Guest Post: Post-Factual Perceptions of Weather

Adversarial interactions between physical and social scientists are sometimes seen around this blog, so I’m happy to report on something different. The occasion is a new paper in the 50th anniversary issue of Sociology, flagship journal of the British Sociological Association. The first author is a sociologist, the second a climatologist, the third a geographer, then two more sociologists.

Flood realities, perceptions and the depth of divisions on climate” employs both physical and social data, as its title suggests. It focuses on the northeastern US state of New Hampshire, where destructive flooding increased in the past decade, for reasons partly related to climate. Figure 1 depicts this increase in terms of disaster expenditures and newspaper reports, but our paper also presents river flow and precipitation data that tell the same story.

flood_fig3p

Figure 1

Might this increase have been noticed by the public, especially in the most affected areas, overshadowing people’s ideological beliefs about climate change? That hopeful hypothesis, posed by climatologist Cameron Wake, is what launched our study. But statewide surveys in summer and fall 2015, involving more than 2,000 telephone interviews, found surprisingly low public awareness. We asked,

Comparing the past 10 years with 20 or 30 years ago, do you think that number and size of destructive floods in New Hampshire have increased, decreased or stayed about the same?

Over the next few decades, do you think that number and size of destructive floods in New Hampshire are likely to increase, decrease or stay about the same?

Figure 2 graphs responses, along with the observed frequency of extreme precipitation events for the decades in question. Only 35 percent of our respondents thought (correctly) that floods in the past decade had increased.

flood_bars1

Figure 2

Moreover, careful analysis found no indication that either past or future flood responses are systematically higher in counties that have been most affected, or are most vulnerable. Disappointingly, responses to the flood questions instead exhibit strong ideological gradients, all too familiar for just about anything related to climate. Figure 3 depicts these gradients for our two flood questions and also a standard question about climate change, which correlates so closely with ideology that it could provide an alternative measure.

flood_bars2

Figure 3

These results on New Hampshire flooding are, unfortunately, not a fluke. Another recent study, also with an interdisciplinary team, focused on northeast Oregon (“Wildfire, climate, and perceptions in northeast Oregon”). In that rural region, summer warming has been accompanied by a dangerous rise in wildfires. We found a similar pattern: public awareness of the warming trend follows a party-line gradient, and is mostly unrelated to experiential factors such as long-term residence or ownership of forest land.

Both of these studies involve decadal trends, however. To learn whether similar patterns affect recollections of recent weather, without mentioning trends, New Hampshire State Climatologist Mary Stampone and I carried out an experiment described in “Was December warm?” December of 2015 had been the state’s warmest by far, for that month or any other (in anomaly terms) looking back more than a century. The whole winter of 2015–16 set historical records as well. Asked about December on a statewide survey in February, or about the winter on a second survey in April, most people correctly said temperatures had been above average. But this accuracy was significantly higher among Democrats, and those who think humans are changing the climate.

We repeated our question a third time on a survey in July. The election campaign was by then in full swing, so we can make a more topical comparison. Trump and Clinton supporters stand 21 points apart in their recollections about the record-setting winter (Figure 4). Political divisions seen earlier had apparently widened as winter became a few months more distant.

trump_winter

Figure 4

While political resistance to climate science in the US appears intractable at times, some observers hope that nature will intervene more persuasively, giving inescapable signs of a problem. For scientists, it has done so already, but many people do not take their word for it, or yet see the problems themselves. That ideology or partisan social identity filters public perceptions about disasters and even mundane weather so powerfully suggests that this process, if it happens, won’t be quick.

Advertisements
This entry was posted in Uncategorized. Bookmark the permalink.

37 Responses to Guest Post: Post-Factual Perceptions of Weather

  1. People experienced a flood, but I guess not many or not anyone experienced a statistically significant increase in floods in their community. The increase is probably only clear state-wide and that information comes from scientists via the media.

    If there were a relationship then it would likely have been because people who experienced a flood informed themselves. Did you also ask knowledge questions? Would be interesting to see if that relationship is there.

    If there a near 100% correlation between the question on past and future increases in flooding or are it different people who make almost the same percentages? Would be curious if there are people who do not see a trend, but expect one in future because There’s Physics.

