• crowdfunding;
  • journalism;
  • news;
  • mass communication;
  • crowdsourcing


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

Crowd-funded journalism is a novel business model in which journalists rely on micropayments from ordinary people to finance their reporting. Based on analyses of the database of, a pioneering crowd-funded journalism website, we examine the impact of crowd-funded journalism on the news produced. We apply a uses and gratifications approach to study consumers' choices when they donate to crowd-funded journalism and find that consumers are more likely to donate to stories that provide them with practical guidance for daily living (e.g., stories about public health or local city infrastructure), as opposed to stories from which they gain a general awareness of the world (e.g., cultural diversity, or government and politics). We discuss the implications for the future of news.

In the United States, the newspaper industry has been facing unprecedented challenges and opportunities, largely because of the rapid evolution of information technology. In 2009, the newspaper industry faced an average of 26% in advertising losses, and advertising revenues fell 43% over the previous 3 years (Pew Project for Excellence in Journalism, 2010). The trend continued into the following 2 years (Pew Project for Excellence in Journalism, 2011).

The financial pressures upon newspapers have led to the downsizing of newsroom staff nationwide, with local newsrooms especially hard hit. Newspaper newsrooms are 30% smaller in 2011 than in 2000 (Pew Project for Excellence in Journalism, 2011). About 800,000 fewer stories are produced each year across the United States as a result of these cutbacks (Doctor, 2010). These cutbacks in local newsrooms are particularly disconcerting because local news is often the way people find out about news and information in a way that shapes their civic and personal identity (Klinenberg, 2005). The financial pressure on the press also hurts the quality of news. As emphasis is placed on increasing web traffic to attract digital revenue, reporters may devote less time to fact-checking and writing in-depth stories (Boczkowski, 2010).

As a response to the financial problems facing a large number of news organizations, journalism practitioners (Overholser, 2006) and scholars (Guensburg, 2008) have advocated for nonprofit media organizations as another way to supplement the news ecosystem. Nonprofit journalism outlets such as the Texas Tribune and ProPublica provide models in which professional journalists supplement traditional news, with their costs paid mainly by sizable endowments from philanthropists and grants.

In this paper, we study another response to the crisis in journalism, namely the flourishing of entrepreneurial projects that showcase crowdfunding—using micropayments by large numbers of people to finance creative projects—as a new business model for news. In particular, we examine the impact of crowdfunding on news as well as on journalism as a profession by focusing on, a nonprofit news organization founded in the San Francisco Bay Area in 2008. Its attempt to use crowdfunding to generate support for local news is unique among other business models for two reasons: 1) it allows individuals to donate to individual stories rather than to, the organization; 2) anyone can register as a reporter and pitch a reporting project for fundraising on the site.1' strategy for funding news was recognized by the Knight News Challenge2 as particularly innovative.3 The founder of, David Cohn, explained that he took his inspiration from the “Obama model” of fundraising (Kershaw, 2008), because about half of President Barack Obama's 2008 presidential campaign contributions came from hundreds of thousands of donations of $200 or less (Isikoff, 2008).

On, the process of pitching and publishing a story works as follows: A registered reporter submits a pitch—a short description of the proposed reporting project with a requested funding amount—which is then, in turn, reviewed by staffers for approval. These pitches are then shared with the public on the Web with the goal of raising money. Individual donors, registered reporters, and news organizations can all donate to these stories. When a story on is funded, it is usually published with a creative commons license,4 and the organization or the journalist may seek an additional venue for publication.5, active for over 4 years, provides an opportunity to study a number of research questions regarding the impact of crowdfunding on journalism—as it stands now—and to evaluate implications for the future. We ask: What type of news articles are likely to be produced in an environment in which news consumers can pay for individual stories? When journalism is crowdfunded, are there systematic patterns in reporters' relative success at getting funded based on their level of professional experience?

From a methodological point of view, provides a rich data set to study crowdfunded journalism because it keeps detailed records on almost all activities on its site, such as attributes of the pitches, reporters' profiles, the amount of each donation, and the attributes of each donor. These data provide us with nonintrusive measures of consumers' choices about news expressed by making monetary donations at the story level, rather than by responding to survey or interview questions about media programs or channels, as in many previous studies.

Through our research, we build on uses and gratifications (U&G) research and literature about crowdfunding. With respect to U&G, we extend the theory to apply to news one can buy (beyond news one simply consumes), and further distill the categories of guidance and surveillance as news needs. We also add to the growing body of research on crowdfunding with the first quantitative study to examine crowdfunding and news—focusing not just on consumers' news preferences but also on professionalism in the news industry.

We had expected donors to be particularly interested in civically minded news about their community because they were paying for news and also participating in this new, civic, online experiment. Instead, we learned that they were far less interested in this kind of news than they were in news that was of immediate concern to them—the news that serves as a practical guide for daily living. We had also expected that people would tend to fund the most experienced reporters, but we found that this was not the case—and, in our discussion, explore why less experienced reporters might do better in a crowd-funded model for news.

Literature Review and Hypotheses Development

  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

Crowd-funded journalism fundamentally shakes up what it means to be a news reader: Users can directly fund the stories they care about reading. In this way, is responsive to recent trends that suggest consumers are playing an increasingly active role in media production. Benkler (2006) argues that the hub and spoke model of mainstream media transmission has been displaced by a more networked public sphere, while scholars such as Gillmor (2004) and Lewis (2012) highlight the increasingly participatory role that consumers play in news creation through everything from building their own news sites to providing user-generated content for mainstream news. In crowd-funded journalism, consumers' financial contributions directly affect which stories actually get written. In this way, takes some of the power of decision-making away from the hands of journalists and places it into the hands of ordinary people.

