A User Perspective on Time Spent: Temporal Experiences of Everyday News Use

ABSTRACT

This paper explores what spending time means from a user perspective. Drawing from three qualitative audience studies that center around the notion of experience, it reveals three complexities regarding time spent in relation to news use. Overall, we find that time spent does not necessarily measure interest in, attention to or engagement with news. First, time spent does not reflect the quality of attention being paid. Second, there is no linear relationship between time spent and interest or engagement. More time spent on news use can be the result of little interest or engagement, and vice versa. Experienced users engage in quick news practices because they are practiced and skillful at using news: they know how to handle and navigate their devices, they can efficiently scan digital environments for new and relevant information, and they are aware of news conventions or title-specific tendencies telling them which parts of news they could skip. Third, different news devices, platforms and genres coincide with different temporal experiences of news.

KEYWORDS:

• Experience
• news use
• news consumption
• time spent
• temporality
• news audiences
• audience studies
• journalism

“Four out of ten Dutch people barely read or watch news.” This rather alarmist headline was published by Dutch newspapers (Trouw, June 22, 2017; Het Parool, June 22, 2017) after The Netherlands Institute for Social Research released their 2017 report “Dutch people and news.” The headline was likely based on the following excerpt from the report’s press release:

More than half of the population (61%) on an average day* uses at least one news medium and spends at least five consecutive minutes on using news media. (SCP.nl 2017)

In the research report, its authors provided more nuance: when also including news use with a duration of less than five consecutive minutes, the percentage of Dutch people who do not consume news every day dropped from 39 to 21 (and on a weekly basis even to 5) (Wennekers and de Haan 2017, 8).

First and foremost an example of how nuance tends to get lost when research turns into news, this anecdote is also illustrative of three interrelated tendencies in journalism and journalism studies. The first tendency is to measure news consumption in terms of how much time people spend on it. Especially since the development of audience analytics it has become common practice to use “time spent” (and adjacent metrics) to measure the consumption of news items, platforms or brands, both in the newsroom (e.g., Cherubini and Nielsen 2016; Cohen 2018) and within journalism studies (Molyneux 2018; Nelson and Lei 2018; Thurman 2018; Thurman and Fletcher 2019). As any researcher will quickly point out, time spent—like any metric—is not a neutral measure of news consumption; one’s choice of metric always impacts the results. This is illustrated by Swedish research comparing news use measured through both page views (clicks) and time spent: in page views, “public sphere” news accounted for 9%, whereas measured in time spent, it accounted for 20% (von Krogh and Andersson 2016). Likewise, studying online news audiences, Nelson and Webster (2016) found no correlation between size (unique visitors) and engagement (time spent). Still, time spent has been recognized as a viable metric for audience attention because it allows for comparisons across platforms (Thurman 2018). Since time spent is likely to become a more dominant measure of news consumption, it is worthwhile to explore in more detail what it does and does not measure.

The second tendency is to use “time spent” to make inferences about news users, especially their interests or preferences. Keeping with the Swedish example, it is easy to see how the page view numbers may be used to support claims about the public’s appetite for “junk news”, whereas the time spent results may serve as evidence of their interest in public affairs news. Potentially problematic about basing audience understanding (solely) on metrics is that these are not intended to capture the interests or experiences of news users, but rather are a measure designed by the news industry, ultimately to quantify and sell users’ attention to advertisers (Ang 1991; Napoli 2011; Webster 2014). What follows is that “problems” and “solutions” regarding news use also tend to be framed in ways beneficial to this industry (cf. Keightley and Downey 2018). For example, in the “attention economy” (Davenport and Beck 2001) which sees news media competing for the finite resource that is the audience’s attention, measuring news use in terms of time spent quickly leads to the question of how the time people spend on news can be increased. An alternative starting point is the perspective of the news users, which instead raises such questions as how and why people engage in short news practices like “checking” and “scanning” (Costera Meijer and Groot Kormelink 2015), and how they might better be served both informationally and experientially.

