Compassion Fatigue and the Elusive Quest for Journalistic Impact

Abstract

Influenced by behavioral research on compassion fatigue, New York Times columnist Nicholas Kristof goes out of his way to find just the right person who illuminates the larger story. Rather than focus on suffering or grim statistics, Kristof seeks to move readers by reporting on people who overcome adversity and offer real solutions. Content analysis and Internet metrics are used to assess whether this kind of reporting engenders reader response. Contrary to expectations, the story-telling techniques had little apparent effect on readers’ online response while story topic and geographic proximity influenced frequency of reader comments, Facebook Likes, and digital sharing.

Keywords

newspaper press performance content analysis

In an age of global cable and Internet communication, more news sources than ever inform us about what is happening in the world, yet mass human tragedy at times receives relatively little attention. Why do people—and the media—become intensely caught up in the story of a girl trapped in a mineshaft yet fail to sustain attention on unending genocide in Sudan, prolonged suffering in the rubble of Haiti’s earthquake, and other wide-scale calamity? One reason is “psychic numbing,” in which people relate to the suffering of one as a tragedy but tune out the loss of thousands as a statistic. Influenced by psychology studies documenting this phenomenon (e.g., Slovic, 2009; Slovic, 2007; Slovic, Zionts, Woods, Goodman, & Jinks, 2013), New York Times columnist Nicholas Kristof (2009) says he turned to experimental research “to understand how I could craft my writing so that it would generate a response rather than a turned page.” For example, Kristof claims he goes to great lengths to find an individual who personifies the larger story that needs to be told. Rather than focus on suffering and grim statistics, Kristof seeks to find people who have triumphed over their privations.

This study appraises how these social psychology principles have been applied and, more importantly, whether the story-telling strategies influence reader response. In a content analysis of a year’s columns, the study measures the extent that Kristof’s reporting focuses on the individual and on victims who overcome adversity. The study also assesses how story personification is counterbalanced with statistical information and mobilizing information. Using a variety of media analytics, the study seeks to gauge the online reader’s digital response. What story approaches prompt readers to pass along Kristof’s columns via email, Twitter, and Facebook? Which columns are most frequently “liked” on Facebook and Google+? When do stories tend to go “viral” as they are linked on blogs and spread across other social media? Taking behavioral research on psychic numbing beyond the experimental laboratory, the study seeks a better understanding of audience reaction to news story-telling formats and to help guide media in coverage of some of the world’s most disturbing and seemingly intractable conditions.

Background and Literature Review

How a story is told can be as important as the information it conveys. Story framing—defined by Entman (2010) as “selecting a few aspects of a perceived reality and connecting them together in a narrative that promotes a particular interpretation”—influences understanding of the information and the response, if any, that follows (p. 391). By depicting the characteristics and dimensions of a public issue, story frames are found to define problems, diagnose causes, make moral judgments, and suggest remedies (Entman, 1993; Scheufele, 2009). The lens used to report humanitarian crises—often distant and hidden from public view—is considered especially important because public knowledge is largely mediated by what is presented by newspapers, television, and the Internet. However, while knowledge is essential, awareness is not necessarily sufficient to engender response (Pruce, 2012). “What is sufficient?” asks Tristan Borer, political scientist and human rights scholar. “The answer is not simple, despite the widely held view that: ‘If only people knew, they would do something.’ There is nothing automatic about the link between information and action” (Borer, 2012a, p. 36).

Scholars have tried to identify why some human catastrophes evoke a strong response and others indifference, and how the media play a role in audience reaction. One explanation for insensitivity to mass suffering is “psychic numbing,” a term coined to characterize the shut-down of feelings experienced by rescue workers dealing with the countless victims of the Hiroshima atomic bombing (see, for example, Kinnick, Krugman, & Cameron, 1996). “Psychic numbing,” defined by BehaveNet as the “reduced emotional responsiveness associated with exposure to traumatic events,” has since been used to explain why caring people go to great lengths to help an individual in dire need but are seemingly indifferent to mass suffering. Both scale and distance have been found to influence human capacity to experience affect—“the most basic form of feeling . . . that something is good or bad”—and to sway people’s judgments, actions, and decisions (Slovic, 2007). As psychologist Paul Slovic (2007) explains, there is considerable evidence that the value placed on saving human lives “may follow the same sort of ‘psychophysical function’ that characterizes our diminished sensitivity to a wide range of perceptual and cognitive entities—brightness, loudness, heaviness, and money—as their underlying magnitude increase” (p. 84). This numbing effect also has been described as “compassion fatigue,” which first was used to portray burnout of service workers but quickly widened to include a jaded public “weary of unrelenting media coverage of human tragedy and ubiquitous fund-raising appeals” (Kinnick et al., 1996, p. 687).

Drawing on communications literature, researchers identified four factors in which mass media particularly contribute to compassion fatigue: emphasis on the sensational, incessant “bad news,” lack of context explaining the underlying crisis, and presentation of problems but not solutions (Kinnick et al., 1996, p. 690). The collapse of compassion may be a protective response because people fear the needs of large groups will be overwhelming (Cameron & Payne, 2011). This paralysis also helps explain society’s reluctance to deal with climate change, famine, poverty, disease, and other global challenges (Slovic, 2009).

Another reason that people tune out reports of calamity is that these events in many instances are geographically or culturally remote. The literature shows that readers are drawn more to local controversy than to reporting on distant, unfamiliar incidents and issues. Newspaper readers retain strong interest in community news (see, for example, Hansen & Hansen, 2011; Heider, McCombs, & Poindexter, 2005; Hollander, 2010), but less than 10% of the U.S. public are “passionate” about international news (American Press Institute, 2014b). According to Pew Research Center surveys tracking public attention to the news, only two foreign events—the Iraq war and the Haiti earthquake—made the public’s list of top 10 news stories 2001-2010; only two other foreign stories (the 2005 tsunami and the war in Afghanistan) made the expanded list of 24 top stories of the millennium’s first decade (Pew Research Center for the People and the Press, 2010). Reflecting a perceived lack of interest in foreign affairs, overseas coverage by mainstream media continues to shrink, declining 53% in U.S. dailies over the past quarter century (Enda, 2011). Moreover, reporters frequently frame foreign affairs in terms of their readers’ national interests and identities, rather than presenting a “cosmopolitan” perspective with a global citizen point of view (Cottle, 2009). In a study of television newscasts and public opinion in 11 countries, researchers found that foreign news coverage was mainly driven by national interest and geographic proximity (Aalberg et al., 2013).

Reader engagement is also driven by the subject of a news event or issue. “Americans are discriminating news consumers whose habits vary by topic,” concludes an American Press Institute report examining U.S. news consumption (Holcomb, Gottfried, & Mitchell, 2013). Politics/government leads the list of news topics that holds the public’s interest, followed by community news, and reports on weather and traffic; at the bottom of the list is foreign news (American Press Institute, 2014b). The public’s news agenda is also shaped by social media, with about half of Facebook and Twitter users getting news on those sites (Holcomb et al., 2013). While bloggers generally took their news cues from mainstream media, hot-button news topics, such as same-sex marriage and abortion, are exceptionally popular on blogs and other new media (Maier, 2010).