Crowd-funded journalism is part of a general phenomenon called crowdsourcing, following Howe (2009), referring to the act of “an organization taking a task once performed by in-house employees and outsourcing it to a large and unclear crowd in the form of an open call.” Crowdsourcing has been applied to user-generated content (Nov, 2007), as well as product innovation and creative designs (Howe, 2009). The subject of the present study shares similarities with crowdsourcing, as they all rely on the participation of a crowd. Nevertheless, they differ in one key aspect: In crowdsourcing, the crowd contributes its time and efforts, but in crowd-funded journalism, or crowdfunding in general, the crowd makes monetary contributions.

Crowdfunding has been in existence since before the rise of the Web. The idea is that through small donations from the crowd, large projects that no ordinary individual could fund on his or her own can be started. A famous example is the community-funded American football team, Green Bay Packers. Recently, crowdfunding became a popular way to raise money for various creative projects (e.g.,, as well as for charity (e.g., Researchers on crowdfunding have examined the geographic patterns of the crowd (Agrawal, Catalini, & Goldfarb, 2011) as well as their contribution patterns over time (Burtch, Ghose, & Wattal, 2013). Our study differs from these studies by focusing on the crowd's topic preferences exhibited in their choice of stories in crowdfunded journalism.

Uses and Gratifications offers us an opportunity to study the kind of news consumers want to read, based on the news they fund. Because of this rationale, we chose to consider the case of through the application of uses and gratifications. U&G “sees media audiences as variably active communicators, rather than passive recipients of information” (Rubin, 2002, p.525), an idea that is congruent with the notion that the online audience is increasingly participatory.

While U&G certainly had its heyday before the Internet became a major source of information and entertainment (Katz, 1959; Katz, Gurevitch, & Hass, 1973), it still holds value in helping us understand why people make the news choices that they make, particularly in the era of the Web. Researchers continue to use U&G to study uses of the Internet (Charney & Greenberg, 2002), information-seeking behaviors and civic engagement (Nguyen, Western, & McKay, 2005), and additional work, particularly in political communication (e.g., Kaye & Johnson, 2002), echoes the legacy of U&G to consider the rationale behind media consumption choices. The theory itself has undergone reformulation across new developments, most radically with the birth of the Web and the rise of niche media (Ruggiero, 2000). LaRose and Eastin (2004), for example, have extended the original U&G frameworks to include such motivations as Internet self-efficacy and the ability to organize and execute online action. But the core premise, that people actively seek out specific media to meet specific needs, still has utility.

Certainly, user donations to crowd-funded journalism are not equivalent to the self-reported measures of media consumption typically employed in U&G research. To fill this gap, we draw on the willingness to pay (WTP) literature and buycott/boycott research to make an argument that U&G can be extended to describe consumers' choices in funding news. Much of the WTP research elicits consumers' willingness to pay for various products by asking them to make hypothetical purchasing decisions (see Horowitz & McConnell, 2002). The premise of this literature is consumer choice theory, a theory of neo-classical economics suggesting that consumers pay for products that best satisfy their needs (Mas-Colell, Whinston, & Green, 1995, p. 17–36). The WTP literature covers a variety of goods (e.g., chocolates and hunting licenses), including public goods that are shared among communities of people (e.g., public parks). Since produces public goods—its stories are freely available to all Internet users—the theory of consumer choice also applies to our study.6 That is, consumer choice theory provides a justification for extending the U&G literature to factor in people's willingness to pay as a reflection of their media consumption motivations.7

The literature on buycott and boycott and political engagement suggests another justification for extending the U&G framework to paying for news. Research in political communication has found that people may assess what they need to buy in terms of their particular stances on issues (Nielson, 2010). These buy and do-not-buy decisions are ways to measure civic engagement. In the context of, donors could be consciously considering whether to buy or not buy information about local news or politics. With this in mind, our work further extends more recent scholarship on the relationship between U&G and civic engagement. For example, Kaye and Johnson (2002) have studied extensively how and why people search online sources for information—and find significant correlations between people who seek political information online and their propensity for civic action.

Predicting Consumers' News Preferences

Early U&G studies (Berelson, 1949; Blumler & McQuail, 1968) identified four key gratifications associated with reading news, including surveillance (the need to understand the world around us), guidance (practical guidance for daily living), escapism (escaping from the routines of life via comics or human interest stories), and entertainment. Further U&G studies showed that newspapers mainly meet people's information needs (surveillance and guidance) rather than affective needs (escapism and entertainment) (Katz, Gurevitch, & Hass, 1973).

Although separating information needs from affective needs was useful, U&G theory does not adequately distinguish between various types of information needs, i.e., guidance and surveillance. For example, Ancu and Cozma (2009) noted that the surveillance function of the Web is stronger than it was in an era of television, but in their study on why people might “friend” political candidates, they fail to distinguish between the kind of information-seeking behaviors associated with surveillance and guidance. Similarly, the distinction between the two terms was also obscured in Kaye and Johnson's U&G study of news (2002, p. 62), in which guidance and surveillance were defined similarly: “actively searching for specific information.”

However, drawing from political communication, the emergence of the monitorial citizen model (explained below) suggests a rationale for keeping guidance and surveillance separate. Guidance is associated with “news you can use” (Zaller, 1999, 2003; Hamilton, 2004, p. 266–275)—fulfilling news readers' needs for practical orientation about daily life. Such news topics might include financial news, weather reports, and shopping guides (Underwood, 1993).