The third tendency is to see more time spent on news use as inherently or automatically desirable. What the exclusion of short news sessions mentioned at the start of the article implies is that practices like “checking” or “scanning” are less legitimate or desirable forms of news use. This might understandably the case for news organizations, for whom more attention minutes equal more revenue or better public service. From a societal standpoint, following the notion of an informed citizenry, it is also generally taken for granted that users spending more time on news is beneficial. However, while it may seem reasonable to qualify people who indicate spending “no time at all” on (traditional) news as “disconnected citizens” (Blekesaune, Elvestad, and Aalberg 2012), it is quite another question whether more time spent on news necessarily translates into being more engaged; i.e., whether the relationship between spending time on news and being an engaged citizen is linear. A related question is whether there is a threshold in terms of time news use that must be crossed in order to count as legitimate news use.

The aim of this paper is to add to and deepen existing research on time spent and news consumption by exploring what spending time on news means from an explicit user perspective. It does so by drawing upon three recent qualitative user studies that center around the notion of experience (Groot Kormelink and Costera Meijer 2017; 2018; 2019).

Literature: Time and News Use

Central as time is to journalism, it has received limited explicit attention in journalism studies (Bødker and Sonnevend 2018; Zelizer 2018). The same is true for research on news use: although time is featured frequently, it is rarely problematized. Time is typically approached in roughly one of three ways. In the first approach, time serves as the unit of analysis. The constructs used most often for measuring news consumption are frequency and duration (time spent). Frequency refers to the times per day, week or month news is used, and is typically measured through surveys (e.g., Molyneux 2018; Strömbäck, Djerf-Pierre, and Shehata 2013). Duration as expressed in minutes spent on news use is typically measured via tools like Google Analytics (von Krogh and Andersson 2016) and comScore (Nelson and Lei 2018), surveys (Aalberg, Blekesaune, and Elvestad 2013; Blekesaune, Elvestad, and Aalberg 2012; Molyneux 2018) or a combination thereof (Thurman 2018; Thurman and Fletcher 2019). A benefit of minutely measuring time spent is that it allows for detailed and precise comparisons between different news media, platforms, brand or genres. This has led to valuable insights, such as Nelson and Lei’s (2018) finding that mobile app users spend significantly more time on news than those who use a mobile browser, and Thurman and Fletcher’s (2019) finding that the impact of the digital distribution of news on how much time people spend on news differs per age group and per newspaper brand.

In the second approach, time is seen in terms of the temporal organization of news use, often in relation to spatial and social dimensions (Peters 2016; Silverstone 1994). Methods include diaries (e.g., Courtois, Verdegem, and De Marez 2013; Dimmick, Feaster, and Hoplamazian 2011; Hoplamazian et al. 2018) and/or (digital) logs (Van Damme et al. 2015). The focus here is typically on how news use is dispersed throughout people’s day or week and/or how it is embedded within other activities. For instance, Dimmick, Feaster, and Hoplamazian (2011) found that the mobile news occupies a new spatiotemporal niche they call “interstices”, defined as “the gaps in the routines of media users between scheduled activities” (23, emphasis added).

In the third approach, the focus is on the temporal characteristics of the news itself, most commonly in terms of speed. Subsequently, these characteristics are used either for inferring about news use or as input for research into news use. In the first variant, people’s experience or engagement with news is read off news production logics; Keightley and Downey (2018) summarize that it is often assumed that “speedily produced news content and fast, flexible technologies of delivery will necessarily produce temporal experiences which are characterised predominantly by speed and, in many cases, that this will routinely produce superficial engagement with the news, or alienation from it altogether” (105). Spending less time with news is thus equated with staying on the surface and being less interested (or even disinterested) in news. Vice versa, Slow Journalism (a critical response to the negative effects associated with speed in journalism practice) is linked to such notions as “responsible citizenship” (Le Masurier 2015, 149). In the second variant, the temporal characteristics of the news are used as input for research on news use. For instance, based on features distilled from academic literature, Drok and Hermans (2016) surveyed users’ interest in Slow Journalism using such items as preferences for in-depth reporting and explanation. Similar examples include research surveying news users’ preferences for temporal characteristics of online news (production), such as continuous updates (Bergström 2008) and immediacy of reporting (van der Wurff and Schönbach 2014).

