Abstract
In the age of spirited debates about the mediating role of technologies, the other side of the coin is the state of direct experience in contemporary news production, that is, cases in which news reporters still rely on traditional channels such as “legwork,” “firsthand witnessing,” or “shoe-leather reporting.” The present study is a systematic attempt to identify journalists’ reasons for engaging in legwork, by recreating item by item the work processes and reasoning behind hundreds of individual news reports produced in the digital age, across Israeli print, television, radio, and online news outlets (N = 859). Insofar as legwork can serve as a proxy for painstaking journalism, journalists’ decisions make some difference in determining if more or less legwork will ensue. The data avail an opportunity to explore scholarly musings about journalists’ motivations behind legwork: be they knowledge related, medium related, or event related. We find support for all three possibilities and discuss the implications of these findings.
As members of a “discipline of verification” (Kovach & Rosenstiel, 2007) aspiring to objective reporting (Gans, 1979; Molotch & Lester, 1974; Schudson, 1981; Tuchman, 1978), journalists have always valued legwork (Allan, 2013; Allan & Zelizer, 2004; Gans, 1979; Peters, 2001; Russell, 1999; Zelizer, 2007; Zelizer & Allan, 2010), which we define here as the nonmediated recording of events. Indeed, according to Alsop and Alsop (1958), “[H]is feet, to cite the first rule, are a much more important part of a reporter’s body than his head” (p. 5). This, of course, is not a formal statement on the nature of legwork (which may involve much more than simply using one’s feet), but an aphorism intended to underscore the importance of the nonmediated recording of events in journalism. Although the emphasis on legwork as a source of valuable and reliable information is very flattering to journalism as an institution and journalists as “professionals,” media sociologists teach us that the evidentiary role of journalistic legwork is not at all obvious (Ekström, 2002; Ericson, 1998; Wang, Lee, & Wang, 2013; Zelizer, 2007).
The goal of the present study is to explore both quantitatively and qualitatively what actually motivates journalists to engage in legwork when they have the opportunity to do so—that is, when legwork is not barred by heteronomous institutional or organizational dictates (e.g., when news organizations fail to invest resources into legwork) or by the nature of the occurrences that are being covered, some of which involve no scene to speak of. The study is carried out in Israel, whose relatively small geographic size makes it potentially an optimal test case of how frequent legwork is in contemporary journalism around the world. Indeed, if it is infrequent in a small country like Israel, then even lower frequencies of legwork can be expected in larger countries. Before we begin, however, it is important to provide explicit justifications for our interest in legwork.
Crisis-related justification: In the face of a debilitating crisis, accompanied by massive layoffs, budget cuts, and heightened competition resulting from the technological transformations (Anderson, Bell, & Shirky, 2012; Downie & Schudson, 2009; McChesney & Nichols, 2011), it makes sense to ask the question “How much painstaking journalism remains?” Insofar as legwork can serve as a proxy for painstaking journalism, determining how frequent it is may be of some value in estimating the extent of quality journalism.
Philosophical justification: For several hundred years, Western philosophers have regarded direct experience as evidence par excellence. David Hume (1902), for instance, thought that sense data—the typical consequence of legwork—was the precondition to knowledge. This view survives even in 20th-century philosophy. Arguably the most important philosopher of the first half of the twentieth century, Bertrand Russell (1927/1992) has argued that “the highest grade of certainty . . . belongs to my own percepts” (p. 388). Albeit somewhat controversial in contemporary philosophy (Audi, 2006; Coady, 1973, 1989; Keren, 2012; Moran, 2005), it is nonetheless a respectable philosophical position, and one that enjoys prima facie plausibility among people without any philosophical training (Loftus, 1996), who believe first and foremost what they (and sometimes others) see with their own eyes.