Surveillance, on the other hand, is associated with the model of monitorial citizenship which suggests that most citizens scan the environment, looking out for events that might affect their lives, but do not devote time to following the details of such news (Schudson, 2003). A monitorial citizen is awakened by a “burglar alarm” style of news which is bound up in the idea of a citizen that ideally will be convinced to actively follow the unfolding drama of news. The burglar alarm is not about orientation for daily life, but rather a wake-up call to pay attention to serious news that requires a response (Zaller, 2003). While Schudson's (2003) and Zaller (2003)'s accounts of citizenship differ, they agree that there is a difference between citizens actively seeking information that is relevant to daily actions – choosing a product, a way to commute, and citizens scanning the news so as not to miss any important issues.

Empirical research on news readers' interests suggests that compared to journalists' choices of news, readers are more interested in “news you can use” than in general information about society. By comparing readers' interests in various topics and editors' importance ratings for news topics, Bogart (1989) finds that “readers want more of certain kinds of content than they presently get,” including topics such as health, nutrition, medical advice, consumer news, travel, and home maintenance and repair (p. 294). Similar findings have been reported in studies supported by the news industry. The Readership Institute8 conducted a series of studies of reader experiences by interviewing more than 37,000 consumers of more than 100 large and small U.S. daily newspapers. The research finds that readers wanted more “go and do” information in news stories (such as websites and address), and “more health, home, food, fashion and travel coverage” (Stepp, 2004). Although some of these topics are not the focus of, these arguments indicate that people might use private utility as a criterion to assess the value of a story before they make a donation on Finally, by interviewing a small number of reporters and donors, Aitamurto (2011) discovered that donors often donate because “the pitch was relevant in their lives, or the topic affected the lives of their friends or relatives.”

To summarize, the monitorial citizen model and the empirical evidence suggest that consumers' interests in funding surveillance news will be low, and their interests in funding guidance news will be high. Therefore, we hypothesize:

Hypothesis 1: News stories associated with the surveillance gratification are less likely to be funded than those that are not associated with it.

Hypothesis 2: News stories associated with the guidance gratification are more likely to be funded than those that are not associated with it.

Journalism Practices

The Internet has brought a lot of changes to the work routines of journalists (Domingo & Paterson, 2011). Multitasking and reporting in multiple media have become common practices (Klinenberg, 2005). Crowd-funded journalism adds yet another task to journalists' responsibilities for those who choose to solicit funds from the public for their reporting projects. In the meantime, it should be noted that in the age of the Internet, the boundary between professional and non-professional journalists has blurred (Gillmor, 2004). Our data indicate that reporters range from Pulitzer Prize winners to part-time English teachers and college students. While the doors might be open to new and veteran journalists, the question is whether the playing field for fundraising remains an equal contest. offers an opportunity to study professional journalists' performance as fundraisers as compared to less experienced journalists, some of whom may have never worked for a professional news organization.

We expect professional journalists, with knowledge of writing pitches and understanding potential newsworthiness, might be more successful in raising funds. We also might expect donors to prefer journalists with more experience. As we have indicated, many local and big-city newspapers have downsized their newsroom staff, which has contributed to an overall loss in news coverage (Doctor, 2010). As a result, people may now desire to read stories that are no longer being covered by their local newspapers—and might want to have this coverage from experienced reporters. Academics, politicians, and journalists themselves bemoan the loss of those who know how to practice the reportorial craft of thorough and impartial investigative reporting (Usher, 2010) and suggest that people should want experienced journalists (Starr, 2009), since the ability to perform such activities, such as detailed investigative reporting, is associated with the experience, tenacity and institutional support of large newspapers. Taken together, we hypothesize the following:

Hypothesis 3: Reporters with more experience working with traditional news organizations are more likely to receive full funding for their pitches.

However, our hypothesis also raises a descriptive question. Given that civic news output is diminishing, donors might be expected to want professional journalists who are trained to gather civic news information to continue doing so. We are curious about whether donors indeed prefer professional journalists to be providing the quality, supplemental journalism that is going unreported in traditional news outlets – or whether they prefer to support less experienced journalists instead. We pose the following research question:

RQ1: Do donors prefer to donate to journalists with more professional experience?


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

We obtained a copy of the entire database of with all sensitive user information removed. The database captures all the attributes of each pitch posted on (e.g., its topics, the reporter's ID, the amount requested, the completion date of the fundraising, and whether the pitch was funded), all the observable attributes of each donation (e.g., the time it was made, the amount, the pitch which received it, and the donor's user ID), and the profile information of each reporter, donor, and participating news organization. The data spanned the period between Oct 20, 2008, the day on which the first donation was made on, and May 16, 2011.

Our data show that editors approved 234 pitches for fundraising on the site,9 out of which 102 garnered sufficient funds for production. Each pitch was about an issue in one of the three geographic regions—the San Francisco Bay area (76%), the Los Angeles area (21%), and the Seattle area (3%). Overall, 10,227 donations were made, the amounts of which ranged from $1 to $2,500. Among the 10,264 registered individual donors (including both reporters and citizens), 5,376 (52%) each made at least one donation, and 3,564 (66%) donated only once.

Before delving into donors' news preferences, it was first necessary to verify that crowdfunding was actually taking place at Since technically does not limit the amount a news organization can donate to any story, if news organizations in fact fund all the stories, would not be an interesting crowdfunding phenomenon to study and more importantly, it would not contribute to crowd-funded journalism as a new funding model. We present descriptive statistics in the appendix and show that, indeed, the news stories published on are crowdfunded and that they mostly reflect the preferences of individual donors.

