Comments on: You go to war with the data you have http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/ EthanZ's musings on Africa, media and international development Wed, 08 Feb 2012 20:45:31 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: …My heart’s in Accra » The Partisan Internet and the Wider World http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-2047752 …My heart’s in Accra » The Partisan Internet and the Wider World Mon, 24 May 2010 21:59:43 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-2047752 [...] evidence that internet users had broader exposure to political arguments than non-users. (I offer my critique of the Horrigan, Resnick and Garrett paper here, relying on the Farrell et al paper for counterarguments. Dr. Garrett was good enough to engage [...] [...] evidence that internet users had broader exposure to political arguments than non-users. (I offer my critique of the Horrigan, Resnick and Garrett paper here, relying on the Farrell et al paper for counterarguments. Dr. Garrett was good enough to engage [...]

]]> By: Science Blogs » Blog Archive » Obligatory Readings of the Day: Science Outreach and Online Behavior http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-1167622 Science Blogs » Blog Archive » Obligatory Readings of the Day: Science Outreach and Online Behavior Tue, 26 Aug 2008 03:42:05 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-1167622 [...] You go to war with the data you have: My interest in these experiments has less to do with questions of political polarization and more to do with interest in international news. Are internet readers more inclined to look for information about other countries, since they've got such a wealth of information at their fingertips? Or are they more inclined towards information on their home countries, since they can easily choose to avoid international news. Extrapolating from Pew's data suggests that wired readers might consult more sources and perhaps consume a more diverse diet; Farrell's research points to a strong homophily effect, which suggests the possibility of geographic cocooning. [...] [...] You go to war with the data you have: My interest in these experiments has less to do with questions of political polarization and more to do with interest in international news. Are internet readers more inclined to look for information about other countries, since they’ve got such a wealth of information at their fingertips? Or are they more inclined towards information on their home countries, since they can easily choose to avoid international news. Extrapolating from Pew’s data suggests that wired readers might consult more sources and perhaps consume a more diverse diet; Farrell’s research points to a strong homophily effect, which suggests the possibility of geographic cocooning. [...]

]]> By: Cos http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-1165690 Cos Sun, 24 Aug 2008 17:22:14 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-1165690 When I first heard and read about Robert Putnam's <em>Bowling Alone</em>, I felt there was something wrong with his thesis that social capital in America was declining. I remember hearing him interviewed on the radio, talking about declining participation in neighborhood groups, bowling leagues, and so on, and thinking "but that's not how my social capital works." How many times this week, he asked, have you held a dinner party, attended a civic group meeting... well, I actually do attend monthly meetings of my neighborhood group sometimes, but my thought was, "he didn't ask how many times a *day* I've commented on the LiveJournal of someone I know in person and see regularly - let alone how many times I've exchanged comments with one of *their* friends". A lot of this clicked together for me when I read Richard Florida's <em>The Rise of the Creative Class</em> a couple of years ago. Florida argues, persuasively, that the kinds of social capital Putnam measured are in fact *barriers* in today's world. Those older forms of social capital are stable and tightly-knit, and make it harder for people to move around and integrate into new places and groups of people. A more powerful form of social capital today, he argues, as a more networked kind of social capital full of shifting masses of loose connections - just like LiveJournal, I thought! Cities & regions that have a more networked style of social capital are more welcoming to newcomers and diversity, according to Florida, and better able to attract & integrate talent, and therefore outperform cities & regions where the old kind of tightly-knit, stable social capital still dominates. The reason I write all this here is that I suspect we're liable to make a similar mistake when looking at whether the Internet is broadening or narrowing people's perspectives, whether homophily or diversity of diet are stronger influences. We may be tempted to look at the broad vs. narrow divide we think we recognize, and in doing so, ignore ways in which breadth & narrowness of perspective is *different* in the Internet age, rather than more or less towards one side of the existing continuum. We may, as Robert Putnam did, be asking the wrong questions. When I first heard and read about Robert Putnam’s Bowling Alone, I felt there was something wrong with his thesis that social capital in America was declining. I remember hearing him interviewed on the radio, talking about declining participation in neighborhood groups, bowling leagues, and so on, and thinking “but that’s not how my social capital works.”

How many times this week, he asked, have you held a dinner party, attended a civic group meeting… well, I actually do attend monthly meetings of my neighborhood group sometimes, but my thought was, “he didn’t ask how many times a *day* I’ve commented on the LiveJournal of someone I know in person and see regularly – let alone how many times I’ve exchanged comments with one of *their* friends”.