Journalistic Strategies

The literature on psychic numbing and compassion fatigue helps explain why media coverage sometimes moves people to action—while other times desensitizes the response to human suffering. How then can the media do better, especially in a digital age of relentless news coverage and information glut? This study focuses on four journalistic strategies—story personification, focus on triumph over adversity, minimal use of statistics, and call for action—deliberately embraced by New York Times columnist Nicholas Kristof as he seeks to engage readers in difficult and remote social concerns. The following section examines the conceptual and empirical underpinning provided in the psychology and communications literature for each of these reporting strategies.

Story personification

Reporters are often encouraged to put a “human face” on a news story to personify tough, complex issues. Providing a strong personal voice is not only a compelling story-telling technique, but communications research suggests it serves as a potent format for influencing people’s beliefs and attitudes. Using a two-step model, scholars have documented that people gain compassion by learning of the perspective of individuals, which in turn leads to compassion toward stigmatized groups (Batson et al., 1997; Oliver, Dillard, Bae, & Tamul, 2012). Narrative stories in particular have been found to encourage empathy toward the group as well as individuals in the story (Oliver, Dillard, Bae, & Tamul, 2012). Personal story-telling has been shown to help deliver health information by promoting character identification, leading to greater shifts in knowledge, attitudes, and behavioral intentions (Murphy, Frank, Chatterjee, & Baezconde-Garbanati, 2013).

In the search for mechanisms to overcome psychic numbing, psychology scholars have also documented the potency of framing a large societal problem through the lens of an individual. Identified as the “singularity effect,” behavioral studies show human preference for helping a single identified victim over a group of victims (e.g., Kogut & Ritov, 2005a, 2005b). For example, Israeli researchers solicited donations for a child requiring life-saving medical treatment and then repeated the process asking for donations to aid eight children needing the treatment. Even though the collective need obviously is greatest for the group of dying children, donations to the single child were nearly twice that given to the group of eight (Kogut & Ritov, 2005a). Conceptually related to psychic numbing, the singularity effect is guided by the psychological principle that increases in magnitude of a stimulus tends to evoke diminishing response (Slovic, 2009).

At what point does the psychic numbing begin? Research suggests that the blurring occurs with as few as two individuals. In a study by Slovic and Vastfjall (2010), people donated less to a pair of needy children in Africa than to either a girl or boy posed singly. People were also found to be much more willing to help a needy child identifiable by name, age, and photo than an anonymous child in need of medical aid (Kogut & Ritov, 2005b). “When it comes to eliciting compassion, the identified individual, with a face and a name, has no peer,” researchers say (Slovic et al., 2013, p. 84). But personalizing a story does not necessarily evoke an empathetic response, especially if the individual’s plight is unsympathetic or seemingly hopeless. What matters, Oliver and her colleagues (2012) report, is how the story is told: “Narrative format does not invariably create either favorable or unfavorable reactions in consumers. Rather, relative to non-narrative format, it amplifies various responses in ways that are consistent with the story line” (p. 217).

Triumph over adversity

One tactic that journalists use against compassion fatigue is to report on success stories, show-casing examples of people overcoming dire circumstances. According to social cognitive theory, such portrayals provide inspiration and model how people against great odds can change their lives for the better. By showing human triumphs over adversity, mass media raise viewers’ sense of efficacy—that they too can make a difference (Bandura, 1997). That is why Save the Children and other charities perennially display a grinning child rather than depressing images of squalor and mass suffering. But hero-victim portrayals also can be a formulaic reporting device that over-simplifies complex challenges and stereotypes victims and relief workers alike, warns communications scholar Susan Moeller, author of the seminal book Compassion Fatigue.

Once the resolution of the individual narrative has been told—the doctor saved the baby, the baby died despite the doctor’s efforts—the public’s attention moves on. And if there is an attempt to retain that attention by retelling and retelling the horror stories, sooner or later—and increasingly it is sooner—the public turns the page, turns away, slipping into a compassion fatigue funk. (Moeller, 1999, p. 28)

Numbed by the numbers

Behavioral research suggests that statistical information can diminish empathy. For example, one such study found that contributions to rescue a starving 7-year-old girl in Africa fell dramatically when statistics disclosed that there were millions of other children facing a similar plight (Small, Loewenstein, & Slovic, 2007). Other studies have also documented “psychophysical numbing” in which lives are less valued as their numbers increase (e.g., Fetherstonhaugh, Slovic, Johnson, & Friedrich, 1997; Friedrich et al., 1999). This paradoxical response—people more readily respond to the aid of single person in need than to mass suffering—offers a cautionary lesson: While statistics can be used to convey the enormity of a crisis, they also may overwhelm and undermine response. “[A]s numbers get larger and larger, we become insensitive; numbers fail to trigger the emotion or feeling necessary to motivate action,” says Slovic (2009, p. 30) in a sobering commentary titled “The More Who Die, the Less We Care.”

Mobilizing information

Studies have documented that the news media influence people’s charitable and emotional responses to humanitarian crises. For example, one study found that just a 700-word story in the New York Times and the Wall Street Journal increased daily donations for 2004 tsunami relief by 18% (Brown & Minty, 2008); another study found charitable giving correlated with media coverage of the 2008 cyclone in Myanmar (Brown & Wong, 2009). People are held to be especially responsive to media reports when given “mobilizing information,” defined as “information which allows people to act on those attitudes which they might already have” (Lemert, Mitzman, Seither, Cook, & Hackett, 1977, p. 721). News attention was the strongest predictor of monetary donations to the victims of the 2010 Haiti earthquake, a factor amplified when news of the disaster spread through social media, and audiences were instructed how to instantly make a donation by texting a keyword on their mobile phones (Heatwole, 2012; Martin, 2013). Examining the impact of mobilizing information more broadly, researchers provided a panel of 755 U.S. adults with news stories that randomly included or excluded a potential response to the social issue described in the article. The results showed that the solutions-based journalism heightened audience interest and desire to share what they read as well as accentuating a sense of efficacy in addressing the issue (Curry & Hammonds, 2014).

Conversely, media have been shown to contribute to compassion fatigue through over-exposure to human suffering without apparent solutions (Kinnick et al., 1996). In a study of 1,600 items in 11 newspapers, mobilizing information that facilitates reader action was found least available where it was considered most needed: controversial political issues and other public affairs reporting (Lemert et al., 1977). The failure by media to provide solutions can lead to a sense of inefficacy “by showing an unending procession of events to which they are expected to react, but which they cannot themselves affect” (Whitehorn, 1989, p. 43). When public attention and concern wither, governments are unlikely to take the strong action needed to address genocide and other calamity (see, for example, Power, 2002).