Measuring Reporter Experience

Each pitch on contains a section called “Qualifications,” which describes the reporter's experience. We invited four independent raters (students at the University of Southern California) to read these qualifications and assign a rating, on the scale of 1 to 10 (10 being the strongest), to each reporter. They were instructed to rate the reporters based on the number of years of reporting experience, the prestige level of the media organizations they worked for, education (if reported), and awards (if any). We obtained excellent inter-rater reliability. The intraclass correlation coefficient (ICC[3,4]10)—a generalization of Cohen's Kappa to multi-rater cases —for the mean of the four raters' ratings was 0.848 (p < 0.01). Overall, the mean experience rating of the reporters in our data was 5.59 (SD = 1.85).

Survey Measures

In order to answer RQ1, whether donors prefer experienced reporters, we piggybacked a question in a survey of donors.11 371 donors responded to the survey. They constituted 85% of the donors who donated during the period when responses were solicited to the survey (from Apr 16 to Apr 30, 2011). The respondents were asked the following question:

Consider two similar pitches: A and B. A's reporter has had 15 years of reporting experience but B's reporter has only two years. Which pitch would you be more likely to donate to?


Two samples of pitches were constructed for our study. First, out of 234 pitches in the dataset, all the pitches that had completed their fundraising (i.e., the expiration date had passed) were selected to create an All Pitch sample with a size of 210. This sample is representative of all the pitches posted on A second sample was constructed in response to possible biases in the All Pitch sample due to repeated observations of a few reporters, i.e., some reporters created multiple stories.12 Specifically, the 210 pitches in the All Pitch sample were created by 152 unique reporters, and thus 58 pitches were created by 34 reporters returning after their first pitch. Unobservable characteristics of the individual reporters could have a large impact on the fundraising outcome. For example, a reporter's friends and family might donate to some or all of his or her pitches. If these individuals favor a particular set of topics, our results on donors' preferences would be biased towards such topics. Therefore, we constructed a Unique Reporter sample with a size of 152 by removing the 58 pitches created by returning reporters. This sample is representative of the reporter population on

Analysis Method

As we focused on the effects of multiple characteristics of a pitch on the likelihood that it would be funded, a multiple regression was used to separately identify the effect of each independent variable. Since our dependent variable was dichotomous (whether a pitch received the amount its reporter requested by its expiration date), we used a logistic regression model in which the binary outcome was modeled in probability terms and was specified as odds. The coefficient (called odds ratio) on each independent variable describes how much the odds of being funded changes as the independent variable increases by one unit.

To examine donors' news preferences with different granularity, we constructed two regression models with different sets of independent variables. One model uses the 13 topics directly as independent variables (listed under the column “Topic” in Table 1), taking advantage of' feature that any story can belong to multiple topics. As our multiple regression analysis simultaneously takes into account the effects of multiple topics on each story, we can separately identify the effect of each topic on a story's fundability.

Table 1. Hypotheses on donors' funding preferences
GuidancePublic healthH1a+
 City infrastructureH1b+
 Consumer protectionH1c+
SurveillanceGovernment and politicsH2a
 Wealth and povertyH2b
 Media accountabilityH2d
 Race & DemographicsH2e
 Cultural diversityH2f
 Local science & businessH2i
 Criminal justiceH2j

A second model was created to test Hypothesis 1 and 2 directly, with two main independent variables: guidance and surveillance. The thirteen topics were mapped onto these two variables. Specifically, guidance is associated with topics that contain information immediately useful in readers' daily lives. Prior research identified three such topics, namely health, consumer news, and travel (Bogart, 1989, p. 294; Stepp, 2004). These topics can directly map onto the following topics: public health (e.g., “Is it time to get serious about coal ash?” (WFPL News, 2011)), consumer protection (e.g., “Law Banning Lead in Children's Clothing Faces Roadblocks Ahead” (Goenka, 2009)), and city infrastructure (e.g., “San Francisco's Muni: Why can't it run on time?” (SF Public Press, 2010a). The last one was because we believed readers' interests in “go and do” information (Stepp, 2004) could be reflected in their interests in information about city infrastructure, a topic that contains many stories on public transportation and local public facilities.

The remaining 10 topics were considered to be associated with surveillance, as they provide information about the societal environment, but do not, in general, offer immediate utility to readers. For example, the topic of government and politics (e.g., “Independent city hall reporting” (SF Appeal, 2009)), as well as media accountability (e.g., “The future of Bay Area newspapers in a digital age and changing economy” (Crowe, 2009)), naturally fall under surveillance. Wealth and poverty (e.g., “Tales of Two Census Tracts: San Francisco, Rich and Poor” (Cook, 2011)), race and demographics (e.g., “Stories from MacArthur Park” (Browne, 2010)), cultural diversity (e.g., “The soundtrack of the Arab Spring” (FSRN Documentaries, 2011)), and environment (e.g., “How green are cruise ship tourism dollars?” (InvestigateWest, 2010)) all focus on the status of various demographics in the society and the general environment, and should fall under surveillance as well. Lastly, we considered education, employment, criminal justice, and local science and business as surveillance as they were not identified in prior literature as particularly relevant to guidance.

Certainly, there exist alternative ways of categorizing the news topics. To ensure the robustness of our results to these alternative categorizations, we conducted a robustness analysis by running the same analysis based on alternative constructions of guidance and surveillance (see appendix13). Overall, our results are robust to these alternative constructions.