Recently, two articles on news use have engaged critically with the notion of time itself. Peters and Schrøder (2018) argue that research has tended to focus on the “here and now” and make the case for a process-based approach to studying news repertories that focuses on “the emergence, maintenance, and (re)formation of audiences’ news repertories in everyday life and across the lifespan” (1079). Keightley and Downey (2018) instead have explored how people themselves experience and navigate the temporal logics surrounding their news use. This aligns with Zelizer’s (2018) critique of how time is typically seen as a “blank slate”, taking “shape more in response to complex settings than as a result of other kinds of interactions” (113). Using Keightley’s (2013) notion of “zones of intermediacy”, they focus on “the experiential arenas in which temporal meaning is produced at the juncture of times—embodied, social, cultural historical and technological” (Keightley and Downey 2018, 100). This concept draws attention to (1) how temporal experiences are produced where several temporalities meet (e.g., clock time, work time), (2) how the temporalities of news texts and technologies impact and set the parameters for users’ experiences, (3) but also how users have agency in navigating and negotiating these times (Keightley and Downey 2018). For instance, the practice of “checking” involves more than a superficial glance at the latest headline: it takes place in a distinct (spatio)temporal context, such as checking one’s phone in the morning to postpone having to get up and get ready for work, or filling up the time between two appointments; it is shaped by how the news is presented to the user (e.g., ordered chronologically and/or ranked by importance) and how the device or platforms is to be operated; and—as this article will show—it is shaped by people’s tactics for using the news.

Methodology: A Case for Experience

As noted, most studies of news use conceptualize time as a given. In this paper, we instead focus on time as part of people’s experience: what spending time on news means from a user perspective. Experience is a fruitful point of departure for four reasons. First, it refers to undergoing or having undergone something (Tuan 1977), and as such helps one move beyond opinions (Costera Meijer 2013) that are not grounded in people’s actual, everyday encounters with news. For instance, for our research purposes, it is less helpful to establish whether news users have a particular opinion about time—e.g., that “accuracy” is more important than “immediacy”—than to establish and understand how they actually go about checking the latest news. Second, focusing on the “undergoing” of an experience enables one to capture how the passing of time is integral to people’s news use.

Third, experience is a broad concept that opens up the possibility of a wide variety of dimensions related to news use to be included, from cognitive to affective, from communicative to esthetic, from material to spatiotemporal (Gentikow 2009; in Ytre-Arne 2011; see also Costera Meijer 2016). Following the notion of “zones of intermediacy” (Keightley 2013), in order to understand what spending time on news means from a user perspective, we must take a broader view that captures how different temporalities overlap and give meaning to an experience.

Fourth, taking experience as point of departure allows for “temporal reflexivity” (Carlson and Lewis 2019) regarding one’s methods and their epistemological consequences. Experience can be described as “a cover-all term for the various modes through which a person knows and constructs a reality” (Tuan 1977, 8). These modes include sensation, perception and conception (Oakeshott in Tuan 1977, 8).¹ Following Throop’s (2003) call for using methods that “differentially access both prereflective and reflective varieties of experience […] [and thus] ensuring that experience is explored ethnographically throughout the entire range of its various articulations” (Throop 2003, 235), we draw from three recent qualitative studies that each center on a different “mode of knowing”. The dominant qualitative method—interviewing—typically depends upon what might be called conception: “those explicit reflective processes that tend to give coherence and definite form to experience” (Throop 2003, 235). In order to include a wide array of dimensions related to how people experience spending time on news, we used a different “temporal orientation” in each of our methods. In study #1, participants reflected on an immediately prior news experience (conception). In study #2 participants verbally reported their news experience in real time while using news (perception). Finally, in study #3, participants were filmed while using news and afterward watched and reflected on these videos, with the aim of capturing sensory and embodied dimensions of their news use (sensation). Table 1 provides further details of each study. The aim, thus, was to use different temporal orientations in order to shine light on different dimensions of what it means to spend time using news.

Table 1. Three user-centered studies.

