heard something negative about a nonprofit increase depolarization scores – implies that the relationship is not so simple. Similar insights are emerging in recent research on polarization¹¹. In this context, it might be more important that people are talking about nonprofits and the causes they serve as this tends to promote more depolarizing views.

¹⁰ See reports from the Urban Institute (https://www.urban.org/urban-wire/political-polarization-affecting-nonprofit-sector) and GivingUSA (https://www.vox.com/future-perfect/359526/charitable-giving-generosity-crisis-report-americans-young)
¹¹ One explanation provided by others is that divisive political advertising that targets a cause aligned with one particular ideology encourages non-ideological giving. See https://www.tandfonline.com/doi/pdf/10.1080/10495142.2024.2360192


Worldview and Political Giving

We measure worldview using a subset of political questions developed and tested by Pew Research. These eight questions capture the greatest differences in worldview in the U.S. among the 30+ questions they tested. 

Depending on combinations of answers to these questions, we can associate respondents with one of nine political groups that broadly span left, middle, and right in U.S. politics. This gives us better resolution in many ways. They capture a variety of positions that do not fall squarely into either of the major political parties (e.g. libertarians, progressives, fiscal vs. social conservatives); and they can more easily allow us to look at how people with opposing political stances align in other ways.

Figure 4.1
The Pew Political Typology¹²: Percent of the public of each type in 2021 (Source:
Pew Research)

In the past, we looked at how generosity compares across the political spectrum using a five point conservative-to-liberal question and observed a less than a 2% difference in the rate at which people gave money to support organizations¹³. The Pew political typology provides more nuance and shows slightly more variance in generosity according to worldview, but its prime utility is in ascribing beliefs and worldview perspectives to the behaviors that we observe. 

The political middle is as engaged as everyone else in 2024

We wanted to explore the extent to which we can understand if political giving is crowding out giving to other causes and issues. Incidence of political giving in the GivingPulse dataset is low (7.1%, n=93)¹⁴, but of the people who did report giving to a political candidate, we find some evidence that they are also giving to other causes, organizations, or individuals. 57% of political donors supported two or more non-profit organizations, compared to 28% of the total sample. 88% of all political donors also gave in some form, including support for formal non-profits, informally organized groups, or individuals.

Pew research found in 2014 that those in the political middle are less engaged politically. They vote less reliably, donate to political parties and candidates less often, and advocate less. We wondered: do our measures of political money and advocacy (for causes, not politicians) follow this same trend? Or, does our data show that the political middle are still engaged in society at the same rate, but are focusing their efforts outside of direct political giving? This may be an issue for future research to explore. 

Unlike Pew's 2014 survey, we found that in Q3 of 2024 the three centrist Pew groups (Stressed Sideliners, Outsider Left, and Ambivalent Right) donated to political campaigns at the same rate or at a higher rate than the population in general¹⁵.

One reason we might not be seeing a disengaged middle is that our results pertain to one specific three-month period at the peak of the U.S. political fundraising season, prior to a hotly contested election, and Pew's 2014 data asked about any donation in the last two years. The percentage of people we see donating politically in Q3 shows slightly more engagement among Establishment Liberals and slightly less engagement among Committed Conservatives and the Populist Right, relative to other groups. Separately, we ask people who gave donations to a nonprofit organization to classify what cause they were supporting, and one of the 19 options was political causes.

This paints a very different picture, where the Outsider Left gives at twice the average rate and the Populist Right, Ambivalent Right, and Stressed Sideliners are underrepresented. Comparing this to the percent of each Pew group that gave in any form underscores that generosity is similar across all groups (61% is the average and 6 of 9 Pew Groups fall within +/-6% of the average), except the Outsider Left and Establishment Liberals, who are a bit more likely to give, and the Ambivalent Right, who are a bit less. If we compare advocacy for any cause (not necessarily political ones), we see the same trend. Taken together, we see no evidence that the ideological poles of the U.S. political spectrum (namely, Progressive Left and Faith & Flag Conservatives on the right) are driving political engagement in 2024, or that the centrist groups are any less engaged. Rather, the Outsider Left and Establishment Liberals are leading on money given to political causes and politicians, respectively.

Civic Intent and Pew Political Types

There are slight differences in civic intent among the Pew political types:

A smaller share of the Outsider Left have low civic intent and a larger share of the Ambivalent Right have low civic intent. Stressed Sideliners nearly match the average for all groups. Establishment Liberals and Committed Conservatives have the highest proportion of individuals with high civic intent (with scores in the top quintile).

Worldview and trust

When looking at trust for subgroups that hold specific beliefs about the world, we find some interesting relationships that tend to make sense:

People who believe there is no real difference in what the main political parties stand for trust people in general quite a bit less, and also trust nonprofits less. These people tend to be outsiders in the political system, according to Pew Research.

People who believe that "America's openness to people from all over the world is essential to who we are as a nation" show the largest increase in trust of nonprofits and people in general (of all the worldview questions we tested).