Based on this grouping, we constructed two independent variables: guidance (MAll = 0.56, SDAll = 0.63; MUnique = 0.49, SDUnique = 0.62) and surveillance (MAll = 2.4, SDAll = 1.82; MUnique = 2.54, SDUnique = 1.75) by adding up the dummy variables for the topics in each of these two groups. For example, if a story was about the city infrastructure and government and politics, it would have a guidance measure of 1 (due to city infrastructure), and a surveillance measure of 1 (due to government and politics). We also generated specific predictions for the individual topics and marked them in the column “Prediction” in Table 1. These hypotheses were numbered based on their groups: H1a ˜ H1c were for the guidance group and H2a ˜ H2j were for the surveillance group. Topics marked with a “+” or “-” were predicted to have a positive or negative impact, respectively, on a story's fundability.

In our regression model, we controlled for the geographic area of the pitch and the requested amount. This was because stories about the San Francisco Bay area, where was initially launched, had a higher funding rate than those about the Los Angeles and Seattle area, where had expanded at a later stage (see details in Table 2). We also clustered the standard errors according to the geographic area, as we expected that based on the local environment in each region, donors' preferences might be correlated in ways unknown to the researchers. Lastly, we also expected the requested amount (MAll = 1,432, SDAll = 3,288; MUnique = 1,617, SDUnique = 3,646) to have an effect on the fundability of a pitch, as the larger the requested amount the harder it would be to raise the necessary funds.

Table 2. Results of the logistic regression predicting the fundability of a pitch
 All Pitch sampleUnique Reporter sample
Model 1Model 2Model 1Model 2
  1. a

    p < 0.1;

  2. b

    p < 0.05;

  3. c

    p < 0.01

  4. Standard errors are clustered by geographic area.

Guidance1.48b 1.54c 
Public Health 1.89c 1.52c
City Infrastructure 1.44c 2.16c
Consumer Protection .72 .67
Surveillance0.89 0.89c 
Government & Politics .85b .76c
Wealth & Poverty .91 .89a
Environment .82c .86
Race and Demographics 1.15 1.29c
Cultural Diversity .64c .67c
Education .8 .6
Employment .82 .8
Media Accountability .42c .43c
Local Science & Business 1.07 1.06
Criminal Justice 1.46 1.49
Requested amount0.99c.99c0.99c.99c
San Francisco Bay Area2.46c2.8c3.72c3.37c
Los Angeles0.52c.58c0.56c.49c
Log Pseudolikelihood−126.39−123.73−86.25−83.86
Pseudo R20.


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

Donors' Preferences for News

Table 2 contains regression results from both the All Pitch and Unique Reporter samples, and from Model 1, which uses guidance and surveillance as the main independent variables, and Model 2, which expands these two variables into individual topics. Overall, results from the two samples are qualitatively similar, except that the significance levels for the coefficients on three variables, namely wealth and poverty, environment, and race and demographics, differ somewhat. To simplify the presentation, we focus on the Unique Reporter sample to discuss our findings, as this sample involves less bias due to multiple pitches by the same reporters.

First, we find supporting evidence for Hypothesis 1—the positive effect of topics associated with the guidance gratification. The coefficients on guidance in both samples are greater than 1 (BAll = 0.39, odds ratioAll = 1.48, p < 0.05; BUnique = 0.43, o.r.Unique = 1.54, p < 0.001). At the individual topic level, Hypothesis 1a (public health) and Hypothesis 1b (city infrastructure) are both supported with fairly large effect sizes. A story about public health has odds of being funded 1.52 times greater than one not about public health (BUnique = 0.42, o.r.Unique = 1.52, p < 0.001). A story about city infrastructure has odds of being funded 2.16 times greater than one not about this topic (BUnique = 0.77, o.r.Unique = 2.16, p = 0.002). Hypothesis 1c (consumer protection), however, is not supported (BUnique = −0.4, o.r.Unique = 0.67, p = 0.65).

We also find support for Hypothesis 2—the negative effect of topics associated with the surveillance gratification. The coefficients on surveillance are similar from both samples, but are of different statistical significance levels (BAll = −0.11, o.r. All = 0.89, p = 0.21; BUnique = −0.11, o.r.Unique = 0.89, p < 0.001). At the individual topic level, as expected, H2a (government and politics) is strongly supported: a government and politics story's odds of being funded are 24% lower than those not relevant to this topic (BUnique = −0.27, o.r.Unique = 0.76, p < 0.001). Similarly, the negative effects of media accountability (H2b, BUnique = −0.84, o.r.Unique =0.43, p < 0.001), wealth and poverty (H2c, BUnique = −0.1, o.r.Unique = 0.89, p = 0.09), and cultural diversity (H2e, BUnique = −0.4, o.r.Unique = 0.67, p < 0.001) are all supported. There is also a negative effect of the topic of environment (BAll = −0.19, o. r.All = 0.82, p < 0.001) as shown in the All-Pitch model. The only exception is the estimated odds ratio on the topic of race and demographics, which is greater than 1 (BUnique = 0.25, o.r.Unique = 1.29, p < 0.001).

Journalism Practices

Hypothesis 3 (more experienced reporters are more successful in fundraising) is not supported. Rather, we find evidence that suggests the opposite: the coefficients on experience in all four models are less than 1 (e.g., BUnique, Model 2 = −0.18, o.r.Unique, Model 2 = 0.84, p = 0.03). In addition, our survey measure aimed at answering RQ1 yields the following results about consumers' preference for experienced reporters.179 (48%) chose the answer “I do not usually look at the reporter's experience.” 84 (23%) chose “I don't know.” The remaining 108 (29%) had a clear preference between the two hypothetical reporters but they were about evenly split: 61 (16%) chose the one with 15 years of experience, and 47 (13%) chose the one with only two years of experience. Taken together, the survey results show that most donors (71%) either did not consider reporters' experience when choosing pitches to donate to, or had no clear preference between experienced and inexperienced reporters.