A lot of this clicked together for me when I read Richard Florida’s The Rise of the Creative Class a couple of years ago. Florida argues, persuasively, that the kinds of social capital Putnam measured are in fact *barriers* in today’s world. Those older forms of social capital are stable and tightly-knit, and make it harder for people to move around and integrate into new places and groups of people. A more powerful form of social capital today, he argues, as a more networked kind of social capital full of shifting masses of loose connections – just like LiveJournal, I thought! Cities & regions that have a more networked style of social capital are more welcoming to newcomers and diversity, according to Florida, and better able to attract & integrate talent, and therefore outperform cities & regions where the old kind of tightly-knit, stable social capital still dominates.

The reason I write all this here is that I suspect we’re liable to make a similar mistake when looking at whether the Internet is broadening or narrowing people’s perspectives, whether homophily or diversity of diet are stronger influences. We may be tempted to look at the broad vs. narrow divide we think we recognize, and in doing so, ignore ways in which breadth & narrowness of perspective is *different* in the Internet age, rather than more or less towards one side of the existing continuum. We may, as Robert Putnam did, be asking the wrong questions.

]]> By: links for 2008-08-22 « andrew golis http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-1163414 links for 2008-08-22 « andrew golis Fri, 22 Aug 2008 21:07:28 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-1163414 [...] …My heart’s in Accra » You go to war with the data you have Ethan Zuckerman on homophily/polarization research. V. important. (tags: empathy partisanship new.media blogging ethan.zuckerman) [...] [...] …My heart’s in Accra » You go to war with the data you have Ethan Zuckerman on homophily/polarization research. V. important. (tags: empathy partisanship new.media blogging ethan.zuckerman) [...]

]]> By: Kelly Garrett http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-1162863 Kelly Garrett Fri, 22 Aug 2008 14:17:17 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-1162863 Thanks for the insightful post. As one of the authors of the Pew study, and as a researcher actively studying the questions you discuss here, I want to offer a couple additional thoughts. First, I wholeheartedly agree with your assertion that the Pew results and the results of Farrell and colleagues are compatible. I think the similarity may be a bit deeper than you have suggested. In their claims about how people seek political information online, both groups of researchers suggest that people prefer opinion reinforcement to opinion challenge. Where our interpretations differ is in how to explain this preference. One common explanation is that people seek opinion reinforcement while actively avoiding opinion challenges. For instance, this appears to be a key element of Sunstein’s “echo chamber.” The problem with this explanation is that there is little evidence that people want to avoid other viewpoints. Yes, people prefer information sources they agree with, but the Pew study (and others before it) suggests that people don’t put much effort into avoiding contact with other perspectives. For example, if online news users really want to avoid other viewpoints, they should abandon the comparatively balanced mainstream media for more partisan alternatives, but as the Pew report indicates, that’s not what people are doing. In your post, you also point out that the Pew report doesn’t provide much detail about the regression analyses. If you’re interested, you can see a more complete analysis in this paper: http://www-personal.umich.edu/~garrettk/Garrett-ICA-Seeking_Similarity-final.pdf The linked paper also describes results of an experiment examining how online news readers choose among diverse news stories about contentious political issues. The bottom line is that individuals’ issue positions do influence their decisions about what they look at, but the effect is small and people are drawn to opinion reinforcement more than they are repelled by opinion challenges. You also point out that the Pew results don’t tell us about the context in which people encounter arguments that differ from their own. Perhaps they learned about the other side on a partisan blog favoring their position. Even when this is the case, these results are important. One concern stemming from Sunstein’s echo chambers is that people will become less aware of other viewpoints and that this will lead them to be more extreme, less accepting of others views, and potentially more violent. People who are familiar with both sides of a debate are more tolerant than those who only know arguments favoring their viewpoint. So exposure to criticism of your political beliefs is an important element of effective democracy. To be sure, you are correct when you observe that hearing the other side’s arguments is not the same as giving them your unbiased consideration, but both activities are important. Thanks for the insightful post. As one of the authors of the Pew study, and as a researcher actively studying the questions you discuss here, I want to offer a couple additional thoughts.
First, I wholeheartedly agree with your assertion that the Pew results and the results of Farrell and colleagues are compatible. I think the similarity may be a bit deeper than you have suggested. In their claims about how people seek political information online, both groups of researchers suggest that people prefer opinion reinforcement to opinion challenge. Where our interpretations differ is in how to explain this preference. One common explanation is that people seek opinion reinforcement while actively avoiding opinion challenges. For instance, this appears to be a key element of Sunstein’s “echo chamber.” The problem with this explanation is that there is little evidence that people want to avoid other viewpoints. Yes, people prefer information sources they agree with, but the Pew study (and others before it) suggests that people don’t put much effort into avoiding contact with other perspectives. For example, if online news users really want to avoid other viewpoints, they should abandon the comparatively balanced mainstream media for more partisan alternatives, but as the Pew report indicates, that’s not what people are doing.
In your post, you also point out that the Pew report doesn’t provide much detail about the regression analyses. If you’re interested, you can see a more complete analysis in this paper: http://www-personal.umich.edu/~garrettk/Garrett-ICA-Seeking_Similarity-final.pdf The linked paper also describes results of an experiment examining how online news readers choose among diverse news stories about contentious political issues. The bottom line is that individuals’ issue positions do influence their decisions about what they look at, but the effect is small and people are drawn to opinion reinforcement more than they are repelled by opinion challenges.
You also point out that the Pew results don’t tell us about the context in which people encounter arguments that differ from their own. Perhaps they learned about the other side on a partisan blog favoring their position. Even when this is the case, these results are important. One concern stemming from Sunstein’s echo chambers is that people will become less aware of other viewpoints and that this will lead them to be more extreme, less accepting of others views, and potentially more violent. People who are familiar with both sides of a debate are more tolerant than those who only know arguments favoring their viewpoint. So exposure to criticism of your political beliefs is an important element of effective democracy. To be sure, you are correct when you observe that hearing the other side’s arguments is not the same as giving them your unbiased consideration, but both activities are important.