  2. izen says:

    So much for the ‘wisdom of crowds’.

  3. L Hamilton says:

    Victor:
    “People experienced a flood, but I guess not many or not anyone experienced a statistically significant increase in floods in their community.”
    To some extent I agree, although it’s surprising how few people answer such questions “I don’t know.” Also, accuracy does not improve with such things as age, years resident in the community, or ownership of forest land (in the Oregon data), which you’d think could be other ways besides statistics to have a long perspective.

    Anyway, for a shorter perspective we asked the “December” and “Winter” questions, starting just a few weeks after the season in question had ended. In New Hampshire, winters are an in-your-face kind of thing, which can complicate everything about daily life (or conversely, improve the skiing and so forth). Christmas day often is snowy here, but in 2015 temperatures exceeded 60 F in many parts of the state — that’s unusual! Winter recreation and tourism, significant in this state, were suffering. In spring, flowers bloomed and trees budded early. So it’s pretty striking to me as a resident, to see a 21-point political gap in what people recall.

    Warmer than average?

  4. Ken Fabian says:

    I can see that if the most recent winter or year or decade is, locally, full of “exceptional” and record breaking weather events, especially if they are damaging and in line with expectations for climate change, that they may persuade some of the unsure that climate change is significant. But, if the decade – or most recent year or winter – was not exceptional or was below trend the persuasive impact would go the other way. Too much of the rhetoric of climate science denial uses too small samples, too short periods – when it isn’t attempting to make current climate change seem unexceptional by invoking too long periods and “geological” time. Too narrow a viewpoint, or too broad can both cause misunderstanding.

    Politics and public perceptions do matter of course and advocates making use of political opportunities extreme weather events present is going to continue. The recent power blackouts in South Australia due to storm damage is being cynically used by Australian conservatives to criticise the state based efforts towards a low emissions transition – attempting, with considerable success, to separate discussion of energy policy from any climate considerations by indulging in renewable energy alarmism. The way they went into a broad and seemingly co-ordinated attack suggests that South Australia’s high RE component was on their radar awaiting the most effective timing – wholesale price spikes provoked a recent previous attempt to discredit RE but didn’t appear to be effective, perhaps because the main culprit was gas generation. This time the culprit was extreme weather, not RE, but that hasn’t stopped them.

    Discussion of changes to the likelihood of such weather events under climate change is outside the range of an appallingly bad mainstream media’s energy policy scrutiny and the Federal Opposition may like to score a few points on the government but don’t have a well thought climate policy alternative or real credibility to fall back on. Of course this is a government that “officially” accepts the validity of science on climate and supposedly intends to commit to the Paris agreement, but it’s hard not to be cynical and think the official position is merely an expedient way to avoid real scrutiny and real debate that would reveal the paucity of commitment to a low emissions transition. I don’t know that Australian politics leads the way in this kind of deceptive and misleading politicking but they do appear to keep getting away with it.

  5. Since your post mentions adversarial interactions between physical and social scientists 🙂 I’ll pose a question that might illustrate why this sometimes occurs. What are the implications of this result for how we should communicate science? Does it mean that we should take cultural aspects into account when doing so, or not? If the goal of a particular communication attempt is to actually convince people of some scientific finding, then it would seem that we should. However, most scientists regard research (and the results of research) as value neutral – or, at least, that we should aim for it be value neutral. Consequently, I think that most physical scientists, at least, regard science communication as something that should also be value neutral. In my view, science communication is simply about communication scientific findings to the public and to policy makers, not to necessarily convince them of something.

    So, it’s certainly my view that some of the adversarial interactions between physical and social scientists is driven by a confusion as to the goals of communication. Social scientists sometimes seem to suggest that these silly physical scientists still don’t understand the significance of cultural factors in determining what some are willing to accept. However, they themselves don’t seem to maybe consider that physical scientists do indeed understand this but just don’t see it as their role to break down these cultural barriers. If anything, some might be very uncomfortable thinking about how to communicate in a way that will be convincing to those who might be pre-disposed to reject some science, as that would seem to be an exercise in marketing and scientists aren’t really trying to sell something.