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

We applied a U&G approach to study consumers' choices in donating to crowd-funded journalism, and found that compared to reporters, consumers favor stories that would provide them with practical guidance for daily living (e.g., public health or city infrastructure), as opposed to stories from which they gain a general awareness of the world (e.g., government and politics). We also discovered that reporters with less experience working with traditional news organizations tended to be more successful in raising funds from the crowd. Using survey data we ruled out the possibility that this was due to donors favoring inexperienced reporters.

Before we discuss the implications of these findings, we noticed that, contrary to our predictions, the effect of race and demographics on a story's fundability was positive (BUnique = 0.25, o.r.Unique = 1.29, p < 0.001). We suspect that this is because some pitches in this category, e.g., “Stories from MacArthur Park” (Browne, 2010), exhibit characteristics of human interest stories, which frequently attract news readers (Curran, Douglas, & Whannel, 1980).


Our finding that donors prefer “news you can use” to surveillance news has implications for the role that crowd-funded journalism will play in the future of news. First of all, does not seem to be a place where people come, en masse, to fund general news about society. Donors to crowdfunded journalism seem to have a taste for specific news topics that are of immediate utility to them in daily living.

This result seems to justify some scholars' concern that if consumers, who are well known to prefer non-public affairs news, play an important role in news production, coverage of general public affairs news would decrease (Boczkowski & Peer, 2011). Nevertheless, these results should not be interpreted as implying that donors do not want general public affairs news at all. Rather, these results should be interpreted as suggesting a difference between consumers' and journalists' news choices, because in our analysis, consumers' preferences were measured by implicitly comparing their news choices with the journalists'. After all, consumers could only choose to fund pitches that were already chosen by journalists. Such a difference between consumers' and journalists' choice of news is well known and has persisted for decades in the United States (Bogart, 1989; Boczkowski & Peer, 2011). Here, we find that it still exists in the context of crowd-funded reporting.

In addition, we argue that the fact that donors prefer specific “news you can use” does not diminish the value of crowdfunded journalism. On the contrary, crowdfunded journalism offers a channel in which news consumers express their preferences for local news that they deem important to their own lives. Such news is essential for thousands of local communities to have their voices heard in the media. In this regard, consumers are actively defining what they deem as important public affairs news.

This study also reveals interesting dynamics in the fundraising activities of journalists. We believe the differences in the success rates of reporters with varying levels of experience might simply reflect how reporters use For example, experienced reporters might have other sources of funding such that they are less financially dependent on, which perhaps means they will exert less effort in fundraising for their pitches. Or, they may be uncomfortable using social media for fundraising (Aitamurto, 2011). Journalists working in established newsrooms feel uncomfortable about using social media in a way that could make their work appear less than objective (Usher, 2011). Even those that put opinions on Twitter are aware that they are potentially representing a news organization's opinion, and try to clearly differentiate their own personal tweets (Usher, 2011).

The fact that donors do not prefer either experienced or inexperienced reporters might offer some hope for young reporters seeking a way to support the reporting that they believe society needs, or to jump-start their careers. In the meantime, we also need experienced journalists' expertise to supplement the losses in news coverage. If these experienced journalists are constrained by their own institutional and generational limitations against raising money directly from individuals (especially through social media), the question remains how we can financially facilitate experienced journalists to report on stories that might benefit from their expertise, while still encouraging the new voices that are now part of journalism's ecosystem.


Our sample size is relatively small (210 pitches), which may affect the reliability of the results. Unobserved fundraising practices might affect the fundability of stories on a case-by-case basis. Further, crowdfunded journalism is a relatively new phenomenon. Thus the news preferences exhibited by donors for a future, perhaps mature, crowdfunding site might be different from what we observe today. Last, has been active mainly in California. The reporter and donor population in this study might not be representative of the wider population in the United States.


This research is the first quantitative analysis of the news preferences of donors for crowd-funded journalism. This study contributes to the greater body of U&G research primarily for the novelty of the data, reflected in two aspects. First, consumers' choices were expressed by making monetary donations (i.e., putting your money where your mouth is), rather than by responding to survey questions, as in previous U&G studies. We argue that consumers' preferences expressed this way are worth studying as they reflect what consumers truly value. Second, as the stories which did or did not receive full funding could be observed, the relative intensities of consumers' preferences can be inferred. This information helps to address, in the context of crowd-funded journalism, one main critique of U&G—that it is silent to the frequency distribution of various motivations for media use in the population (Ruggiero, 2000).

Our finding on consumers' news preferences extends the research on the gap between what journalists and consumers deem as newsworthy (Boczkowski & Peer, 2011) from simply examining what consumers read to what consumers actually fund. As consumers assume an increasingly active role in producing news, we find that the choice gap between consumers and journalists still exists. In crowd-funded journalism in particular, consumers prefer specific news useful in their daily living whereas journalists tend to focus on general public affairs news.

Last, this research also addresses changing practices of journalists. We provide a preliminary analysis of journalists' performance as fundraisers, and discuss implications of the different success rates among journalists with varying levels of professional experiences.


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

This research extends the study of news consumers' preferences in the context of crowd-funded journalism, and confirms prior findings that consumers are much more interested in news that provides them with guidance to daily living than news from which they gain a general understanding of the world. This work provides evidence of the impact of crowdfunding on journalism, as news consumers start to play an active role in the production of news.