Study	Method and procedure	Temporal orientation	Dominant mode of knowing
#1 (Groot Kormelink and Costera Meijer 2017)	Viewing and discussing clips: we watched items from current affairs TV shows and immediately afterward interviewed each participant about the extent to which they felt captivated by each item (N = 54).	Reflective	Conception
#2 (Groot Kormelink and Costera Meijer 2018)	Think-aloud protocol: participants browsed digital news as they normally would and said out loud all the steps they made and thoughts they had (N = 56).	Real-time	Perception
#3 (Groot Kormelink and Costera Meijer 2019)	Video-ethnography: participants were filmed from two sides while using news in their own home, on their own devices; immediately afterward the videos were watched and made sense of with each participant (N = 13).	Real-time; reflective	Sensation

Results

By exploring what spending time means from a user perspective, we generated three main findings that add to our understanding of time (spent) as measure of news consumption, in particular as it relates to people’s interest in, attention to or engagement with news. First, time spent does not reflect the quality of attention. Second, time spent is not necessarily linearly related to interest or engagement. Third, all time spent is not equal, as different news media and platforms coincide with different temporal experiences.

Finding 1: “Time Spent” Does Not Reflect Quality of Attention

Our first finding is that time spent does not reflect how that time is spent: the quality of attention. In particular, time spent—measured in minutes or seconds on a page—does not (always) take into account whether users are actively engaging with content (Cherubini and Nielsen 2016). While perhaps not surprising, our data provide more insight into why this is the case.

First, participants regularly opened news items in new tabs without reading them (yet). When visiting news websites on his laptop, Kevin used the following tactic: first, he scanned the headlines on the homepage and pre-selected potentially interesting articles by using ctrl + left mouse click to open each article in a new, separate tab. Then, when finished with this pre-selection, he went through the opened tabs one by one, closing each after having read them. This means that each article was open for a considerable time without having been paid any attention to; indeed, the articles that caught his eye first were open the shortest amount of time. A more extreme example, Myra had a large number of tabs open in her browser, some of which for days and some of which they did not expect to end up reading at all. They explained:

Yes, also very often it’s things I don’t read. Then I think, “oh I’ll read that later” […] and then there’s like 15 things open and then sometimes I don’t read them at all, and they’re open for like two weeks.

In the latter case, then, articles collecting more “time spent” actually points to Myra’s not being very interested in reading them.

This complication can partially be overcome by looking at engaged time: a more sophisticated metric that measures the time users actively spend engaging with an item, for instance by registering scrolling activity (Cherubini and Nielsen 2016). However, even this metric is not faultless. Filming Melanie, we observed that as she was visiting a news site, she was simultaneously watching a series on her television. Her gaze constantly shifted between the article she was reading and the TV series. Yet, because she was nonetheless repeatedly “moving” through the article by constantly having her finger on her mouse, even the “engaged time” metric would not have picked up her shifting attention. Here too, her longer “engaged time” was actually the result of her limited engagement with and attention to the news.

Time spent thus does not capture the quality of users’ attention to news. We also found this in study #1, where we interviewed viewers about their experiences watching items from two different current affairs TV shows. We found a marked difference on how the two shows were watched. One show was appointment television: participants watched it intently in a concentrated, lean-forward fashion, because they wanted to learn about politics. The other show was watched in a lean-back fashion: participants often had it play in the background while doing other activities, paying attention with one ear. Carmen illustrates: “It’s on in the background, but then I’m also cooking and [unpacking] the groceries, so doing all kinds of things at home, and then if it’s really interesting I will watch a bit.” The amount of time spent viewing these shows does not capture the varying levels of attention and interest involved.

It is also worth noting that spending time on news is not a stamp of approval from users. Previous research showed that people consume news that they do not see as quality news (Costera Meijer 2007), news that does not lead to a satisfying viewing experience (Stanca, Gui, and Gallucci 2013), and news that they do not trust (Tsfati and Cappella 2003). Similarly, our participants had no problem admitting that their attention to news did not correspond with their judgement of quality. For instance, Walter (27) criticized the journalistic quality of one news item but watched attentively because it was entertaining: “I think at least television-wise it’s well done, and as a result I paid attention.” In study #2, which explored why users (do not) click on news, we also found that spending time with news did not necessarily mean participants appreciated it. This was especially the case with “clickbaity” headlines:

Joe (26):

“Then I click it if it’s a very sensational headline, so I’m more like ‘oh what the hell’ […] sometimes there are headlines that make you go like ‘what is this ridiculousness’ or ‘what the hell is that supposed to mean’, and then I’m sometimes inclined to click”.