  6. Dikran Marsupial says:

    Cool, genuine collaboration between academics is often very productive. Not only do you get new insights by seeing things a different way, but also it is difficult to really master more than one field of research.

    “Adversarial interactions between physical and social scientists are sometimes seen around this blog” which ought to be considered decidedly “vieux chapeau” IMHO. The difficulties of social and physical sciences are different (especially the awkwardness of their subjects) so they have different characters and methods, but neither is superior to the other. I was very impressed by the elegance of the experiments in Kahneman’s “Thinking, fast and slow”, for example. It is a bit like the “two cultures” banter between humanities and sciences, which seems mostly the result of Dunning-Kruger based underestimation of the sophistication of the other camp, just an aspect of human nature, e.g.

    “Once or twice I have been provoked and have asked the company how many of them could describe the Second Law of Thermodynamics. The response was cold: it was also negative. Yet I was asking something which is the scientific equivalent of: Have you read a work of Shakespeare’s?”

    Personally I’d say that is more like “have you understood a work of Shakespeare’s?”, you can quote “to be or not to be” and not know what Othello meant by it. ;o)

  7. Pingback: Da Montreal a Kigali - Ocasapiens - Blog - Repubblica.it

  8. Michael 2 says:

    “giving inescapable signs of a problem.”

    Or inescapable signs of a non-problem. “Problem” is in the eye of the beholder. Science cannot say whether a thing is a problem, it can only describe the thing.

    This relates to ATTP’s followup question of whether science communication should be value neutral. I tend to agree that it should; but this does not mean it will be seen as value neutral by the readers/hearers of it.

    Where you see such huge differences in recollections I interpret that in fact most people did not find the past winter exceptional and merely repeat what they’ve heard from their preferred sources of information which will most likely be biased. In other words, this much difference should itself be an indicator of unreliable personal recollection, which itself is an indicator that the past winter was largely unremarkable.

  9. BBD says:

    M2

    Ooh, selective quotation! Let’s have the context (my bold):

    While political resistance to climate science in the US appears intractable at times, some observers hope that nature will intervene more persuasively, giving inescapable signs of a problem. For scientists, it has done so already, but many people do not take their word for it, or yet see the problems themselves. That ideology or partisan social identity filters public perceptions about disasters and even mundane weather so powerfully suggests that this process, if it happens, won’t be quick.

    Thanks, M2, for the demonstration of ideology or partisan social identity filtering:

    which itself is an indicator that the past winter was largely unremarkable.

  10. L Hamilton says:

    ATTP:
    “What are the implications of this result for how we should communicate science?”

    Key question! A few thoughts…

    One thing: The traditional paths for disseminating research results — peer reviewed papers and conference presentations — are necessary but not sufficient in this rapidly spinning world. Many researchers recognize this, inspiring diverse outreach from statements to videos, op-eds or blogging. That’s terrific and there needs to be more, although it raises workloads and hassles.

    Second thing: Cumulatively, the diverse efforts are working. That is, public acceptance of the reality of anthropogenic climate change, and awareness that scientists agree on this point, have been drifting upwards (paper coming soon). With frustrating slowness because many people vote by social identity rather than issues, so shifts in opinion on an issue may not translate into electoral or policy movement. Also, people favor symbolic actions more than costly ones.

    Third: Scientists need communication allies, and there are many capable citizens who are glad to be engaged. They are empowered when scientists participate in public discussions, post clear and rapid-response information, or make data and articles publicly accessible.

    Fourth: It’s very difficult to change the minds of true believers, so that can’t be a main goal. Science advocates who manifestly share the social identity of an audience may have better luck with this than “outsiders” seen to represent some other group.

  11. Larry,
    Thanks. It sounds like, overall, you’re suggesting that it’s difficult, but it is working (if slowly) and that help from other sources also makes a difference? Certainly gels with my impression. I think science communication can be difficult and expecting noticeable results in a short space of time is probably unrealistic. So, it’s not so much doing anything different, but just continuing as we are and accepting help if offered.