Our findings suggest that crowdfunded journalism could work to fill an essential gap in public knowledge—if given the chance to do so. As Kim and Ball-Rokeach (2006) observed, quality local news can create shared ties among community members, facilitating a sense of belonging and a desire to participate, volunteer, and engage in the community. Thus, crowdfunded journalism signal the growing recognition that something must be done to replenish the dwindling supply of local news, not just for accountability's sake, but for the larger nature of civic engagement.


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies

We thank David Cohn for providing us the server-side data of Spot.Us, and Sean Aday, Catie Bailard, Cliff Lampe, Rick Wash, and two anonymous reviewers for their helpful comments. Lian Jian acknowledges the financial support of the Advancing Scholarship in the Humanities and Social Sciences program by the Office of the Provost at the University of Southern California.

  1. 1

    The main criterion for determining whether a story is approved is whether “it consists of real original reporting”— whether the reporter proposes to discover new facts rather than offer his/her opinion, e.g., a movie review (D. Cohn, personal communication, June 20, 2011).

  2. 2

    The Knight News Challenge is a “media innovation contest that aims to advance the future of news by finding new ways to digitally inform communities” (Knight News Challenge, n.d.). was awarded an initial grant of $340,000 in 2008. See details at

  3. 3

    In late 2011, was acquired by American Public Media's Public Insight Network.

  4. 4

    A creative commons license essentially means that the work can be distributed worldwide free of charge, under certain terms and conditions. More details can be found at

  5. 5 has over 105 publishing partners, which have included The New York Times, McSweeney's, Miller-McCune, as well as smaller organizations such as the San Francisco Public Press. For more details about, see

  6. 6

    Applying consumer choice theory to donations seems to imply that donations are necessarily tied to consuming the news story later. This is not necessarily the case, as in consumer choice theory, consumers choose to pay based on their perceived value of the goods, which may not be the realized value after consumption.

  7. 7

    WTP studies about news have been limited to people's willingness to pay for online news, suggesting that most people do not wish to pay for online news, in part because it is widely available and free online (Chyi, 2005). Yet provides news that is not available elsewhere. Thus these findings do not apply to stories on

  8. 8

    The Readership Institute is associated with the Newspaper Association of America and the American Society of News Editors.

  9. 9

    A small number of pitches (13) were not labeled a specific geographic area. As this article focuses on local news, these pitches were not included in our analyses.

  10. 10

    The first number in the square brackets indicates the statistical model assumed and the second number the number of raters. Statistical model 3 assumes that the raters are fixed and not randomly drawn from a target population. More details on ICC can be found in Shrout, P., & Fleiss, J. (1979). Intraclass correlations: uses in assessing rater reliability. Psychological bulletin, 86 (2), 420–428.

  11. 11

    The details of the survey can be found in Jian and Shin's (2013) study.

  12. 12

    Although one way to account for repeated observations of the same reporter would be to create a panel, the unbalanced distribution of the number of pitches each reporter created, especially given that the majority (78%) of the reporters only had one pitch and some had as many as seven pitches, prevented us from creating an effective panel.