Nicholas Kristof

Twice awarded the Pulitzer Prize, Kristof has devoted much of his career at the New York Times to covering global disease, poverty, and violence. Sobered by the world’s muted reaction to atrocities in Darfur, Kristof turned to psychology studies and other academic research for clues on how to overcome “psychic numbing” when reporting on mass suffering. Kristof was particularly struck by Slovic’s research showing that the story of one person suffering is more likely to draw an empathetic response than the story of many (e.g., Chong, 2012; Kristof, 2007). “To me, the lessons of this research are twofold,” Kristof says. “First, tell an engaging individual story to suck people in. Second, show that it’s not hopeless, but that progress is possible” (as cited in Chong, 2012). But Kristof stops short when it comes to the behavioral research finding that statistical information stifles emotional response. “If you follow this research, you would leave out the context,” Kristof said in an interview on National Public Radio.

All you would do would be telling individual stories, and that would be one step too far for me. I do want to connect with people and inform them about these larger problems. So my compromise is that I do try to find a story that will resonate with people. (Gladstone, 2009)

Although Kristof is applauded for his perseverance in seeking the untold personal story of overcoming adversity, even some of his admirers suggest that this journalistic approach has become too formulaic, too predictable. For example, Moeller (1999) criticizes the “virtual template” in which complex problems are reduced to a “morality play” of heroes, rescuers, and victims (usually women and children), and solutions are overly simplified and cast through a Western lens of “What does this mean for us?” Other human rights scholars raise concern that Kristof’s “shock” portrayals of starving children and mutilated bodies “exoticize” distant suffering, diminishing the connection between sufferer and the spectator and the sense of close ties that would evoke a humanitarian response. Says Borer (2012b),

There is a fine line between turning the page because of insufficient shock . . . and turning the page because of too much shock (“I can’t bear to look at these pictures, so I must look away”). Either way, the message is tuned out and the page is turned, defeating the behavior-inducing purpose. (p. 157)

Media Metrics

Communication scholars have long relied on content analysis and survey data to gauge agenda-setting effects, media credibility, and other means that media exposure influences audience behavior and attitudes. In recent years, Internet analytics have introduced new ways of tracking audience response, providing real-time data down to the individual story. One of the first uses of digital metrics to measure readership was to count “hits”—that is, the number of visitors to a story posted on a digital website (Gerlitz & Helmond, 2013). Other metrics followed seeking to measure how audiences engage with news stories, from measuring the number of comments or “likes” a story elicits to the extent that readers share the story via Twitter, Facebook, and other social media (see, for example, Fischer, 2014; Krall, 2009; Thiruvengadam, 2013). The utility of these metrics is that they not only indicate which stories readers choose to view but they also provide a rich array of measures tracking pro-active responses as online readers digitally discuss, endorse, and post stories that catch their attention. Analytic metrics help guide news organizations because they allow standardized comparisons of reader response across stories, across time, and product investment (Stray, 2012). As noted by Aron Pilhofer (2014), who headed the new technology team at the New York Times,

We are awash in metrics, and we have the ability to engage with readers at scale in ways that would have been impossible (or impossibly expensive) in an analog world. The problem now is figuring out which data to pay attention to and which to ignore.

Digital tools of engagement potentially amplify a story’s impact (Kessler, 2012). When a story is presented in such a way as to evoke an empathetic response, all a reader has to do is click “Like” on Facebook or briefly comment to engage in a modest form of collective action. Kessler (2012) and other new media experts contend that these digital responses should not be dismissed as “feel-good clicking,” noting that social media enable readers to easily align themselves with a distant cause, to cross the line from private to public discourse, and to share their feelings across “weak ties” (pp. 205-207). The online response represents “the first step on the ladder of engagement,” particularly important when dealing with human rights disasters where the suffering is often far removed from a reader’s daily life (Borer, 2012, p. 31; Kessler, 2012).

Hypotheses and Research Questions

In summary, this study draws on experimental psychology research that suggests the “story of one” is more likely to draw an empathetic response than the story of mass suffering. Research also indicates that statistics and portrayal of overwhelming need tend to diminish empathy, while mobilizing information providing corrective action encourages personal engagement. The value of using Kristof’s work as a case study is several-fold: Kristof is arguably the pre-eminent U.S. journalist covering genocide and other mass atrocity, he explicitly draws on social psychology research to guide his reporting, and his twice-weekly columns offer consistency in format, placement, timing, and length, providing a “level playing field” for this research. Moreover, reader online response to each of his columns can be independently monitored by a variety of Internet metrics that publicly track Kristof’s work.

This investigation opens with a series of research questions examining the threshold issue of whether Kristof accomplished his stated intentions to focus on the individual and to apply other specified social psychology principles to his reporting.

RQ1: To what extent do the columns by New York Times reporter Nicholas Kristof focus on one or two individuals who depict a larger story of social significance?

RQ2: To what extent do Kristof’s columns tell stories of people overcoming adversity?

RQ3: To what extent do Kristof’s columns offer mobilizing information?

RQ4: To what extent do Kristof’s columns offer quantitative information?

Drawing from social psychology research and the literature on overcoming psychic numbing and compassion fatigue, the following hypotheses are posed:

H1: Reader response, as tracked by online media analytics, positively corresponds with the extent to which Kristof’s reporting focuses on one or two individuals.

H2: Reader response positively corresponds with stories focused on personal triumph.

H3: Reader response positively corresponds with stories offering mobilizing information.

H4: Reader response negatively corresponds with the extent to which Kristof’s reporting focuses on statistics.

The literature also suggests that the U.S. public is drawn more to media coverage of local controversy than foreign affairs.

H5: Reader response positively corresponds with geographic proximity.

Examining the data broadly, the study assesses the relationship of reader response with story topic.

RQ5: What story topics engender strong reader response?

Method

Data Collection

Nicholas Kristof writes a column twice weekly (Thursday and Sunday) in the New York Times. Each column was collected over a 13-month period ending in September, 2012, yielding a constructed year-long sample of 104 columns (the extra month of data collection was needed to make up for days that Kristof did not publish because of vacation or other circumstances). This comprehensive approach to story collection was chosen over random sampling of archival columns because the study required real-time examination of Internet analytics tracking reader response. For each column studied, reader response was recorded over a 3-day period—the day of publication and the 2 days following.¹ The resulting database provided a broad array of story topics as well as more than 1,400 individual measurements of online story response. Table 1 provides a listing of the top 25 columns as ranked by number of reader comments.

Table 1.

Top 25 Kristof Columns, Ranked by Number of Reader Comments.