Scholarly justification: It appears that the journalism studies literature has been characterized by a “time bias” (Barnhurst, 2011; Phillips, 2011; Schlesinger, 1977; Tenenboim-Weinblatt & Neiger, 2014; Waisbord, 2013; Witschge, 2012). That is, it devoted a lot of attention to the adverse effects of time shortage on the quality of newswork, all the while neglecting to some extent the space dimension of journalism. Although there have been fruitful conceptual and ethical treatments of the space dimension, there seems to be a scarcity of empirical work on this question (albeit not an absence; see, for example, Allan, 2013; Tait, 2011; Wang et al., 2013; and below).
In the next section, we review the literature that has touched on journalists’ conscious motivations for visiting the scene and formulate research questions. We then lay out the methodology used to study journalists’ motivations for visiting the scene, which is based on (mostly quantitative) face-to-face reconstruction interviews designed to recreate the work processes behind hundreds of individual news items (N = 859). Next, we present our main findings, and finally, we discuss their possible implications for the evidentiary standards used in journalism.
Theory
The scholarly literature has expressed many worries regarding the decline of legwork in the digital age. Due to the combination of a growing assortment of newsgathering technologies, the “hamsterization” of newswork (Starkman, 2010), and growing workloads (Phillips, 2011, 2012; Witschge, 2012), different scholars suggest that journalists develop “deskbound,” “screen-bound” or a “computer-bound” work style of “mouse monkeys” (Fenton, 2009; García Avilés, León, Sanders, & Harrison, 2004; MacGregor, 1997; Manning, 2001; Middleberg & Ross, 2000), becoming heavy users of technology (Fenton, 2009; Garrison, 2004; Hermida, 2010; McClure, 2009) and moderators of citizen copy (Singer et al., 2011).
The question of why journalists engage in legwork as individuals or avoid doing so, however, is quite distinct from the question of why news organizations, qua institutions, invest in legwork or avoid doing so (Schudson, 2005). Although organizational limitations of resources would certainly constrain the ability of reporters to do legwork (Anderson et al., 2012; Downie & Schudson, 2009; McChesney & Nichols, 2011; Van Leuven, Deprez, & Raeymaeckers, 2014), journalists’ own decisions, under conditions when such resources are available, also have an impact on legwork’s incidence. We therefore turn to the hints provided by the extant literature in search for journalists’ own rationales behind the decision to engage in legwork.
Recent empirical works have attended to journalists’ rationales for legwork more or less tangentially (e.g., Allan, 2014; D. Bennett & Lane, 2015; Reich, 2009; Rodgers, 2014; Wang et al., 2013; for an exception, see Bradbury, 2003). But outside of journalism studies, the broader sociological literature has focused on the social meaning of space and time (Elias, 1992; Innis, 2004; Schütz, 1945; Urry, 2012).
Urry (2012), for instance, studied how modern forms of mobility functioned to restructure social life. As modern life became more ramified, copresence became viewed as vital in the face of sensitive, complex, and uncertain information, which was susceptible to misunderstanding, requiring intimacy, trust, assessment of commitment, and detection of lies (Boden & Molotch, 1994; Urry, 2002). These insights point to both fact-related and emotional motivations for aspiring to proximity in space, above and beyond journalism practice. However, as a rare attempt to theorize space on the basis of a large-scale empirical study of journalism practice has concluded, “. . . [r]eporters work much less in space than they do in time” (Reich, 2009, p. 139). It seems, then, that journalists’ motivations for engaging in legwork are quite circumscribed. But what are these motivations?
There are three main journalistic motivations for engaging in legwork that the journalism studies literature has identified, although it has often done so quite obliquely and indirectly. Let us list them.
Knowledge acquisition: In line with the above-mentioned philosophical argument (Hume, 1902; Russell, 1927/1992), journalists’ visits to the scene are said to be motivated by the desire to enhance the reliability of their reports, obtain better evidence, and thereby form better knowledge of the goings-on at the scene. This motivation is attributed to journalists primarily by news reporters themselves, as well as by journalism educators (Godler & Reich, 2013; Harcup, 2014; Kovach & Rosenstiel, 2007; McKane, 2006; Meyer, 2002; Patterson, 2013).