  12. L Hamilton says:

    Yes, there seems to be a gradual rise in public acceptance of anthropogenic climate change, and of the existence of a scientific consensus — on the order of 10 points over the years (since 2010) I’ve been watching. There were no sharp jumps coinciding with Hurricane Sandy, the Pope’s statement, IPCC AR5, or other obvious events. Public understanding remains somewhat limited. Consequently, it seems plausible that acceptance of ACC is growing due to rising awareness that there *is* a consensus among scientists, which in turn reflects cumulative impacts from the diverse science communication efforts I mentioned above.

    Here’s an image lifted from the forthcoming paper:

  13. Eli Rabett says:

    Larry, curious if you separated out the guys that tap maple trees something where there are a) a lot of folk in NH and b) subject to seasonal changes in the winter and early spring.

  14. Larry Hamilton, is the green line of the right panel more variable because of larger sampling errors (less moderates) or do these people actually have a more open mind and change their position more often?

  15. izen says:

    Given the human propensity for pessimism I wonder what the answers would be if the questions about more flooding/fires were ask of a population that had experienced no change in these metrics.

    While it looks from some of the results that ideological position causes conservatives to wrongly think things are better than they are, perhaps a moderate or liberal would think things are worse than they really are. Conservative more optimistic than liberals?

    Both sides may suffer from a Morton’s demon that prevents them gaining an accurate picture of reality.

  16. L Hamilton says:

    Eli:
    “curious if you separated out the guys that tap maple trees”

    That’s definitely an issue regarding warming winters in New England generally, and this past winter in particular. Here’s one news story:
    http://www.climatecentral.org/news/climate-change-maple-syrup-20178

    More detail in the Northeastern Climate Impacts Assessment, available here:
    http://www.ucsusa.org/global_warming/science_and_impacts/impacts/northeast-climate-impacts.html#.V_zoz8kpdps

    Syrup producers are too small a demographic to resolve on surveys, but their plight makes the news. As do other segments involved with winter tourism (skiing, snowmobiling etc.).

  17. L Hamilton says:

    Victor:
    “is the green line of the right panel more variable because of larger sampling errors (less moderates) or do these people actually have a more open mind and change their position more often?”

    Some of both perhaps, but sampling error plays a key part. Most of the individual New Hampshire surveys in the (B) panel above are about 500 interviews, of which “moderates” form the smallest group — around 15% overall, whereas other groups are around 20%. So really none of these individual-survey ideology groups are large, hence the wilder swings in panel (B) compared with the whole-sample results of the main line in (A).

  18. L Hamilton says:

    izen:
    “Given the human propensity for pessimism I wonder what the answers would be if the questions about more flooding/fires were ask of a population that had experienced no change in these metrics.”

    Haven’t tried that but Cameron and I did have another question that kind of gets at your idea. On one survey we asked New Hampshire residents whether they thought the area covered by forests now in this state is greater, less or about the same as it was a hundred years ago. About half took a pessimistic view, which happens to be wrong — a century ago, most of the state was deforested (hence those stone fences you come across deep in the woods today).

  19. L Hamilton says:

    izen:
    “While it looks from some of the results that ideological position causes conservatives to wrongly think things are better than they are, perhaps a moderate or liberal would think things are worse than they really are. Conservative more optimistic than liberals?”

    Following up on your other point with respect to my forest question above, there at least the answer seems to be no. Conservatives & liberals, or Democrats & Tea Partiers, were equally prone to get the forest-cover question wrong.

  20. Roger Jones says:

    Thanks Larry – that paper looks really interesting and relevant

  21. Willard says:

    Speaking of post-facticity, here’s the greatest:

    I have no idea how 10% of Clinton fans trust Fox, but so be it.

  22. Steven Mosher says:

    “It focuses on the northeastern US state of New Hampshire, where destructive flooding increased in the past decade, for reasons partly related to climate.”

    And I’ll note that the oregon study also took place against a background of increased disasters.

    I would think the opposite kind of study would be very interesting

    That is, where there hasnt been an increase or where there has been a decrease in disasters,
    what will the answers look like.

    Example: But for Maththew it would have been interesting to talk to floridians about hurricanes.

    the point would be to also look for folks being “alarmed” into thinking things are worse than they are

  23. L Hamilton says:

    Willard:
    “Speaking of post-facticity, here’s the greatest”

    And speaking of post-facticity, Tom Safford and I have a new topic (survey in progress) — Zika virus. That appears to be politicized too, although we don’t yet know what to say about it. Stay tuned.