Headline	Topic	Comments	Facebook likes	Google citations	Focus on one
Beyond Pelvic Politics	Birth control	632	2,203	136	11%
It Takes One to Know One	Election	501	2,912	90	39%
When States Abuse Women	Birth control	494	1,924	173	21%
Value of Teachers	Education	478	2,243	271	18%
Crony Capitalism	Economy	424	511	458	20%
Students Over Unions	Education	378	881	166	11%
We Are all Nuns	Humanitarianism	370	3,484	377	21%
American’s Primal Scream	Economy	354	953	188	10%
Obama’s First Term Report Card	Election	337	1,055	155	83%
Politics, Odor, and Soap	Politics	332	676	137	30%
The Bankers and the Revolutionaries	Economy	318	449	361	19%
Is an Egg for Breakfast	Agriculture	315	1,468	230	26%
Is Israel Its Own Worst Enemy?	Middle East	315	350	249	38%
The White Underclass	Inequality	312	1,245	140	29%
Where Are the Romney Republicans?	Election	298	424	86	19%
Markets and Morals	Inequality	295		135	14%
A Veteran’s Death, the Nation’s Shame	Military United States	283	1,616	199	50%
The Winning Essays	Bullying	281	693	131	43%
Why Is Europe a Dirty Word?	Election	269	1,394	118	17%
Is Banking Bad?	Economy	268	690	110	28%
Obama AWOL in Syria	Arab spring	265	448	164	30%
Born not to Be Bullied	Bullying	260	1,784	129	94%
President as Piñata	Election	243	2,567	221	75%
Scaring the Voters in the Middle	Election	241	2,203	57	26%
The Secret Weapon: All of Us	Economy	239	2,211	93	30%
A Banker Speaks With Regret	Economy	165	1,183	260	60%

Coding

In a content analysis, each Kristof column was coded to assess the role of human sources, kinds of information provided, and the topic, scope, and length of the article. Coders identified each person (up to four) who was named or specifically described (e.g., 13-year-old rape victim) in the column. Excluded were incidental references to persons mentioned only in passing in the story. Coders designated each person by source role (e.g., victim, official, observer), by gender, and by child/adult classification (persons 17 and younger were considered children). Personification was calculated by the number of paragraphs that name, describe, quote, or otherwise cite a person in a story divided by the story’s total number of paragraphs. For this category, personification included both positive and negative portrayals as the intent is to assess overall reader response to the personal story, a refined examination distinguishing between triumph and tragedy follows.

Coders also determined the frequency, measured by number of paragraphs, that the column offered quantitative/statistical information (i.e., count of casualties, percentages, rates, financial figures). Excluded were quantitative references to ages (i.e., 87-year-old victim), incidental counting (e.g., 20-mile hike, three police encounters), or incidental costs (e.g., US$7 dress, 46 cent postage stamp). Coders identified columns that provided mobilizing information, which was defined as facts and recommendations specifically advising readers what should be done to address a situation described in the story. For example, mobilizing information would include a call to support “safe harbor” legislation that treats prostitute teenage girls as victims rather than criminals; excluded would be a generalized plea to take climate change more seriously. In addition, coders identified columns that reported “triumph over adversity” in people, event, or situation described.² For example, stories of personal triumph included a Kenyan woman who escaped prostitution to start a micro-financed sewing business (Kristof, 2011b), a library program that made a profound difference in an impoverished Vietnamese girl’s education (Kristof, 2011d), and the story of “Mrs. Grady” and how she and other great teachers have transformed lives (Kristof, 2012a). Finally, each column also was assigned one or more descriptors to provide an accounting of story topics (e.g., the economy, sexual abuse, presidential election, reproductive rights, “Arab spring”). The story descriptors, emergent from the news covered in the year studied, were reviewed on completion of the coding process and in some instances refined or combined. From this list, story topics considered especially emotionally or politically charged (i.e., abortion, Israel/Palestine conflict) were designated ad hoc as “hot-button” issues.

Intercoder reliability

A six-page codebook was developed to provide standard measures and protocols for this study. Following coder training and refinement of procedures, an intercoder reliability test was conducted prior to the start of the coding process. Simple agreement across all coding variables was 88% or higher,³ except for two variables, “mobilizing information” and “big picture/context.” On retraining, 100% agreement was achieved for “mobilizing information” by clarifying that the category required specific guidance beyond a general call for action. But coding for “big picture/context” remained problematic (agreement generally below 80%) due to the ambiguous nature of the category; the variable was dropped from this study.

Reader response

A broad array of Internet analytics were selected to gauge readers’ online response to Kristof’s columns, from Facebook “Likes” to blog references on Google, from the New York Times accounting of its top stories to Technorati’s algorithm quantifying a story’s standing in the blogosphere. The metrics for this study were chosen because they provide unobtrusive measures that independently tracked online reader response without requiring the use of the newspaper’s internal analytics or otherwise potentially influencing results by alerting Kristof that his work was under study. Recognizing that each analytic offers a different perspective of online behavior, the study examines the metrics individually as well as collectively to assess multi-faceted reader response.

Metrics for this study are as follows:

New York Times “Most Popular”: The New York Times ranks its top 20 stories by most viewed (articles most frequently viewed by its online readers), most emailed (articles most frequently emailed by its online readers), and most blogged (articles most frequently linked to by bloggers as determined by Blogrunner.com). Ranks were collected for three time periods (last 24 hours, last 3 days, last 30 days), providing longitudinal data for each category as well as the combined measure of reader response referred to as New York Times Top 20.⁴ Approximately, 80% of Kristof’s columns made at least one of the New York Times Top 20 lists. In addition, coders recorded the number of online reader comments that each column tallied over a 3-day period. Over the year-long study, Kristof’s columns drew 17,742 reader comments; the mean was 172 comments per column.

Google: On the day of publication and 2 following days, the headline of the Kristof story was entered into Google’s search engine to track Internet traffic that the story generates. Kristof’s columns generated, on average, 146 Google results over a 24-hr time period. Using Google’s blog search engine, the process was repeated.

Facebook: Kristof posts his columns to his professional Facebook page, providing access to his half-million subscribers. On average, a Kristof column drew 1,046 Facebook “Likes,” 201 comments, and 416 shares.

Google+: Similar to Facebook, Google+ tracks the number of comments, Likes (signified by the plus sign), and frequency that Kristof’s post is shared. Kristof’s following on Google+ was relatively sparse, with each column attracting, on average, 48 “Likes,” 21 comments, and 25 shares.

Topsy Labs: Offering a search system similar to what Google provides for the Web, Topsy has indexed every Twitter message since the first tweet posted in 2006 (Topsy, 2013).⁵ Cited by the New York Times as the best search engine to monitor Twitter traffic (Boutin, 2011), Topsy was used to track how many tweets and other social media posts Kristof’s columns generated. A Kristof column generated, on average, 1,292 social media hits.

Technorati: Measuring digital influence and standing, Technorati calculates the “Authority” and the rank that a blog holds in the blogosphere. Technorati Authority is computed according to linking behavior indicating how often a site generates content referenced by other Internet sites.⁶ On a scale of 1 to 1,000 (the highest “authority”), the mean Technorati authority score for Kristof’s blog was 495; the mean rank was 2,962.