Medium format: The attribution of knowledge motivations to journalists, as noted at the outset, raises some suspicions within sociological circles. Suspicions that question not only the possibility of obtaining knowledge about reality but also the set of mythologized journalistic practices of which legwork is one (alongside cross-verification, poring over documents, and adversarial interview tactics; Ekström, 2002; Tuchman, 1978; Zelizer, 2007). Such practices, it is claimed, may be performed ritualistically, without the goal of knowledge acquisition in mind (Reich & Godler, 2016). In line with the suspicions about journalists’ knowledge-related motivations, sociologists raise the possibility that journalists engage in legwork simply to establish their image as authoritative knowers, both in the eyes of the public and in the eyes of editors and supervisors, rather than to obtain evidence (e.g., Ekström, 2002). Indeed, if people generally accept the proposition that direct experience is evidence par excellence, it should not surprise that journalists draw on this intuition in their self-presentation (Bradbury, 2003). Although it is unlikely that journalists will openly admit such a motivation, it can be gauged indirectly by the degree to which their legwork depends on the prosaic and logistical format requirements of their medium. Thus, for instance, television reporters may engage in legwork simply because of the visual nature of their medium, because news audiences expect visuals, and because their editors expect them to do a stand-up in front of the fire or the murder scene, rather than to shed light on additional evidence (Ekström, 2002; Ericson, 1998; Lipschultz & Hilt, 2014).
Event: Less critically, there is some evidence in the literature that the nature of the events that journalists happen to cover has an effect on their level of attentiveness to the scene. Thus, for example, Livingston and Bennett (2003) demonstrate, among other things, that unexpected events tend to be covered with greater emphasis on the actual goings-on comprising the event.
Correspondingly, and in line with the above-mentioned putative motivations identified in the literature, we offer the following research questions:
Method
Research consisted primarily of a series of face-to-face reconstruction interviews, during which a sample of reporters (N = 108; of which, 81 were males and 27 females) from different beats and media had recreated source by source and news scene by news scene how they obtained a random sample of their routine items that were published or aired during the month preceding the interviews (N = 859). The interviews contained mostly quantitative questions, but also some qualitative ones. We also made sure that no major security and political events took place during the month of sampling, which would compromise the focus of the study on routine news items.
The method used has demonstrated its viability in exploring different facets of news processes (Brueggemann, 2011; Phillips, 2012; Reich, 2009; Shapiro, Brin, Bédard-Brûlé, & Mychajlowycz, 2013). The reconstruction interviews were preceded by three consecutive stages:
Random selection of beats and reporters: This was based on a full list of reporters and newsbeats at the 10 leading Israeli news organizations, representing all four media. The list was prepared in 2 months of byline monitoring, in which research assistants tracked the frequency of publication of various reporters in search for reporters who publish regularly. Making sure that regular reporters’ news items had comprised the sample of news items was necessary, once again, to study legwork as part of routine news rather than exceptional news.
Identification of all published items: The sampling period extended for more than 4 weeks—long enough to supply a rich mix of stories, but not long enough to tax participants’ memories during the interviews, in which they had to recall details about the news items they produced.
Random sampling of news items: To limit interview duration to 60 to 75 min, the sample included eight to 11 items per reporter (the exact number varied according to medium and size of news organization—less items per reporter on television, where reconstructions are more complex and extensive, and more items in news organizations with a smaller workforce of reporters). We sought to limit the duration of the interview because its quantitative and repetitive nature was found to tax journalists’ patience and tolerance to boredom (note that reporters had to answer the same set of—mostly quantitative—questions for every one of their items). Still, we sought as large a sample as we could obtain under these limitations, and calculated on the basis of previous rounds of the study and pilot interviews that eight to 11 items per reporter would yield a sizable sample (N = 859), without crossing reporters’ threshold of tolerance (see Table 1).
Sample Breakdown (N = 859).
Note. The numbers represent absolute quantities of news items from each medium and newsbeat.