Jack (56):

“‘Metal band wants money for music use during torture Guantanamo’, then I’m like, god what an item (clicks) then I click it and I think, what kind of band is that, what’s the story behind it”

Danny (25):

“Pretty often Nu.nl has these stupid news items about, I don’t know, a new year’s day dive. Couldn’t care less, but if it happens to have a picture of a lady, I do click on it”.

From a user’s perspective, spending time with news is no quality endorsement.

Finding #2: Less Time Spent Can Point to Not Less but More Interest or Engagement

Above we showed that time spent is not an unflawed measure of interest or engagement because it does not capture how that time is spent. A perhaps more surprising insight is that shorter news sessions can indicate not the lack but the presence of interest or engagement. In the methodology section, we listed four reasons for taking experience as point of departure, one of which was that experience suggests actually undergoing something. Study #3 revealed the particular relevance of “the ability to learn from what one has undergone” (Tuan 1977, 9, emphasis added). Indeed, we found three ways in which what might call people’s “being experienced” with news was the reason their news use was so quick or short.

First, “experienced” news users have embodied knowledge of how to most efficiently use their devices (see also Groot Kormelink and Costera Meijer 2019). Their tactics included quickly swiping downward to refresh and update a news app to see the very latest headlines, rearranging the icons on their smartphone so that their favorite apps are within thumb’s reach, and scrolling with a specific finger so they can scroll faster. Consider the extreme detail (and proficiency) in Regina’s explanation of why she uses different fingers for different actions on Twitter:

For scrolling et cetera I just use my thumb on Twitter. […] Sometimes when I’m ALL the way down and I have to go all the way up uhm then it’s faster to scroll with my forefinger because […] with my right hand I hold [my phone] so […] with my forefinger I have access to the entire screen so that […] I can make the movement of scrolling to the top bigger, because my reach is larger than with my thumb, because my thumb is not large enough to get all the way to the top of the screen.

More directly related to time spent, when coming across a potentially interesting long read while using Facebook on his smartphone, Ferdinand very quickly—within ten seconds—decided whether or not to save the article by using the save-for-later app Pocket installed on his phone. He clicked on the article, scrolled through it very quickly to pick up words indicating the essence of the article (“When I saw this [section] “what to do” [it] told me […] that there was more depth to the news”), clicked on “copy link”, opened his Pocket-app, and tapped “add” on the latter’s pop-up suggestion “Add copied URL to your list?”. Although—so he claims—he would read the full article in Pocket later, the “engaged time” metric would register a mere ten seconds spent in the original article. For these experienced users, then, the very quickness of their news use attests to their skills and resourcefulness rather than their disengagement: their sessions were short because they are efficient and proficient in navigating and handling their apps and devices.

Second, more experienced and especially avid news users had “quick” news sessions not because they were disinterested or disengaged, but because they were very efficient at scanning the environment and picking out news relevant information. For instance, Robert quickly went through several news app successively. Within each app, he scanned all the headlines added since his last “checking cycle” (Costera Meijer and Groot Kormelink 2015) earlier that day, and picked out several articles he wanted to read. Within these articles, he was also focused on scanning for new information. He explained:

If I read things in an item I already know, I tend to skip past it, so I don’t read everything in the item because often it contains things that are a repetition of something I had already read elsewhere or [that I] already know.

Such scanning might best be typified as a burst of news use: a short, intense session characterized by highly efficient information scanning (cf. Ytre-Arne and Moe 2018). Here, too, the short time spent points to intense engagement with news rather than superficial engagement or disengagement.

Finally, some participants relied on what we might call their meta-knowledge of news: they are aware of general conventions or title-specific tendencies of the news and adjust their news use accordingly. For instance, Regina referenced the inverted pyramid (“that pyramid model: first the very most important news and then it becomes this little funnel”) when explaining why she repeatedly skipped the last paragraphs of news items. Consider also Fiona, who in her e-paper often first reads the lead and the conclusion, and only then decides whether it is worth reading the rest of the article. This way she does not waste her time reading what she described as “exaggerated” articles. As she explains:

[…] I often do that, the beginning and the end, sometimes […] you read the headline and the conclusion [and that] is enough. Then it’s an enormously suggestive article and then in the end it concludes: it’s all not so bad.