Results

The first stage of this research examines the threshold issue of whether Kristof accomplished his stated intentions to focus on the individual and to apply other social psychology principles to his reporting.

RQ1: Personification

A preponderance of Kristof’s stories focused on one or two individuals who represent a larger story. Although considerable variation was found, Kristof devoted on average more than half of each column to telling the story of two individuals; more than a third of each column featured, on average, just one person’s story. About one in five of Kristof’s columns was so highly personified that the story focused almost exclusively (90% or more of the column) on one or two individuals. Kristof turned to personal stories most frequently when reporting on sexual violence and other human rights violations. The most commonly used sources for Kristof’s stories were ordinary people—victims and witnesses (often women and children) who personally experienced or saw the events described in the story. Victims and witnesses represented 30.5% of Kristof’s sources, while officials represented 23.8%, and experts/observers represented 19.0% of sources in Kristof’s columns. Kristof himself also was a frequent character with first-person commentary accounting for 11.1% of his columns.

RQ2: Overcoming Adversity

Kristof clearly went out of his way to tell stories of triumph over privation. Even though Kristof reported on some of the world’s grimmest circumstances, more than a third of his stories (35.6%) included examples of people overcoming extreme adversity. These included a story about a former prostitute’s efforts to rescue young girls from a Cambodian brothel (Kristof, 2011e), a story about a Florida man who braves bombs to document atrocities in Sudan’s Nuba Mountains (Kristof, 2011a), and a story about a young Sierra Leone civil war victim who survived to become a symbol of the conflict’s depravity (Kristof, 2012b).

RQ3: Mobilizing Information

Kristof’s columns also commonly took on an advocacy role, offering in 43.3% of his columns a “call for action” with specific advice on what can and should be done to ameliorate the conditions he described.

RQ4: Statistics

Kristof injected (often in small amounts) quantitative information in nearly all of his columns. Only 13.5% of his columns lacked statistical data. Nearly half of the columns provided three or more paragraphs with quantitative information.

H1: Focus on the individual

The first hypothesis predicts that reader online response corresponds with how strongly a story is personified, that is, how much of it is focused on one or two individuals. By a number of measures, the hypothesis is not supported (refer to Table 2 for key correlations).

Table 2.

Assessing Reader Response: Pearson Product–Moment Correlations (Select Examples).

	Focus on one person	Focus on two people	Use of statistics
New York Times Top 20	−.140	−.199*	.162
New York Times Comments	−.218*	−.273**	.139
Facebook Likes	−.088	−.125	.192
Facebook Comments	−.100	−.132	.044
Facebook Shares	−.187	−.189	.133
Google Cites	−.213*	−.247*	.022
Google+ Likes	−.062	−.041	.193
Google+ Shares	−.082	−.085	.197*
Google+ Comments	−.180	−.167	.042
Topsy Cites	−.055	−.024	.152
Technorati Authority	−.107	−.128	−.018

Correlation is significant at the .05 level (two-tailed).

Correlation is significant at the .01 level (two-tailed).

New York Times Top 20

Under the heading “Most Popular,” the New York Times posts the relative rankings of the top 20 online news stories most viewed, emailed, or blogged. The Pearson product–moment correlation was weak and not significant between each of these “popularity” measures and the extent to which a column centered on either one or two individuals. Only the relationship between a collective measure of the New York Times Top 20 lists and a focus on two individuals yielded a significant though weak correlation, r(102) = −.199, p = .044, but the correspondence was in the opposite direction predicted.

The New York Times also tracks (though does not rank) the number of comments that an article draws on its website. By this measure, the correlation between column focus on one individual and frequency of comments was significant but weak, and in the opposite direction predicted, r(102) = −.218, p = .028. Focus on two individuals yielded not only a somewhat stronger correlation, r(102) = −.273, p = .006, but also a negative correspondence.

Social networking

Kristof routinely posts his column on his Facebook page and Google+, two social networking sites that track the number of subscribers who indicated they “like” the post, “share” the post with their own followers, or comment on the post. None of these measures of reader engagement correlated significantly with the proportion of Kristof’s columns focused on either one or two individuals. Moreover, the relationships were all in the opposite direction predicted.

Blogosphere

Several metrics were used to indicate the extent to which Kristof’s columns went “viral,” that is, to circulate rapidly on the Internet. The search engine Topsy was used to track Twitter and Internet traffic citing Kristof’s columns, Google provided a gauge of how often Kristof’s column was referenced on the Web or posted on a blog, and Technorati Authority measured the standing and influence of Kristof’s blog on the blogosphere. Only one of these metrics, Google’s tracking of Internet references to Kristof’s columns, yielded a significant correlation with focus on an individual, r(102) = −.213, p = .031, and on two individuals, r(102) = −.247, p = .012. But once again, the correlations were weak and consistently in the opposite direction predicted.

H2: Overcoming adversity

Contrary to expectations, stories of personal triumph did not engender reader online response. In fact, Kristof columns without the element of personal triumph elicited significantly more reader comments in the New York Times, t(104) = 2.59, p = .011; on Facebook, t(104) = 2.06, p = .042; and on Google+, t(104) = 2.65, p = .009, than stories focused on overcoming adversity. Another metric, the New York Times tracking of blog postings, indicated that bloggers were less likely to refer to the column when Kristof focused on a victim overcoming adversity, t(70.5) = 2.98, p = .027. None of the other metrics of reader digital response indicated significant differences between the two story types.

H3: Mobilizing information

By nearly every metric, columns that included a call for action elicited higher levels of reader online response than columns devoid of advice. However, independent-samples t tests indicate that none of the differences are statistically significant. The hypothesis is not supported.

H4: Statistics

Drawing from experimental research indicating that quantitative information diminishes civic response, the fourth hypothesis predicts that reader response negatively corresponds with Kristof’s focus on statistics. Contrary to the hypothesis, Google+ users were somewhat more likely to share Kristof’s columns when an abundance of statistical data was provided, r(102) = .197, p = .048. No other significant correlations were found between Kristof’s use of statistics and propensity of readers to comment, “like,” email, or otherwise share his columns. However, the correlations were uniformly positive (opposite to the direction generally found in the preceding analyses).

The analysis above examines whether numbers and statistics of any kind had a numbing effect. But how do readers respond when given statistical information that explicitly expresses the magnitude of the issue or event described? Nearly 60% of Kristof’s columns provided specific quantitative information regarding lives at stake (i.e., number of victims killed, deported, hospitalized), longevity (i.e., years of conflict), and cost (i.e., financial losses, governmental expenditure). When comparing whether or not the story provided statistics indicating magnitude of the issue or event, the apparent effect on reader response comes into somewhat sharper focus. Independent-samples t tests indicated that when given statistics on the magnitude of a problem, readers more frequently shared the column on Facebook, t(95.6) = −2.76, p = .007; referred to the column on Twitter or other social media, t(102) = −2.32, p = .023; viewed the story online, t(97.1) = −1.99, p = .049; clicked on “like” on Google+, t(84.6) = −2.35, p = .021; shared on Google+, t(75.4) = −3.60, p = .001; and commented on Google+, t(74.0) = −2.21, p = .030. By every other measure, reader response was also greater when the story quantified the issue than when magnitude was left out, though the differences were not statistically significant.