Special seating arrangements were used to avoid infringement of source confidentiality. For print, online, and radio items, the reporter (with a pile of sampled stories or printouts) and interviewer (with a pile of questionnaires) sat on opposite sides of a table with a screen placed between them to give the reporter privacy each time he or she was asked to choose another item from the sample pile and detail how it was obtained. Television reporters were given a laptop with video clips of their sampled items and a pair of earphones, enabling them to watch their items in privacy.
Consistent with our research questions, the following variables were defined:
Legwork: The presence of a journalist at the scene, excluding face-to-face interviews and press conferences.
Legwork rationale: The justification given by the reporter for visiting the scene is an open-ended question that was asked whenever the reporter mentioned physical presence (however, reporters were not asked to explain why they did not visit the scene 1 ). The replies were recoded into a finite number of nominal categories corresponding to the aforementioned motivations identified in the literature: knowledge-related, medium-related, and event-related motivations. For instance, one of the reporters’ responses, “to check in person and to see with my own eyes,” was classified as “knowledge related”; another response, “because it was a large and interesting event,” was classified as “event related”; and a third response, “the medium demands it from a technical point of view—I needed a recording in real time,” was classified as “medium related.” Intercoder reliability was performed by the two authors who classified each response according to these nominal categories. The rate of agreement was rather high and amounted to 90.6% (Krippendorff’s α = .88).
Event type: Reporters were asked to provide a general description of the extent of organizing and coordination behind the events that were covered. Qualitative descriptions were, once again, recoded into the categories of “planned events,” “unplanned events,” or other (for instance, a paradigmatic planned event would be a press conference, an unplanned event would be an accident, and the category “other” would apply to mixed and indeterminate cases such as political agreements containing unanticipated elements). Eighty-five randomly selected cases that were coded according to the above categories were tested for intercoder reliability—one of the authors coded all cases, and another coded a randomly chosen 10%—producing agreement in 94% of the cases (Krippendorff’s α = .91).
Cross-verification: This is the juxtaposition by journalists of two news sources against each other with the express intention of ascertaining the information’s reliability.
Medium type: This refers to the technological format of the news outlet, that is, whether it is television, radio, online, or print news.
Sources’ perceived credibility: This is the degree to which journalists judge the sources to be trustworthy, on the scale of 1 to 6 (corresponding to the following categories: highly credible, credible, quite credible, not so credible, not credible, and not credible at all).
Work locus: Beyond the reconstruction of specific news items, we asked reporters to fill out a personal data questionnaire in which one of the questions asked them to indicate in a generalized fashion whether most of their work was done from the scene, from the office, or from home.
Differences in the frequency of legwork as a function of nominal variables such as medium type, as well as ordinal variables such as the sources’ perceived credibility, were tested for statistical significance using chi-square (χ2) tests.
Findings
The present data are the product of a systematic, large-scale analysis of the relationship between the mythologized yet understudied practice of journalistic legwork, and the different conditions of news production that may inspire more or less legwork, based on hundreds of their recently published news items. Descriptively, it is worth pointing out that legwork was present in 131 of the news items (amounting to 15% of the sample), thus lending credence to Reich’s (2009) observation that journalism works “much less in space than . . . in time” (p. 139). The average percentage of legwork in each journalist’s portfolio of news items is M = 15.4 (SD = 16.4). In contrast, we have evidence from a supplementary personal data questionnaire, which, albeit based on generalized self-reports, suggests that the scene was the leading locus from which the journalists in our sample did their work (21% of them to be exact), with work split between home and office coming in second, and amounting to 18% of the sample (work done exclusively from home and exclusively from the office amounted to 12% and 13%, respectively, and the remaining 36% split their work between various loci). Beyond these descriptive data, our research also delved into journalists’ rationales for being physically present at news scenes in an era of technology-mediated coverage.
With respect to
Responding to
Legwork Across Sources’ Levels of Perceived Credibility and Media.
The N represents the overall number of sources in the news items that contained legwork.