  24. L Hamilton says:

    Steven Mosher:
    “the point would be to also look for folks being “alarmed” into thinking things are worse than they are”

    The New Hampshire forest cover question I mentioned upthread might be the closest thing I have in this vein, and you’re right, most people think things are worse than they are (in this case, most think NH forests have declined in past century, whereas they are actually more extensive now).

    But, unlike the clear gradient for everything climate-related, that misperception exhibits an odd partisan breakdown — pessimism strongest at *both* ends of the political spectrum. I hadn’t closely looked at this until now.

  25. Willard says:

    > Stay tuned.

    I think you should redo your “who do you trust” graph with blue and orange, L.

  26. Steven Mosher says:

    “But, unlike the clear gradient for everything climate-related, that misperception exhibits an odd partisan breakdown — pessimism strongest at *both* ends of the political spectrum. I hadn’t closely looked at this until now.”

    I think just for methodological reasons it would be interesting to always “pair” the questions you
    ask, such that if you are looking at post factual perceptions of weather you would ask one of each type of question… not sure how you do that.. but I think it might be enlightening.

    anyways.. always enjoy your work

  27. Joshua says:

    Lawrence –

    ==> Zika virus. That appears to be politicized too, ==>

    It would be fun to see information on the politicization around Ebola. One of the most amusing phenomena I’ve seen recently was when righwingers, who are putatively concerned about “government overreach” got on the Trump train to ban immigrants, quarantine nurses, etc. to address Ebola, despite input from the “expert” community that such measures would be ineffective AND counterproductive.

  28. izen says:

    @-“But, unlike the clear gradient for everything climate-related, that misperception exhibits an odd partisan breakdown — pessimism strongest at *both* ends of the political spectrum. I hadn’t closely looked at this until now.”

    With the caveat it is speculative psychobabble…
    The NH forest results look as though having a strong partisan ideology that causes you to believe the world can be a better place in the future has as a consequence that you perceive the present as worse than it really is.

    So if you think things would be better if Bernie Saunders/Donald Trump was in charge, then you will also think the present is more corrupt and closer to collapse than is warranted.

  29. izen says:

    @-Joshua
    “…ban immigrants, quarantine nurses, etc. to address Ebola, despite input from the “expert” community that such measures would be ineffective AND counterproductive.”

    While that response may be more prevalent amongst authoritarians, it is a a widespread reaction to a disease epidemic. Historical examples abound.
    While the nature of a disease and the treatment available may make such actions counterproductive, in cases where the primary vector is human and there is little effective treatment, exclusion and quarantine is not an inappropriate response.

  30. Willard says:

    Still no orange:

  31. L Hamilton says:

    With apologies, I misread my own graphs in the forest/partisan plots above. What they show is not pessimism strongest at both ends of the spectrum, but *realism* (forests greater) at both ends. It’s still not very strong (<40%). A better way to put this is that pessimism or just inaccuracy is strongest in the middle. Independent or undecided voters often are, by other criteria, a relatively low-information group.

    Willard, I've got some blue/orange graphs now on your inspiration, may find a use for them soon.

  32. Joshua says:

    izen -min
    ==> While that response may be more prevalent amongst authoritarians, it is a a widespread reaction to a disease epidemic. ==>

    I’m not entirely sure what your point is. Mine was that the reaction to Ebola was highly politicized, with the call for immigration bans and quarantining dominated by Republicans, as a ploy to exploit
    Ebola and polarization for political expediency, and at the expense of effective way to address the threat.

  33. izen says:

    @-“A better way to put this is that pessimism or just inaccuracy is strongest in the middle. Independent or undecided voters often are, by other criteria, a relatively low-information group.”

    Another beautiful theory ( speculative phsychobabble ) destroyed by ugly fact!

  34. Mal Adapted says:

    Speaking again of post-facticity, I propose to name the second 10 years of the 21st century “the post-factual decade” in dishonor of Karl”Turdblossom” Rove.

  35. Pingback: Week in review – science edition | Climate Etc.

  36. Pingback: 2016: A year in blogging | …and Then There's Physics

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s