¹² Pew Research has studied the differences between political factions that are not fully aligned with the two major political parties since 1987, and they have found either 8 or 9 groups spanning the liberal-to-conservative spectrum in the US: https://www.pewresearch.org/short-reads/2022/01/05/americans-at-the-ends-of-the-ideological-spectrum-are-the-most-active-in-national-politics/ They provide access to all the raw data for over 30 questions at https://docs.google.com/spreadsheets/d/1-BjWhA-McZyEvjDp_kry9WPjK07GHrfw7werH4MeGGs/edit?gid=194263473#gid=194263473
¹³ Sample was over 10,000 responses from 2022-2023
¹⁴  [rectify with opensecrets et al]
¹⁵ Exact wording of our question was: [YES/NO] "Made a financial contribution to a political party, any political advocacy organization, a politician, or for any other person running for an elected position in public service." In a separate question, anyone who donated to a nonprofit organization, informal group or to help an individual was asked to choose the purpose or cause of that donation from 19 categories, and one of those was "Political/Politician/Public Policy."


Conclusions

Examining our data from previously unexplored angles has revealed evidence that people have more prosocial and depolarizing views than is commonly understood. We find that most people act prosocially and want to help others. People who strive to improve their communities and depolarize society can be found in similar numbers across the political spectrum. We find that the enabling environment that promotes generous action is a better predictor of giving than demographic differences. We also see prosocial behaviors evenly distributed across people with different worldviews and identities. In our data, we find indications that civic intent, as we define it, exists across various groups and segments of the U.S. population. 

We also developed a new indicator that captures the combination of prosocial behavior and good intentions that may lead people to participate in their communities. We present evidence that this civic intent measure is internally consistent, based on other measures of prosocial behavior. We hope that others will adopt the best parts of this tool and apply them in other contexts to advance our ability to understand attitudes and behaviors through this lens.

Mapping the motivations behind the generosity helps us develop more nuanced models of identity and associated behavior. We found that a combination of attitudes about helping and good intentions towards building broader community cohesion are closely associated with doing good, at least for the forms of prosocial behavior we track. With this research, we hope to provoke new research questions that will inform efforts to promote social cohesion and generous action.


Appendix

Relative contributions to Civic Intent measure

Figure A.1 is an expanded version of Figure 2.2, showing the extent to which each component question matters to overall civic intent. This unpacks more nuance about the underlying respondents and the way civic intent is calculated. Whereas Figure 2.2. shows the essential measure of "contribution" — the difference in scores between those who agreed versus disagreed with a statement – this also illustrates sample size effects.

Take the first question: Only 1 in 10 people initiated a helping effort in the past 7 days; 9 out of 10 people did not. The left starting point for this line represents the average civic intent score for those that did not, and the right end point, the score for those who did. At the other end, 9 out of 10 people have performed a random act of kindness for someone at some point. This is the reason the first question about "initiating" starts close to the overall civic intent average of 62, and the "random act of kindness" ends just over that average. Taking the average of 90% of the overall sample yields almost the same score. About half of the people affirmed a question in the middle, such as "I acted generously in the past month" so the line is centered about the overall mean.

Additionally, this shows how important a single question is in understanding what matters or drives civic intent, but the calculated contribution of each question to the score is essentially the same. All questions are weighted equally, though a few things may be counted twice because they appear twice on the list, or, in the case of acts of generosity, we factor in the variety of acts and the recency of acts.

So once it is clear why these difference lines move about the overall mean, and that affirming each statement only contributes to 1/16 of the overall civic intent score (based on weighting), we understand that the useful signal is the magnitude of the difference. Those who affirm a statement necessarily have much greater civic intent because of the many other things they report doing, that we asked about in other questions, which contribute to their average score being much higher. This reveals which questions matter most because they cross-correlate most with the overall measure. If you had to ask just a few questions, those with the largest differences shown would be the ones that best encompass all the other, related questions you might ask, and serve as a proxy for those answers, and thus are the essential core questions to measuring overall civic intent.

One of our main takeaways is that, broadly speaking, prosocial behavior in the context of community involvement and civic engagement can be best defined and summarized by just six questions that stand in for the rest. That argument follows from this line of thinking: ask many questions, find those that correlate best with as many others, and refine your survey instrument to only ask those questions. Our survey tested dozens of possible questions to arrive at these.


Another conclusion we draw from Figure A.1 is that these 16 questions each contribute something meaningful to the whole. The question with the weakest signal – about one's future intentions – shows a 10-point difference between those who affirm or reject it. The questions we excluded were primarily about attitudes and behaviors adjacent to prosociality but still correlated, such as awareness of the needs of others, world events, crises, and nonprofit news. Other questions related to the social part of prosocial, such as taking advice from friends and family when giving, did not measure helping acts directly. Likewise, though we found that while larger gifts and more involved acts of generosity correlated with higher civil intent, we limited the index to tracking whether a person took the action, and how often/recently, in part, to mitigate introducing confounding variables, such as access to wealth and access to time.

Validation: Civic Intent predicts the results our other prosocial survey questions

One indication that civic intent reliably captures prosocial behavior in a community context is that it leads to internally consistent results when we compare it against other survey questions that were not used in its construction.