These are percentages of news items containing legwork within each medium and hence do not add up to a 100%.
Regarding
Discussion
Contrary to the myth (Ekström, 2002; Tuchman, 1978; Zelizer, 2007), legwork does not serve the singular journalistic purpose of obtaining evidence and acquiring knowledge. However, journalistic knowledge acquisition through legwork is not merely a myth—there are both self-reported and statistical indications that obtaining evidence is the most important role ascribed by journalists to legwork (Kovach & Rosenstiel, 2007).
Although journalists’ self-reports corresponding to the myth about legwork ought to be treated, at least prima facie, with skepticism, this does not mean that they should be dismissed. Indeed, the fact that legwork was relatively infrequent according to journalists’ own self-reports and the fact that evaluations of reliability and instances of cross-verification were correlated with legwork lend credence to the above conclusion that the evidentiary role of legwork ought not be taken lightly (Harcup, 2014; Kovach & Rosenstiel, 2007; McKane, 2006).
But there is even lesser reason to doubt reporters’ admissions about the importance of medium format and the nature of the event in encouraging them to engage in legwork (Ekström, 2002; Ericson, 1998; Lipschultz & Hilt, 2014; Livingston & Bennett, 2003). In these instances, reporters were clearly expressing their readiness to engage in legwork in circumstances where no evidentiary value was anticipated from their being present at the scene.
However, evidentiary value is a concept that may need some unpacking. It is not as if every story requires the same amount of evidentiary work. It is likely that the levels of uncertainty presented by each story combined with journalists’ considerations of public significance and their estimations of public and professional risk in the event of an error affected the standards of evidence journalists felt they had to satisfy.
The combined effects of uncertainty and risk on evidentiary standards have been theorized to some extent within journalism studies, but much more so outside it, in a particular strand of contemporary philosophy. Within journalism studies, it was theorized by the late Wolfgang Donsbach. According to Donsbach (2004), in situations of uncertainty, journalistic confidence in information is affected by social forces, particularly by the dominant view within one’s social group (primarily reporters’ colleagues). In this context, Donsbach (2004) borrowed the concept of “shared reality” (p. 139) from social psychology. As he explained, “journalists often find themselves in what psychologists call uncertain or undetermined situations . . . [S]ocial psychologists have long described the function of groups in solving such uncertain or indefinite situations” (p. 138).
This is, however, merely one way in which social forces may impinge on journalistic evidentiary standards. A literature committed to exploring the possibility of “pragmatic encroachment,” that is, the capacity of practical considerations and values to set the threshold beyond which evidence is regarded as supporting knowledge claims, provides additional examples. This literature argues that (at least in some cases) social values and goals determine the standards of evidence by which one acquires factual knowledge. The oft-repeated demonstrations of this claim are examples from work in the applied sciences (e.g., cancer studies), in which error carries not only knowledge-related risks (i.e., failing to acquire factual knowledge) but also social and ethical risks (e.g., increasing the risk of cancer in society). For example, cancer studies may classify borderline or indeterminate samples of tissue (drawn from rats that have been exposed to environmentally pervasive chemicals) as cancerous to minimize risks to public health. Thus, arguably, a value and a social goal (i.e., the desire to lower the risk of cancer in society) dictate that ambiguous evidence—that is, particularly uncertain evidence—should be regarded as a positive finding (Douglas, 2000; Fantl & McGrath, 2007).
It is important to point out, however, that the findings of our study do not fully substantiate the theoretical possibility suggested by “pragmatic encroachment” theorists. In other words, we do not yet know if “pragmatic encroachment” actually plays a role in journalistic evidence assessment. Greater detail about the instances in which legwork served an overtly knowledge-related function would be necessary to establish such a possibility conclusively.