She illustrated this when encountering an article about a high ranking leader of the Taliban being killed. Reading the piece in its entirety would have been waste of time:

That bearded dude on the right has been killed, but then I read the beginning [points to the first paragraphs] and then in the end it concludes: his successor is already standing by, then I’m like well (laughs) That’s how it goes (laughs) Then I’m like yeah, this one is has been wiped out and the next one is ready, problem is not solved.

Again, here Fiona’s and Regina’s limited reading time per article points not to their disinterest or disengagement but to their savviness and experience as news users.

Finding 3: All Time Spent Is Not Equal

Whereas the first two findings concerned the question of what the metric “time spent” does and does not measure, the third finding relates to the comparability of news use based on time. To be sure, since attention is a finite resource that people can only spend once, time is a most useful measure for studying how people allocate their overall attention between different activities, devices or platforms. However, what it means to spend more time on one medium versus another is a more complicated matter.

First, while a very rudimentary division, digital news use seems to call up different expectations of time than traditional news use (print newspapers, TV), not necessarily in terms of the immediacy of content, but regarding the speed or—to be more precise—the efficiency of the user practice itself. Participants expected quick, smooth user experiences and were frustrated when these were interrupted. This is illustrated by Kevin, who while reflecting on a past experience recounted a deep, bodily felt frustration concerning an error on the Dutch news site NOS.nl:

Can I leave a complaint with you about the NOS website? Those videos don’t load properly anymore, you have to, VERY annoyingly, you have to reload them and then it gives you an error message, reload and only then it works. […] That’s really fucking annoying. […] No, but it’s really been annoying me. Because often it’s clips of like 30 seconds and then I’m like “oh nooo” [throws up hands and rocks back and forth, conveying a physical expression of frustration] […] then on one page they have three clips and then you keep having to reload and click again and reload and wehhh. […] I’ve noticed I’m watching less because of that, then with the third clip I’m like never mind […] so it does have a negative impact on my news consumption.

While in objective time reloading a page may only takes a few seconds, it apparently felt like a major interruption of his browsing experience. Mangen (2008, 412) describes the impatience users experience when using digital media as “an experiential situation bereft of both physical and phenomenological presence”. When a digital medium freezes or falters, the technology is brought to the fore and is experienced as an obstruction or intrusion (Mangen 2018; see also Ihde 1990). Although a print newspaper, too, can break down and reveal itself as a technology—e.g., when a drop of coffee falls from our mug onto the paper, making the text suddenly unreadable—digital media experiences are more easily interrupted. Participants’ expectations of smoothness were also apparent in study #2, where some did not want to click on videos because they did not want to sit through a commercial, which was seen as an interruption of the flow of their news use. Compared to traditional news use (especially the print newspaper), during digital news use the experience of time seems to become intensified.

Second, different devices and platforms co-produce different temporal experiences. A prime example here is a distinct user practice on Facebook we have previously called “scrolling”, characterized by an urge to keep the movement of scrolling going, regardless of one’s interest in the content (Groot Kormelink and Costera Meijer 2019). This is illustrated by Ferdinand, who watched a news video on Facebook for only twenty seconds before feeling the urge to move on: “I thought [the video] was really nice but I don’t wanna spend too much time doing it.” Although he found it difficult to verbalize what exactly made him want to limit his time with content and keep scrolling other than feeling an urge “to keep it going”, it appears that the platform—including the way it is navigated and physically handled with one move (a mouse scroll or a thumb scroll)—invites a sense of restlessness that encourages light, quick engagement with content. Rushkoff (2013) has described this as a “perpetual now”, in which we ignore what is before us and “compulsively anticipate the next decision point” (116). As Ash (2015) clarifies, this is not to be confused with Csikszentmihalyi’s (1990) notion of “flow” which refers to the positive state of being in the zone, absorbed in the activity at hand. Indeed, participants expressed no joy regarding their urge to keep scrolling.