H5: Proximity

Independent-samples t tests were conducted to assess reader response for stories reported in or focused on the United States or Canada versus stories focused on locales outside North America. As predicted, North American stories by nearly every metric engendered greater reader online response than foreign stories; the differences are statistically significant for all digital measures except Google+ and Google search. The hypothesis is supported.

RQ5: Story topic

Reader online response to topical issues of high public interest (i.e., abortion, Israel, presidential election) was almost uniformly strong. Highly emotionally or politically charged issues evoked markedly more views, more comments, more “Likes,” and more blog posts than less sensational stories. Independent-samples t tests indicate the differences are significant for all metrics except Technorati’s measure of blogosphere “authority” and rank. While one would expect hot-button issues to elicit strong reader response, these results offer a sharp contrast with the generally negative and weak relationships reported above.

The findings suggest that reader response often has more to do with story topic and locale than how strongly a story is personified or how it incorporates mobilizing or statistical information. But what happens if the data are controlled for the effects of story type and story location? The answer is not much. For example, a multiple regression analysis indicated that story topic and locale accounted for a significant amount of the variability in the volume of online comments that Kristof columns generated, R² = .453, F(2, 99) = 40.96, p ≤ .001. A second analysis was conducted to evaluate whether story personification affects reader comment over and above story topic and locale. Even when controlling for story topic and locale, story personification had little apparent effect on volume of reader comments, R² change = .048, F(3, 98) = 32.74, p = .003. Only one other measure, Google search, indicated that story personification significantly affects reader response above and beyond story topic and locale, R² change = .05, F(3, 99) = 5.159, p = .019.

Discussion

The study shows that Kristof “practices what he preaches,” applying the social psychology principles he publicly espouses. His columns routinely apply a human face (or two) to tough and distant social issues. Kristof’s stories widely include elements of triumph over adversity as well as mobilizing information, providing prescriptions for positive change even when dealing with some of the world’s most difficult challenges. But this does not mean that Kristof does—or should—unwaveringly adhere to these practices. Depending on the story to be told, his column at times concentrates more on broad issues than individual endeavors. Kristof also eschews experimental research that suggests statistics stifle emotional response. His focus on individual stories is frequently counterbalanced with quantitative information.

While Kristof clearly uses reporting methods intended to overcome compassion fatigue, the effect of these efforts for the most part is not apparent. None of the metrics studied showed a significant positive relationship with personification of story, triumph over adversity, or mobilizing information. The interplay of reader response with statistics also contradicted what experimental psychology posits; this study found that reader response positively corresponds with quantitative information, though the results generally were not statistically conclusive.

Proximity and story topic were found to be stronger predictors of reader response than the social psychology elements evaluated by this study. It is no surprise that Kristof’s column on Lady Gaga’s campaign against bullying elicits greater response than a column about a young woman who had escaped from a Cambodian brothel. But the results underscore why the media gravitate to superstars and sensational topics while remote, difficult distant issues such as human trafficking get relatively little notice. Seeking to overcome these challenges, Kristof purposely turns to individual narratives when the story line is least likely to have popular appeal. Kristof explained in an interview with the Poynter Institute:

The issues I care most passionately about, from Sudan to sex trafficking, aren’t ever going to do well on the most emailed list, because they’re off the agenda. That’s precisely the reason I’m writing about them, trying to get people to care more about neglected issues. (As cited in Beujon, 2012)

This also may be one reason that the relationship between personification and reader response is statistically inconclusive and often in the negative direction predicted.

Further analysis was conducted to assess reader response among stories that focused on violence and other human rights abuses. Even within this select sub-sample, story personification did not significantly affect response from readers. However, it is noteworthy that stories offering mobilizing information to end the abuse elicited more than twice as many reader comments as stories without a call for action, a significant difference in response.

It is disappointing that the “power of one” story-telling so vividly identified in experimental research did not yield similar audience reaction by Kristof’s real-life reporting. One wishes there were a simple story formula in which the focus on the individual would trigger a direct emotional response, as it did in the experimental conditions posed by the social psychologists. In this regard, this study follows a long line of communication research showing there is no “magic bullet” or “hypodermic needle” in which the media’s message activates a powerful uniform response (see, for example, Lowery & DeFleur, 1995, pp. 13-14). The results also give weight to critics who warn that repetitive use of victim-hero archetypes and over-exposure of graphic accounts of suffering may numb audience response. Whether explicit portrayals of suffering motivates or desensitizes response deserves further scrutiny. As Borer (2012b) contends, “In a world that is ever more rife with suffering, sorting out the answers to this debate is more important than ever and deserves serious and sustained analysis” (p. 173).

This is not to say that Kristof’s approach to overcoming compassion fatigue should be abandoned. Despite Kristof’s penchant for reporting on remote and grisly topics, his columns routinely make the New York Times’ Top 20 “most popular” lists. Kristof’s half-million subscribers on Facebook is another indication that his reporting elicits broad reader interest. Several charities reported that his column generated more than US$100,000 in contributions when Kristof wrote about their organizations (see, for example, Kristof, 2011c). Clearly, Kristof has built a large, engaged following as he reports on the disenfranchised around the world.

Kristof also may be on the right path when he insists on providing quantitative information. Unlike the findings of experimental research, this study revealed no evidence that statistics stifled audience response. When presented statistics indicating the magnitude of an issue or event, reader response was significantly stronger by a number of measures, including frequency of online comments, blog linkages, and Google+ likes and shares. These results should be viewed with caution; needed is further study examining how different ways of conveying statistics and other contextual information might mitigate rather than accentuate psychic numbing.

It is possible that Kristof’s unique readership tended to mute the impact of the columnist’s narrative emphasis. As engaged readers of the New York Times, those most likely to comment or otherwise respond may already be well attuned to the issues raised by Kristof’s columns. If so, these informed readers may react more to hard information than a narrative story of an individual’s plight. Supporting this supposition is the finding that the strongest reader response to human rights stories occurred when columns provided information that specifically advised readers what should be done.

The results also may reflect the limitations of the online media metrics used by this study. To maintain independence, this investigation relied on digital reader-response measures that were publicly available. Future research could benefit by gaining access to proprietary analytics used internally by news organizations. Moreover, media metrics reveal a limited view of reader response. As New York Times technologist Pilhofer (2014) cautions, “What we do not have are ways of measuring how a piece of journalism changes the way people think or act. We don’t have a metric for impact” (also see Stray, 2012).