Still, it appears that the possibility of “pragmatic encroachment” in journalism is at the very least suggested by a combination of several facts that are apparent in the data. Note that the relative infrequency of legwork appears alongside significant correlations between sources’ perceived credibility, cross-verification, and legwork. Thus, although it cannot be doubted that knowledge acquisition guides the decision to engage in legwork in cases in which the veracity of information is at stake, it is clear that the stakes—that is, the risks of error—are in flux and do not always lead to legwork. The alternative is, of course, that journalists failed to engage in legwork simply because they did not seek knowledge in some of the cases. Although possible, we do not have evidence to substantiate this alternative. It is also inconsistent with how most journalists view their practice (Schudson, 2005).
If, however, reporters sought knowledge even in the cases when legwork was not present, then, at least in some cases, they have likely set for themselves lower thresholds of evidence. Fleshing out this possibility may be a worthy task for future, more qualitatively oriented, studies.
Finally, we should make a few comments about the transferability of the Israeli experience. Indeed, it could be argued that Israel’s rather small geographic size—not an irrelevant factor in any study of journalistic legwork—as well as turbulent political and security circumstances do not allow for generalization. Israeli reporters may have an easier time reaching various locations within the country than their counterparts elsewhere, and they may also be driven to various locations outside the newsroom by frequent cases of violent attacks in the context of Israel’s security situation, to an extent unfamiliar elsewhere.
We argue, however, that Israel’s small size does not undermine the transferability of the results but rather strengthens it, while the turbulent political and security situation is controlled for under the conditions of the existing research set up, and even were it not controlled for, it would not undermine the study’s transferability.
It is true that Israel’s total area is relatively small amounting to 20,770 square kilometers within the internationally recognized borders, and thus the geography is, it seems, especially conducive to legwork compared with larger territories. Thus, the Israeli case constitutes, in an important sense, an optimal environment for legwork. However, as the findings demonstrate, even under these optimal conditions, legwork is rather sparse, suggesting that news cultures enjoying less legwork-friendly geographies are likely to produce even a smaller incidence of legwork.
With regard to Israel’s security and political situation, it is worth reminding that the precondition for the present study, and part of the sampling strategy, involved a conscious attempt to single out routine news. Thus, as noted in the methodology, the precondition for conducting the study on the news items sampled herein was that no major security or political event should take place during the sampling period. However, if security-related events have the feature of spurring reporters to engage in legwork, then even if the inclusion of such events had not been controlled for, the rather low incidence of legwork discovered in this study would be a fortiori transferable to countries with less turbulent security situations.
Beyond these comments, we may close off by offering a general description of the Israeli news market. The Israeli news market is dominated by national-level mainstream news organizations, serving a population of nearly eight million citizens from no more than 20 main newsrooms, many of which are located within approximately eight miles of one another in Tel Aviv, along with offshoots of the two main television channels whose headquarters are in Jerusalem and Neve Ilan. The Israeli press is highly competitive, with the race to scoop competitors becoming more severe in the wake of sudden and unpredictable security-related events (Lavi, 2012). As noted, we sought to avoid news items dealing with such events in this study.
Being overwhelmingly privately owned, the Israeli press suffers from the commercial crisis that characterizes Western media (Davidson, 2012; Tsfati & Meyers, 2012), which apparently has the same effect on the dwindling levels of legwork as it does elsewhere. Israeli journalists are technologically up-to-date, just as their audiences are compared with their counterparts in other Organisation for Economic Co-Operation and Development (OECD) countries. 2 Therefore, at least some of the reduction in legwork that results from digital technologies used by journalists around the world is likely to be afoot in Israel as well.
Israeli print news in the four privately owned newspapers studied is published in a single daily edition, representing a nonupdatable product. In terms of length, online news is similar to print items. The two radio stations included in this study, which are publicly owned, air an hourly bulletin comprised of extremely short news items. Meanwhile, the television channels included in this study—one public, two commercial—air a daily evening newscast.
Footnotes
Acknowledgements
The authors thank Inbal Avraham and Yifat Naim for their assistance in data collection, as well as Tali Avishay-Arbel for her statistical advice.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Israel Science Foundation (Grant 1104/11).