A specific affordance of the video-ethnography (study #3) was that it captured—without interruption—the impact of the passing of time on news use. The videos showed that as time went by, some participants grew satiated with news, in the sense that they felt they could not absorb any more information. This was especially the case with participants using media that did not let them (as readily) decide themselves in which order to consume news items. On news websites and apps, participants selected and read the items they deemed most interesting first and stopped when they had read everything they wanted to read. On the contrary, on devices and platforms where users tended to follow the order suggested by the creator (e.g., newspaper or e-paper) or could not at all choose the order of news items (e.g., Facebook), satiation impacted their selection behavior and reading style (e.g., reading versus scanning) more pronouncedly. For instance, after having read her newspaper (chronologically) for a while, Norah leafed through her newspaper faster and faster and only glanced at headlines. She even skipped articles that she found very interesting, such as an article regarding dark matter:

I watched the [documentary] series Cosmos [A Spacetime Odyssey], I find that SUPER interesting how all of that works, astronomy and physics. I’d actually want to read that, but now I’m thinking “pff, it doesn’t fit [into my head] anymore.” […] Because then I have to think about dark matter which is already a VERY complicated concept which no one can actually explain what it is, and then I have to read about that now […] pff, I can’t anymore (sighs).

She did not read an article she found “super interesting” because gradually she had become satiated: she felt she could not absorb any more information. Similarly, while viewing the recording of his own Facebook scrolling session, Ferdinand noted:

Yeah ‘cause then I got tired of checking. Yeah this, ah yeah, the Time article was about uh uh … 

Interviewer: [tries to read the headline as captured on video] “6 things”

Yeah how to get things done. And I thought it could be nice to click on it just yeah for inspiration, but then I was already tired of clicking so [I] just kept scrolling.

Like Norah, Ferdinand came across an article that interested him, but due to having become “tired” of clicking, he did not select it. This suggests that when studying people’s selection or time-spending behavior, the “reading order” suggested or enabled by the platform and device must be taken into account, in particular when inferring people’s interests or preferences. For news media with a more or less “predetermined” reading order, the passing of time has greater impact: during the beginning of a user’s news session, their selection or time-spending behavior might be more indicative of their (lack of) interest than toward the end of that session. That is, news people come across later—regardless of how interesting they find it—is more prone to being skipped due to them having become satiated.

Even when comparing use of the same news medium, time spent is not always revealing. An illustration is the difference between allocating one’s time for news and restricting one’s time with news. Fiona earmarks her Saturday morning for her newspaper. She sees this as a treat, a moment to relax on her day off. Norah, on the other hand, limits her time with her Saturday paper, actively taking time into consideration when reading individual articles:

And maybe half way [the article] I’ll think, yeah now I get it, let me go on, because I still have a whole newspaper and I don’t plan on spending two hours on the newspaper, because I have things to do.

Norah’s not wanting to spend too much time on the newspaper is an example of her navigating and negotiating different temporalities, including news delivery time (Saturday morning), leisure time (since it is weekend she does not want to spend too much time on “troublesome” things like news) and chore time (having “things to do” like groceries). The actual time she spends is insufficient to make sense of this practice: in this “zone of intermediacy” (Keightley 2013) it is the juncture of these different temporalities that gives her experience its particular meaning.

Conclusion

Aiming to explore what spending time means from a user perspective, this article added three nuances regarding “time spent” as a measure of news consumption. Overall, our results suggest that time spent does not necessarily measure interest in, attention to or engagement with news. First, time spent does not reflect the quality of attention being paid. Second, there is no linear relationship between time spent and interest or engagement. More time spent on news use can be the result of little interest or engagement, and less time spent can be an indicator of more interest or engagement. The more “experienced” news users tended to engage in quicker news practices exactly because they were “practiced” and skillful at using news: they knew how to handle and navigate their devices, they could efficiently scan digital environments for new and relevant information, and they were aware of news conventions or title-specific tendencies telling them which parts of news articles they could skip. Therefore, discounting news use of less than five minutes does not do justice to the intensity and efficiency of the “bursts of news use” we found; these point not to disengaged citizens but to experienced, skillful, inquisitive news users.

