Internet politics myth-busting, part 2

Part 1 discussed a survey by the Institute of Policy Studies (IPS) just after the May 2011 general election and how it failed to find any direct impact from alternative online media on it. Part 2 here discusses the possibility of indirect effects, with special reference to the two-step flow model and agenda-setting, angles that were touched upon at the conference held on 4 October 2011.

Quite early in the proceedings, Professor Ang Peng Hwa from the Nanyang Technological University (NTU) spoke about a long-standing theory in communication research: the two-step flow. In the course of the various presentations, this theory was mentioned a few more times. This is not a new theory; it’s been around for decades. In that period there has been empirical support for it, but also criticisms. Two-step flow posits that ideas put out by mass media are not immediately absorbed by the public. Ideas need to be reinforced by opinion leaders whom each section of the public trusts or who are similar in profile with the recipient.

If this was in operation, it would explain why it would prove difficult to trace how alternative online media influenced political discourse. Associate Professor Cherian George, also from NTU, explained that ideas and perspectives might originate on one platform, and be passed from one person to another via other modes of communication, e.g. the mainstream press, social media or personal conversation. The final recipient, whose views might have been changed as a result, would not even realise that the idea, perspective or opinion originated from an alternative blog. (By the way, a transcript of Cherian’s presentation on Diverse strategies in political blogs’ election coverage is available online.)

Other speakers gave tantalising clues as to how such phenomena can be studied (perhaps in future elections). Paul Wu and Randolph Tan from SIM University and Carol Soon from the National University of Singapore attempted a study titled Agenda setting: What and How, in which they monitored which media platform raised which issues and when. Who was driving the election agenda?

None of the papers presented gave conclusive answers to the question of how influential alternative online media was. Mostly this was because I don’t think anyone was specifically testing for the theory. Below, however, I will mention some papers which may have a bearing on indirect transmission of viewpoints.

Political talk

Zhang Weiyu looked at how often ordinary people engaged in political talk in everyday conversations (i.e. not with politicians).  Responses were gathered via a five-point scale (1 = never, 2 = rarely, 3 = sometimes, 4 = often, 5 = very often), and here are the means for talking to four classes of conversational contacts :

People talk more about politics with those who are close to them.

Comparing conversation various kinds of media,

People judged that talking with others was of third most importance after newspapers and TV in obtaining political and election-related information. However, they trusted what they learned from conversations rather less.

Dealing with conflicting information

Natalie Pang spoke twice at the conference. The second time she came on, she described an experiment she conducted involving 60 participants. In a pre-experiment survey, their views on an issue (foreign labour) were ascertained. Then they were split into two groups: Group A were asked to view sites with conflicting information; Group B were asked to view sites with consistent information. They were surveyed again post-experiment.

I could be wrong, but from what I understood of her presentation, she found these outcomes:

Strangely, she labelled her outcomes slide “Increased certainty in supporting a party” which didn’t make a lot of sense to me since the topic she visited on her subjects was foreign labour. You will also notice that as far as my notes go, she didn’t state what outcomes came from participants who initially had uncertain views and been given consistent information. Nonetheless what was interesting was that participants who were initially uncertain, even when given conflicting information, progressed to greater certainty. People seem to be able to make sense of information to arrive at an opinion even when the information was inconsistent.

Moreover, their level of knowledge increased when exposed to conflicting information than consistent information.

When asked how they would try to cope with conflicting information, these were the responses she obtained (She didn’t indicate what her X axis represented; I assume it’s the number of participants):

I thought it was interesting that four out of the six coping strategies named by her subjects (shown in blue) were of a form that could be described as seeking out an opinion leader. This appears to dovetail with the two-step flow theory.

Taken together, her study seems to suggest that in an election campaign, when voters are faced with claims and counter-claims, their responses would be to seek out more diverse (even conflicting) information and to seek out opinion leaders. Both these responses would be consistent with an argument in favour of an important role played by alternative online media in helping people organise their thoughts.

Blogged issues and valence

The earlier paper presented by Natalie Pang was equally interesting. It stemmed from a detailed analysis of blog contents during the election. 833 blog posts from 200 blogsites were reviewed, of which (after dropping those that turned out to be “not relevant, because they were links from the mainstream media, Yahoo, etc”) 764 posts were coded for factual content (topic) and qualitative content (positive / neutral / negative).

The top five topics found in the analysed blogposts were:

1. Governance
2. Candidate’s quality
3. Political system
4. Hearing people’s voices,
5. Housing cost

She remarked that it was interesting how “the top issues were political rather than bread and butter concerns.”

Below is her slide showing how these topics were reflected onto the People’s Action Party (PAP) and opposition parties.

What is meant by “governance”? I suppose it means the way the PAP operates, its arrogance and lack of transparency and accountability. And this is where Paul Wu et al’s presentation needs a mention.

Causality and inter-correlation across media

His results came from an enormous number-crunching exercise: “Close to 1,200 relevant online media sites, including websites belonging to traditional media (e.g. The Straits Times and Channel NewsAsia), and new media, including blogs, forums, Facebook and Twitter (there existed 650 streams with more than 50 tweets) were archived.”