Figures A.2, A.3, and A.4 summarize the effect of segregating groups based on various survey questions that were not used to calculate civic intent. We used the same calculations here as in Figure A.1, except that we also display the effect of each question on belonging scores (green arrows) beside civic intent (blue) for comparison. Arrow direction denotes an increase or decrease (red) of civic intent or belonging, respectively, among those who agreed with the corresponding statement. Those with greater awareness of recent events and exhibiting more prosocial behaviors (Figure A.2) saw larger increases in civil intent. By comparison, demographic variables were not correlated with increasing civic intent to the same extent. (Figure A.3).

There are many ways to segment a population based on demographics. Figure A.3 shows that most common demographic splits —age, income, gender, having children, rural/urban/suburban, religiosity, and employment status— have little effect on civic intent and either increase or decrease it slightly. Of these, religiosity has the biggest effect in both directions, for both belonging and civic intent.

If we examine additional contexts surrounding a person's last generous act we see that many traits that otherwise affect giving patterns do not have any real correlation with civic intent or belonging. Note that for all conditions shown in Figure A.4, the baseline civic intent or belonging is the average among people who gave in any form. In each case, the arrows represent differences in how a person gave, but the comparison was against givers who gave differently.

More about the correlation between civic intent and GivingTuesday

We reported that those with higher levels of civic intent also appeared to be more aware of and inspired by GivingTuesday. Here is an expanded version of that finding, showing the relationship for each quintile of civic intent score. At the lowest levels, people are mostly not aware of GivingTuesday and none who were aware were inspired. At the highest quintile, the majority were aware of GivingTuesday and 45% were either somewhat or very inspired by the event.

How civic intent shifts for populations with negative views or behaviors

It is useful to see that abstention from giving in any form in the previous week moves the distribution to lower values but the shape remains mostly intact. In contrast, lack of trust in people seems to flatten out the histogram everywhere and create spiky patterns, as multiple other questions probably shift for this group together. It appears that deeming generosity unimportant is a greater barrier to civic intent than is lack of trust; we see practically no one with civic intent scores above 65 with the former, but the high cutoff appears to be around 85-90 for the latter.

Other attitudes closely tied to depolarizing views

Figure A.7: Depolarization Figure 3.4, expanded to show the percentage that disagrees or agrees with each position (at the left or right end, respectively). The bars shown in Figure 3.4 are the differences between the groups.

Comparable prosociality measurement instruments

Those highlighted in red are questions that bear the greatest similarity to questions used in our GivingPulse survey and used to calculate civic intent.

Prosocial Tendencies Measure (revised) PTM-R:

  1. I can help others best when people are watching me.

  2. It makes me feel good when I can comfort someone who is very upset.

  3. When other people are around, it is easier for me to help others in need.

  4. I think that one of the best things about helping others is that it makes me look good.

  5. I get the most out of helping others when it is done in front of other people.

  6. I tend to help people who are in a real crisis or need.

  7. When people ask me to help them, I don’t hesitate.

  8. I prefer to donate money without anyone knowing.

  9. I tend to help people who are hurt badly.

  10. I believe that donating goods or money works best when I get some benefit.

  11. I tend to help others in need when they do not know who helped them.

  12. I tend to help others especially when they are really emotional.

  13. Helping others when I am being watched is when I work best.

  14. It is easy for me to help others when they are in a bad situation.

  15. Most of the time, I help others when they do not know who helped them.

  16. I believe I should receive more rewards for the time and energy I spend on volunteer service.

  17. I respond to helping others best when the situation is highly emotional.

  18. I never wait to help others when they ask for it.

  19. I think that helping others without them knowing is the best type of situation.

  20. One of the best things about doing charity work is that it looks good on my resume.

  21. Emotional situations make me want to help others in need.

  22. I often make donations without anyone knowing because they make me feel good.

  23. I feel that if I help someone, they should help me in the future.

  24. I often help even if I don’t think I will get anything out of helping.

  25. I usually help others when they are very upset.

Prosocial Scale for Adults (16 questions)

The following statements describe a large number of common situations. There are no right or wrong answers; the best answer is the immediate, spontaneous one. Read each phrase carefully and fill in the number that reflects your first reaction.

1 - Never/Almost Never | 2 - Rarely | 3 - Occasionally | 4 - Often | 5 - Always/Almost Always

  1. I am pleased to help my friends/colleagues in their activities.

  2. I share the things that I have with my friends.

  3. I try to help others.

  4. I am available for volunteer activities to help those who are in need.

  5. I am empathic with those who are in need.

  6. I help immediately those who are in need.

  7. I do what I can to help others avoid getting into trouble.

  8. I intensely feel what others feel.

  9. I am willing to make my knowledge and abilities available to others.

  10. I try to console those who are sad.

  11. I easily lend money or other things.

  12. I easily put myself in the shoes of those who are in discomfort.

  13. I try to be close to and take care of those who are in need.

  14. I easily share with friends any good opportunity that comes to me.

  15. I spend time with those friends who feel lonely.

  16. I immediately sense my friends’ discomfort even when it is not directly communicated to me.