It is hoped that this line of research encourages others to evaluate how these principles of social psychology apply more widely to news reporting on genocide and other mass suffering. Research that goes beyond that of a single columnist or newspaper is needed. While media analytics open promising new avenues for communication research, tracking online response is an inadequate surrogate for gauging emotional engagement. Other kinds of measures, such as psychological testing of emotional affect, could perform as better barometers of empathetic response. Qualitative analysis of reader comments also would likely reveal contextual insight missed by numerical accounting. The answers that communication scholars provide could help garner more effective news coverage and elicit a stronger civic response to human suffering.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biography

Scott R. Maier is associate professor in the School of Journalism and Communication at the University of Oregon.

References

Aalberg

Papathanassopoulos

Curran

Hayashi

Iyengar

Jones

P. K.

. . . Tiffen

(2013). International TV news, foreign affairs interest and public knowledge: A comparative study of foreign news and public opinion in 11 countries. Journalism Studies, 14, 387-406. doi:10.1080/1461670X.2013.765636

American Press Institute. (2014a, March 17). The personal news cycle: How Americans choose to get their news. Retrieved July 27, 2014, from http://www.americanpressinstitute.org/publications/reports/survey-research/personal-news-cycle/

American Press Institute. (2014b, March 17). The rational and attentive news consumer. Retrieved April 8, 2014, from http://www.americanpressinstitute.org/publications/reports/survey-research/rational-attentive-news-consumer/

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.

Batson

C. D.

Polycarpou

M. P.

Harmon-Jones

Imhoff

H. J.

Mitchener

E. C.

Bednar

L. L.

. . . Highberger

(1997). Empathy and attitudes: Can feeling for a member of a stigmatized group improve feelings toward the group? Journal of Personality and Social Psychology, 72, 105-118. doi:10.1177/014616702237647

Beujon

(2012, September 25). Kristof: “The U.S. is losing interest” in foreign reporting. Poynter Online. Retrieved April 11, 2014, from http://www.poynter.org/latest-news/mediawire/189532/kristof-the-u-s-is-losing-interest-in-foreign-reporting/

Borer

T. A.

(2012a). Introduction: Willful ignorance—news production, audience reception and responses to suffering. In Borer

T. A.

(Ed.), Media, mobilization, and human rights: Mediating suffering (pp. 1-41). New York, NY: Zed Books.

Borer

T. A.

(2012b). “Fresh wet tears”: Shock media and human rights awareness campaigns. In Borer

T. A.

(Ed.), Media, mobilization, and human rights: Mediating suffering (pp. 143-180). New York, NY: Zed Books.

Boutin

(2011, July 26). A better way to search Twitter. The New York Times. Retrieved October 13, 2013, from http://gadgetwise.blogs.nytimes.com/2011/07/26/a-better-way-to-search-twitter/

10.

Brown

P. H.

Minty

J. H.

(2008). Media coverage and charitable giving after the 2004 Tsunami. Southern Economic Journal, 75, 9-25.

11.

Brown

P. H.

Wong

P. Y.

(2009). Type of news coverage and donations to disaster relief: Evidence from the 2008 cyclone in Myanmar (Working paper). Retrieved October 19, 2013, from http://ssrn.com/abstract=1489909

12.

Cameron

C. D.

Payne

B. K.

(2011). Escaping affect: How motivated emotion regulation creates insensitivity to mass suffering. Journal of Personality and Social Psychology, 100, 1-15. doi:10.1037/a0021643

13.

Chong

(2012, November 26). How Nicholas Kristof uses his pulpit to engage people with empathy. Co.Exist. Retrieved October 15, 2013, from http://www.fastcoexist.com/1680966/how-nicholas-kristof-uses-his-pulpit-to-engage-people-with-empathy

14.

Cottle

(2009). Global crises in the news: Staging new wars, disasters, and climate change. International Journal of Communication, 3, 494-516. Retrieved November 14, 2014, from http://ijoc.org/index.php/ijoc/article/view/473

15.

Curry

A. L.

Hammonds

K. H.

(2014, June 2). The power of solutions journalism. Solutions Journalism Network. Retrieved July 30, 2014, from http://solutionsjournalism.org/wp-content/uploads/2014/06/ENP_SJN-report.pdf

16.

Doctor

(2013, November 21). The newsonomics of The New York Times’ Paywalls 2.0. Nieman Journalism Lab. Retrieved April 2, 2014, from http://www.niemanlab.org/2013/11/the-newsonomics-of-the-new-york-times-paywalls-2-0/

17.

Enda

(2011, December/January). Retreating from the World. American Journalism Review. Retrieved October 17, 2013, from http://www.ajr.org/article.asp?id=4985

18.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58. doi:10.1111/j.1460-2466.1993.tb01304.x

19.

Entman

R. M.

(2010). Media framing biases and political power: Explaining slant in news of Campaign 2008. Journalism, 11, 389-408. doi:10.1177/1464884910367587

20.

Fetherstonhaugh

Slovic

Johnson

S. M.

Friedrich

(1997). Insensitivity to the value of human life: A study of psychophysical numbing. Journal of Risk and Uncertainty, 14, 283-300. doi:10.1023/A:1007744326393

21.

Fischer

M. C.

(2014, March 18). The pageview is passé: New metrics emerge to measure audiences. American Journalism Review. Retrieved March 24, 2014, from http://ajr.org/2014/03/18/pageview-passe-new-metrics-emerge-measure-audiences/

22.

Friedrich

Barnes

Chapin

Dawson

Garst

Kerr

(1999). Psychophysical numbing: When lives are valued less as the lives at risk increase. Journal of Consumer Psychology, 8, 277-299. doi:10.1207/s15327663jcp0803_05

23.

Gerlitz

Helmond

(2013). The like economy: Social buttons and the data-intensive web. New Media & Society, 15, 1348-1365. doi:10.1177/1461444812472322

24.

Gladstone

(2009, December 11). [Transcript]. On the Media (broadcast on National Public Radio). Retrieved October 17, 2013, from http://www.onthemedia.org/2009/dec/11/follow-for-now/transcript/

25.

Hansen

E. H.

Hansen

G. L.

(2011). Newspaper improves reader satisfaction by refocusing on local issues. Newspaper Research Journal, 32, 98-106.

26.

Heatwole

(2012, January 12). Pew studies the power of text-based donations after Haiti Quake. IdeaLab. Retrieved April 8, 2014, from http://www.pbs.org/idealab/2012/01/pew-studies-the-power-of-text-based-donations-after-haiti-quake011/

27.

Heider

McCombs

Poindexter

(2005). What the public expects of local news: Views on public and traditional journalism. Journalism and Mass Communication Quarterly, 82, 952-967. doi:10.1177/107769900508200412

28.

Holcomb

Gottfried

Mitchell

(2013, November 14). News use across social platforms. Pew Research Journalism Project. Retrieved August 6, 2014, from http://www.journalism.org/2013/11/14/news-use-across-social-media-platforms

29.

Hollander

(2010). Local government news drives print readership. Newspaper Research Journal, 31(4), 6-16.

30.