Because Wu et al were searching a different data set from Natalie Pang, they found a different list of major election issues:

Then they did some kind of analysis — all very mysterious to me — and found that causality was all bi-directional, meaning each class of media was influencing what was said  in the other classes.

From my understanding of the talk this was arrived at by looking at something called an “Agenda time series”, picking up when a certain topic first appeared strongly on which platform. One issue (accountability) was particularly interesting in that it was talked about strongly on alternative online media (what in the slide he called “new media”) several days before the mainstream media picked it up.

In that sense, one could say that alternative online media was driving this issue of accountability. As you will recall, this is related to the top issue (governance) that Natalie Pang found in her analysis of  794 blogposts.

* * * * *

While it is not possible to discern any definitive picture as to the role played by alternative media in the general election, what is becoming clear is that there is no simple answer. Alternative and mainstream media mutually influence each other. Consumers of information, far from being turned off by conflicting information, actually progress their thoughts better when exposed to it. In other words, to the degree that the mainstream media tend to present consistent, pro-establishment information, however trustworthy they may be, there is a flight away to alternative media precisely for their difference.

—

ADDENDUM

I received an email from Natalie Pang, with several clarifications. It would help towards a better understanding of her presentation if I presented those clarifications here:

Yawning Bread wrote: “She labelled her outcomes slide “Increased certainty in supporting a party” which didn’t make a lot of sense to me since the topic she visited on her subjects was foreign labour.”

Natalie Pang: To clarify, subjects were asked if the topic of foreign labour was significant enough to make them support a political party. My apologies in the ambiguity of the slide’s title.

Yawning Bread wrote: “You will also notice that as far as my notes go, she didn’t state what outcomes came from participants who initially had uncertain views and been given consistent information.”

Natalie Pang: For the purpose of time, I had presented only significant findings. In other words, there was no significant change for participants who were given consistent information and initially uncertain. This was particularly interesting for me, as for this group, consistent information did not actually reinforce opinions nor increase participants’ engagement/inclination towards any party. Whilst some may argue that it may be that the topic of foreign labour may not be of enough interest – the finding from the other group (those who were given conflicting information, and were also initially uncertain) overruled this possibility as they became more certain that the issue was significant enough to influence support for a party (although there was no significant directions in this certainty).

Yawning Bread wrote: “Nonetheless what was interesting was that participants who were initially uncertain, even when given conflicting information, progressed to greater certainty. People seem to be able to make sense of information to arriive at an opinion even when the information was inconsistent.”

Natalie Pang: I share your observations here. I think we may also speculate from this, that the online electorate will continue to be an opinionated lot and contribute much to political discourse, if not already obvious. How they develop and the extent to which such developments become constructive (and not destructive) is a question of great importance.

Yawning Bread wrote: “When asked how they would try to cope with conflicting information, these were the responses she obtained (She didn’t indicate what her X axis represented; I assume it’s the number of participants)”

Natalie Pang: My apologies again for the lack of clarity. X refers to the number of times a coping strategy was mentioned, as participants can indicate more than one coping strategy.

Yawning Bread wrote: “I thought it was interesting that four out of the six coping strategies named by her subjects (shown in blue) were of a form that could be described as seeking out an opinion leader…Taken together, her study seems to suggest that in an election campaign, when voters are faced with claims and counter-claims, their responses would be to seek out more diverse (even conflicting) information and to seek out opinion leaders. Both these responses would be consistent with an argument in favour of an important role played by alternative online media in helping people organise their thoughts.”

Natalie Pang: I did not see the results in the way you did – I saw them as part of another theory from philosophy, which theorized how people who deal with conflicting information. Seeking information from others (personal networks), experts, or participate in activities with the goal of making personal observations were all part of that theory. Your perspective is a fresh one for me – thanks for that. I wonder if I can quote you on this?

And I like your conclusion here:

Yawning Bread wrote: “While it is not possible to discern any definitive picture as to the role played by alternative media in the general election, what is becoming clear is that there is no simple answer. Alternative and mainstream media mutually influence each other. Consumers of information, far from being turned off by conflicting information, actually progress their thoughts better when exposed to it. In other words, to the degree that the mainstream media tend to present consistent, pro-establishment information, however trustworthy they may be, there is a flight away to alternative media precisely for their difference.”

Natalie Pang: I agree with the point that no media platform should be looked at independently. Media and information should (and needs to) be looked at as a fluid system. They are recursive, very much like how we cannot separate our limbs from our senses. People consume information coming from various media channels/platforms as a whole, rather than singling them out based on the platform(s) they originate from. In a sense, this is also because media/communication channels are increasingly integrative. We read news from mainstream media off our mobile phones, and are able to share + comment on it using social media (such as Facebook and Twitter) etc – the boundary between mainstream and alternative media is already blurred and I think it will continue in that fashion with increased use of technological devices and a very connected/online nation.