  13. 13


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies
  • Agrawal, A., Catalini, C., & Goldfarb, A. (2011). Friends, family, and the flat world: the geography of crowdfunding. NBER Working Paper.
  • Aitamurto, T. (2011). The impact of crowdfunding on journalism. Journalism Practice, 5(4), 429445.
  • Ancu, M., & Cozma, R. (2009). MySpace politics: Uses and gratifications of befriending candidates. Journal of Broadcasting & Electronic Media, 53(4), 2009.
  • Benkler, Y. (2006). The wealth of networks: How social production transforms markets and freedom. New Haven, CT: Yale University Press.
  • Berelson, B. (1949). What “missing the newspaper” means. In P. F. Lazarsfeld & F. N. Stanton (Eds.), Communication research 1948–1949 (pp. 111129). New York: Harper.
  • Blumler, J. G., & McQuail, D. (1968). Television in politics: Its uses and influence. London: Faber and Faber.
  • Boczkowski, P. J. (2010). News at Work: Imitation in an age of information abundance. Chicago: University of Chicago Press.
  • Boczkowski, P. J., & Peer, L. (2011). The choice gap: The divergent online news preferences of journalists and consumers. Journal of Communication, 61(5), 857876.
  • Bogart, L. (1989). Press and public: Who reads what, when, where, and why in American newspapers (2nd ed.). Hillsdale, NJ: Erlbaum.
  • Browne, D. (2010). Stories from MacArthur Park. URL:
  • Burtch, G., Ghose, A., & Wattal, S. (2013). An empirical examination of the antecedents and consequences of contribution patterns in crowd-funded markets. Information Systems Research. Advance online publication. doi: 10.1287/isre.1120.0468
  • Charney, T., & Greenberg, B. S. (2002). Uses and gratifications of the internet. In C. A. Lin & D. J. Atkin (Eds.), Communication technology and society: Audience adoption and uses (pp. 379407). Cresskill, NJ: Hampton Press, Incorporated.
  • Chyi, H. I. (2005). Willingness to pay for online news: An empirical study on the viability of the subscription model. Journal of Media Economics, 18(2), 131142.
  • Cook, C. (2011). Tales of two census tracts: San Francisco, Rich and Poor.
  • Crowe, A. (2009) The future of Bay Area newspapers in a digital age and changing economy.
  • Curran, J., Douglas, A., & Whannel, G. (1980). The political economy of the human-interest story. In A. Smith (Ed.), Newspapers and democracy. Cambridge, MA: MIT Press.
  • Doctor, K. (2010). Newsonomics: Twelve new trends that will shape the news you get. New York, NY: St. Martin's Press.
  • Domingo, D. & Paterson, C. (Eds.). (2011). Making online news – Volume 2: Newsroom ethnography in the second decade of Internet journalism. New York: Peter Lang.
  • FSRN Documentaries (2011). The soundtrack of the Arab Spring.
  • Gillmor, D. (2004). We the media: Grassroots journalism by the people, for the people. Sebastopol, CA: O'Reilly.
  • Goenka, A. (2009). Law banning lead in children's clothing faces roadblocks ahead. Retrieved from' archive as it was not online.
  • Guensburg, C. (2008). Nonprofit news. American Journalism Review, 30(1), 2633.
  • Hamilton, J. T. (2004). All the news that's fit to sell: how the market transforms information into news. Princeton, NJ: Princeton University Press.
  • Horowitz, J. K., & McConnell, K. E. (2002). A review of WTA/WTP studies. Journal of Environmental Economics and Management, 44, 426447.
  • Howe, J. (2009). Crowdsourcing: Why the power of the crowd is driving the future of business. New York, NJ: Three Rivers Press.
  • InvestigateWest (2010). How green are cruise ship tourism dollars?
  • Isikoff, M. (2008, October 4). Obama's “good will” hunting. Newsweek.
  • Kaye, B. and Johnson, T. (2002). Online and in the know: Uses and gratifications of the web for political information. Journal of Broadcasting & Electronic Media. 46(1), 5472.
  • Katz, E. (1959). Mass communication research and the study of culture. Studies in Public Communication, 2, 16.
  • Katz, E., Gurevitch, M., & Haas, H. (1973). On the use of mass media for important things. American Sociological Review, 38, 164181.
  • Kershaw, S. (2008, 23 Aug). A different way to pay for news you want. The New York Times.
  • Kim, Y. and Ball-Rokeach, S. J. (2006). Civic engagement from a communication infrastructure perspective. Communication Theory, 16(2), 173197.
  • Klinenberg, E. (2005). Convergence: News production in a digital age. The ANNALS of the American Academy of Political and Social Science, 597, 4864.
  • LaRose, R. & Eastin, M. S. (2004). A social cognitive theory of Internet uses and gratifications: Toward a new model of media attendance. Journal of Broadcasting & Electronic Media, 48(3), 358377.
  • Lewis, S. C. (2012). The tension between professional control and open participation. Information, Communication & Society, 15(6), 836866.
  • Mas-Colell, A., Whinston, M. D., & Green, J. R. (1995). Microeconomic theory. Oxford University Press: New York, NY.
  • Neilson, L. A. (2010). Boycott or buycott? Understanding political consumerism. Journal of Consumer Behaviour, 9(3), 214227.
  • Nov, O. (2007). What motivates Wikipedians? Communications of the ACM, 50(11).
  • Nguyen, A. F. E., Western, M. & McKay, S. (2005). Online news in Australia: Patterns of use and gratification. Australian Studies in Journalism, 15, 534.
  • Overholser, G. (2006). On behalf of journalism: A manifesto for change. Philadelphia: Annenberg Public Policy Center.
  • Pew Project for Excellence in Journalism (2010). The state of the news media: An annual report on American journalism.URL:
  • Pew Project for Excellence in Journalism. (2011). The state of the news media: An annual report on American journalism.URL:
  • Rubin, A. M. (2002). The uses-and-gratifications perspective of media effects. In Bryant, J. & Zillman, D., & Oliver, M. eds., Media effects: Advances in theory and research, 2nd ed., pp. 52548. Mahwah, NJ: Erlbaum.
  • Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass Communication & Society, 3(1), 337.
  • Schudson, M. (2003). The sociology of the news. New York: Norton.
  • SF Public Press (2010). San Francisco's Muni: Why can't it run on time?
  • SF Appeal (2009). Independent city hall reporting.
  • Jian, L., & Shin, J. (2012). Driving forces behind donors' contribution to crowdfunded journalism. Working paper.
  • Starr, P. (2009, March 4). Goodbye to the era of newspapers (hello to a new era of corruption). The New Republic, 28–35.
  • Stepp, C. S. (2004, January). What they like. American Journalism Review.URL:
  • Underwood, D. (1993). When MBAs rule the newsroom. New York: Free Press.
  • Usher, N. (2010). Goodbye to the news: How out of work journalists assess enduring news values and the new media landscape. New Media & Society, 12(6), 911928.
  • Usher, N. (2011). Making business news in the digital age. Unpublished doctoral dissertation, University of Southern California, Los Angeles, California.
  • WFPL News (2011). Is it time to get serious about coal ash?
  • Zaller, J. (1999). Market competition and news quality. Paper presented at the 1999 annual meetings of the American Political Science Association, Atlanta, GA.
  • Zaller, J. (2003). A new standard of news quality: Burglar alarms for the monitorial citizen. Political Communication, 20(2), 109.


  1. Top of page
  2. Abstract
  3. Literature Review and Hypotheses Development
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgement
  9. References
  10. Biographies
  • Lian Jian is an assistant professor in the Annenberg School for Communication at the University of Southern California. Her primary research interests focus on the economics of online information systems. She studies collective intelligence, the phenomena of crowd-sourcing and crowd-funding, and the business models of news.

    Address: 3502 Watt Way, Los Angeles, CA 90089, USA

  • Nikki Usher is an assistant professor at the School of Media and Public Affairs at George Washington University. Her primary research interest is news production in the digital age. She studies newsrooms, journalism outside newsrooms, and the changing definition of journalism through intensive fieldwork.

    Address: 805 21st NW, Suite 400, Washington DC 20052, USA