]]> By: Tino http://www.ethanzuckerman.com/blog/2008/08/21/you-go-to-war-with-the-data-you-have/comment-page-1/#comment-1160608 Tino Thu, 21 Aug 2008 10:57:04 +0000 http://www.ethanzuckerman.com/blog/?p=2169#comment-1160608 Great post!! Does the internet forge a larger consumption of international news stories? Good question. But as you point out, the causal relationship between interest and available sources has never been proven. What Horrigan and Farrell have shown is merely that internet users access more news sources, even if it's not necessarily a broader political perspective they gain. As Wu (e.g. http://gaz.sagepub.com/cgi/content/abstract/69/6/539) keeps showing us, the other factor is what international news stories you _can_ get, even from a plethora of sources: if LAT and NYT and CNN give the exact same story on a Chadian rebel group (because they all use the same wire AP report), users will probably be less inclined to use more than one source for international news. I think in the Pew study you can see that Americans still prefer the big networks and portals for news, and it's still very commonplace for (left-leaning) blogs to use the Times and other biggies to source their information. So, as international stories go, how much of a diverse story _can_ you get? (yes, GV is the beautiful solution, but unfortunately just not that popular yet...) I'm working on a similar research design with South African students, focusing on their mobile media consumption (the cell phone is the most widely used internet technology by this point). We're doing a mix between quantitative surveys and experimental, almost ethnographic methods. I'll send you the results. Great post!! Does the internet forge a larger consumption of international news stories? Good question. But as you point out, the causal relationship between interest and available sources has never been proven. What Horrigan and Farrell have shown is merely that internet users access more news sources, even if it’s not necessarily a broader political perspective they gain.

As Wu (e.g. http://gaz.sagepub.com/cgi/content/abstract/69/6/539) keeps showing us, the other factor is what international news stories you _can_ get, even from a plethora of sources: if LAT and NYT and CNN give the exact same story on a Chadian rebel group (because they all use the same wire AP report), users will probably be less inclined to use more than one source for international news. I think in the Pew study you can see that Americans still prefer the big networks and portals for news, and it’s still very commonplace for (left-leaning) blogs to use the Times and other biggies to source their information.

So, as international stories go, how much of a diverse story _can_ you get? (yes, GV is the beautiful solution, but unfortunately just not that popular yet…)

I’m working on a similar research design with South African students, focusing on their mobile media consumption (the cell phone is the most widely used internet technology by this point). We’re doing a mix between quantitative surveys and experimental, almost ethnographic methods. I’ll send you the results.

]]>