Third, different news devices, platforms and genres coincide with different temporal experiences. Time is experienced differently depending on several factors, one of them being the type of device or platform used. Television and radio are often used in a lean-back mode (Lull 1990; Larsen 2000) and reading the newspaper is often experienced as a moment of relaxation, a ritual treat (e.g., Berelson 1949). With digital news media, on the other hand, speed and efficiency appear to be valued more—or perhaps more precisely, the lack thereof is disliked—making the experience of time become intensified. Although time spent is certainly useful to measure how people divide the finite resource that is their attention, our results suggest that we must be mindful of what we can infer from differences between platforms in terms of time spent. What do these findings mean in light of the decline in time people spend on digital news as compared to print news (Thurman 2018; Thurman and Fletcher 2019)? Our studies are ill-equipped to speculate about the democratic implications—such as learning from news on different platforms (Kruikemeier, Lecheler, and Boyer 2018)—and economic implications of the transition to digital news. However, it is worth emphasizing that while the digitalization of journalism has certainly broadened people’s arsenal of news user practices (Costera Meijer and Groot Kormelink 2015), enabling quicker, more efficient news use, there is also a sense of restlessness and compulsiveness to such digital news practices as scrolling. We therefore propose further research on the affective qualities of these newer forms of news use (Ytre-Arne and Moe 2018).

Our results suggest that when measuring time spent on individual (digital) articles, differences between devices and platforms should also be taken into account. Whereas on news apps users typically select and read articles in succession, on news websites users can open several articles simultaneously. This first of all makes it especially urgent to use “engaged time” rather than “time spent” on news websites. What is more, on platforms where users have less (e-paper, paper) or no (perceived) freedom (Facebook, Twitter) to choose their own order of reading, it is important to take into account that readers can grow tired or satiated as their news session progresses. What this implies is that for these platforms, “time spent” becomes a less reliable measure of interest the “further” users are; toward the end of their session they will be more inclined to skip content they do find interesting.

Our results suggest that a newsroom strategy of retaining users’ attention as long as possible may make sense from an attention economy perspective, but does not necessarily match with the experience of (digital) news use which is often—although certainly not always—characterized by smoothness and efficiency. One alternative strategy is to provide news in such a way that it simultaneously affords different user practices, from “scanning” to “reading” (Costera Meijer and Groot Kormelink 2015). For instance, by summarizing a news article in a couple of succinct bullet points, “scanners” are enabled to quickly get the gist of the story, while not stopping the “readers” from consuming the full story (indeed, it could be argued that such a schema can also help readers to better understand the core of the article). This strategy maximizes not the time that people spent on the story but the number of people that can get value out of it.

A different challenge is how to get news users to take their time to consume content they are already interested in. What the logic of “scrolling” (i.e., “keeping it moving” even if one likes the content) suggests is that users find it (increasingly?) challenging to “commit” to one news item when there is a plethora of other content to consume. One strategy could be to provide content that is experienced as valuable (Costera Meijer, 2019). The small sample from our video-ethnography tentatively indicates that news users are more willing to invest their time in news that generates insights. For instance, articles that made participants curious about specific but insignificant details were clicked and scanned until the answer was found (“I’m just gonna check which supermarket it was [that was robbed]”), whereas articles that were experienced as more constructive or insightful (“it gave me new insights [and] a broader and richer picture of the issue”) tended to be read more fully. This finding is corroborated by our current affairs TV study (study #1) that showed that users were willing to invest their time and attention in exchange for insight into complex political matters. Finally, following the example of the app Pocket, another strategy is to make it easier to save and access articles for later consumption, so that interesting content the user comes across but cannot consume at that moment can be efficiently accessed during a more opportune moment.

It should be emphasized that our findings are based on qualitative research and that quantitative research is needed to establish how dominant the patterns we found are. E.g., we cannot say how prevalent such practices as efficient information scanning or having multiple tabs with news open at once are. Indeed, our aim was to add to our understanding of “time spent” by exploring spending time on news from a user perspective.

Acknowledgements

The authors thank the two anonymous reviewers for their insightful and constructive feedback.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This article is part of the research project “The New News Consumer: User-Based Innovation to Meet Paradigmatic Change in News Use and Media Habits,” supported by The Netherlands Organisation for Scientific Research (NWO) [314-99-103] and ten Dutch journalism organizations (see http://www.news-use.com).

Notes

1 Whether these modes of knowing are empirically separable is beyond the scope of this paper; we use them for their heuristic value.