Kessler

(2012). Amplifying individual impact: Social media’s emerging role in activism. In Borer

T. A.

(Ed.), Media, mobilization, and human rights: Mediating suffering (pp. 205-215). New York, NY: Zed Books.

31.

Kinnick

K. N.

Krugman

D. M.

Cameron

G. T.

(1996). Compassion fatigue: Communication and burnout toward social problems. Journalism and Mass Communication Quarterly, 73, 687-706. doi:10.1177/107769909607300314

32.

Kogut

Ritov

(2005a). The singularity effect of identified victims in separate and joint evaluations. Organizational Behavior and Human Decision Processes, 97, 106-116. doi:10.1016/j.obhdp.2005.02.003

33.

Kogut

Ritov

(2005b). The “identified victim” effect: An identified group or just a single individual? Journal of Behavioral Decision Making, 18(3), 157-167. doi:10.1002/bdm.492

34.

Krall

(2009). Using social metrics to evaluate the impact of online healthcare communications. Journal of Communication in Healthcare, 2, 387-394.

35.

Kristof

(2007, May 10). Save the Darfur Puppy. The New York Times, p. A33.

36.

Kristof

(2009, November 30). Nicholas Kristof’s advice for saving the world. Outsideonline. Retrieved October 17, 2013, from http://www.outsideonline.com/outdoor-adventure/Nicholas-Kristof-s-Advice-for-Saving-the-World.html

37.

Kristof

(2011a, February 26). Battling Sudan’s bombs with videos. The New York Times, p. SR11.

38.

Kristof

(2011b, September 15). Sewing her way out of poverty. The New York Times, p. A35.

39.

Kristof

(2011c, September 24). Helping Somali famine victims [On the Ground, a New York Times blog]. Retrieved from http://kristof.blogs.nytimes.com/2011/09/24/helping-somali-famine-victims

40.

Kristof

(2011d, November 6). His libraries, 120,000 so far, change lives. The New York Times, p. SR11.

41.

Kristof

(2011e, November 13). Fighting back, one brothel raid at a time. The New York Times, p. SR11.

42.

Kristof

(2012a, January 22). How Mrs. Grady transformed Olly Neal. The New York Times, p. SR13.

43.

Kristof

(2012b, September 25). From tragedy to triumph. The New York Times, p. SR11.

44.

Lemert

J. B.

Mitzman

B. N.

Seither

M. A.

Cook

R. H.

Hackett

(1997). Journalists and mobilizing information. Journalism Quarterly, 54, 721-726.

45.

Lowery

S. A.

DeFleur

M. L.

(1995). Milestones in mass communication research (3rd ed.). White Plains, NY: Longman.

46.

Maier

S. R.

(2010). All the news fit to post? Comparing news content on the web to newspapers, television and the web. Journalism and Mass Communication Quarterly, 87, 548-562. doi:10.1177/107769901008700307

47.

Martin

J. A.

(2013). Closing gaps in international knowledge and participation: News attention, online expression, and the 2010 Haiti Earthquake. Mass Communication and Society, 16, 417-440. doi:10.1080/15205436.2012.713149

48.

Moeller

(1999, December). Four habits of international news reporting. Retrieved April 11, 2014, from http://www.frameworksinstitute.org/assets/files/PDF_GII/four_habits_of_news_reporting.pdf

49.

Murphy

S. T.

Frank

L. B.

Chatterjee

J. S.

Baezconde-Garbanati

(2013). Narrative versus nonnarrative: The role of identification, transportation, and emotion in reducing health disparities. Journal of Communication, 63, 116-137. doi:10.1111/jcom.12007

50.

Oliver

M. B.

Dillard

J. P.

Bae

Tamul

D. J.

(2012). The effect of narrative on empathy for stigmatized groups. Journalism and Mass Communication Quarterly, 89, 205-224. doi:10.1177/1077699012439020

51.

Pew Research Center for the People and the Press. (2010, December 30). Public’s top stories of the decade—9/11 and Katrina. Retrieved July 29, 2014, from http://www.people-press.org/2010/12/30/publics-top-stories-of-the-decade-911-and-katrina/

52.

Pilhofer

(2014, July 25). Finding the right metric for news. aronpilhofer.com. Retrieved March 24, 2014, from http://aronpilhofer.com/post/27993980039/the-right-metric-for-news

53.

Potter

W. J.

Levine-Donnerstein

(1999). Rethinking validity and reliability in content analysis. Journal of Applied Communication Research, 27, 258-284. doi:10.1080/00909889909365539

54.

Power

(2002). A problem from hell: America and the age of genocide. New York, NY: Harper.

55.

Pruce

J. R.

(2012). The spectacle of suffering and humanitarian intervention in Somalia. In Borer

T. A.

(Ed.), Media, mobilization, and human rights: Mediating suffering (pp. 216-239). New York, NY: Zed Books.

56.

Scheufele

(2009). Framing as theory of media effects. Journal of Communication, 43(4), 103-122. doi:10.1111/j.1460-2466.1999.tb02784.x

57.

Slovic

(2007). If I look at the mass I shall never act: Psychic numbing and genocide. Judgment and Decision Making, 2, 79-95. Retrieved from http://ssrn.com/abstract=1569098

58.

Slovic

(2009). The more who die, the less we care. In Michel-Kerjan

Slovic

(Eds.), The irrational economist: Decision making in a dangerous world (pp. 30-40). New York, NY: Public Affairs Press.

59.

Slovic

Vastfjall

(2010). Affect, moral intuition and risk. Psychological Inquiry, 21, 387-398. doi:10.1080/1047840X.2010.521119

60.

Slovic

Zionts

Woods

A. K.

Goodman

Jinks

(2013). Psychic numbing and mass atrocity. In Shafir

(Ed.), The behavioral foundations of policy (pp. 126-142). Princeton, NJ: Princeton University Press.

61.

Small

D. A.

Loewenstein

Slovic

(2007). Sympathy and callousness: The impact of deliberative thought on the donations to identifiable and statistical victims. Organizational Behavior and Human Decision Processes, 102, 143-153. doi:10.1016/j.obhdp.2006.01.005

62.

Stray

(2012, August 17). Metrics, metrics everywhere: How do we measure the impact of journalism? Nieman Journalism Lab. Retrieved March 24, 2014, from http://www.niemanlab.org/2012/08/metrics-metrics-everywhere-how-do-we-measure-the-impact-of-journalism/

63.

Thiruvengadam

(2013, April 9). How journalists can measure engagement. Poynter Online. Retrieved March 23, 2014, from http://www.poynter.org/how-tos/digital-strategies/209695/how-journalists-can-measure-engagement/

64.

Topsy. (2013). Every tweet ever published. Retrieved from http://about.topsy.com/2013/09/04/every-tweet-ever-published-now-at-your-fingertips

65.

Whitehorn

(1989). The use of the media in the promotion of mental health. International Journal of Mental Health, 18, 40-46.