The Dunning-Kruger effect

One of the biggest challenges I’ve faced in communicating science (apart from my own struggles to explain things clearly without losing the interest of my audience), is the Dunning-Kruger effect, which Wikipedia gleefully defines as

cognitive bias in which an unskilled person makes poor decisions and reaches erroneous conclusions, but their incompetence denies them the metacognitive ability to realize their mistakes.

It leads to stupid people thinking they are way smarter than they actually are and to smart people thinking they aren’t as smart as they actually are.  I’ve wanted to rant write about this effect for awhile now, but Barbara beat me to it.  And lucky for you, too – she does it better than I ever could have, even though we’ve had many of the same experiences, like

There was a guy on my favorite online dating website that answered the question, “What do you think about healing crystals?” with the answer, “I think they work.” But he claimed to be an intellectual and he only wanted to meet women “if their smart.” He was online, so I started a chat with him. I opened with a remark that he’d attract more smart women if he knew the difference in “their” and “they’re.” Yet he kept talking to me. Turns out he believes in astrology, thinks the government has a secret 20 year plan to try to take away all rural land ownership to force people to live in cities, and he thinks global warming is a hoax to increase taxes. He honestly believes he is smarter than everybody else and wants to meet a smart woman.

Funny, and so true. Go read the rest!

Communicating Science

I went to DC this weekend for a really awesome workshop on using GIS as a tool to communicate science to non-scientists.  I’m pretty excited about giving my next presentation now – it’s amazing how much more information gets across when you use colors on a map instead of a bar chart.

Most of the workshop was inside a convention center, but we did get a little time to explore.  DC is pretty in the fall.

At Rock Creek Park

Another Berry Go Round!

October’s Berry Go Round was posted last week and I only just now noticed.  A lucky turn of events since it was a fantastic distraction from today’s impending  migraine.  I had three favorites in this round of the carnival.  The first is on flowers that parasitize fungi complete with fantastic pictures of these weird, weird plants.  I’d also recommend this post on British soldier lichen and this one on a fringed orchid in the same genus I saw this summer, but a way cooler color.

Redirect – Goddess Edition

Remember that grant proposal I was freaking out about last week? I submitted it with plenty of time to spare and started on another one.  The one I’m working on now is kind of a big deal.  So far, things are going slowly because I’m so darn intimidated.

Luckily, the blog of Dr. Isis, the domestic and laboratory goddess is a confidence booster/kick-in-the-ass, especially her many advice columns.  I think my absolute favorite post by Dr. Isis is Trampy Toys and What We Tell Our Daughters, but I also love her comments on the day to day life of a scientist who also happens to have a vagina.

Figures Lie and Liars Figure – Climate Change in 2007

My aunt has a hard time believing that climate change could be occurring. Her questions and concerns aren’t that different from many Americans’, so I’ve been responding to her here in a series called Figures Lie and Liars Figure.  You probably thought I was all done since my last post was more than a month ago.

You obviously don’t know my aunt.  She had quite a few questions after looking at the International Geosphere-Biosphere Programme Climate Change Index and a slideshow of climate change indicators from the EPA.

The IGBP uses four indicators of climate change – carbon dioxide levels, global average temperature, sea level, and arctic sea-ice cover – to build an index of climate change. Here’s how it works:

Each parameter is normalised between -100 and +100. Zero is no annual change. One hundred is the maximum-recorded annual change since 1980. The normalised parameters are averaged. This gives the index for the year. The value for each year is added to that of the previous year to show the cumulative effect of annual change.

So, higher global average temperatures and sea level and lower arctic sea-ice cover INCREASE the climate change index.  The climate change index isn’t using any new data or fancy techniques to analyze it.  It’s not part of a scientific analysis trying to prove or disprove climate change.  It was developed by scientists to try to communicate that climate change is happening and it’s happening fast.  Based on my aunt’s response, it’s not entirely successful.

I’m showing graphs of three of the indicators here so you can follow along with my aunt’s question.  If you want to see bigger graphs, just click to go to the IGBP site.

Three of the indicators used in the IGBP climate change index. Click to see larger graphs at the IGBP

Based those graphs my aunt had this to say:

The slides show Global Average Temperatures ~2007 down significantly, Arctic Ice increased and Sea Level down.

In the context of her email, she’s saying that that because the trend in these three indicators seems to reverse in one year, the case for climate change is weakened.

My aunt is right about sea level – it did go down from 2006 to 2007.  But arctic sea ice cover actually decreased from 2006 to 2007 and temperature was actually up in 2007 compared to 2006.  So, overall the climate change index from the IGBP goes up in 2007.

It’s a bit hard to see exactly which year corresponds to which point on the graphs provided by the IGBP – my aunt did say approximately 2007.  So let’s look at 2005 to 2006 – both temperature and arctic sea-ice cover changed in the direction opposite the trend we’d expect with climate change, but sea level went up in that year and the index stayed positive.

What about years where the index really does fall?  Those reversals were probably because of volcanic activity which dumps particulates into the atmosphere that reflect sunlight and help to cool things down:

The index dips in just three years, 1982, 1992 and 1996 . . . . The dip in the curve in 1992 may have been caused by the massive Mount Pinatubo volcanic eruption in the Philippines in 1991. The eruption was large enough to affect temperature and sea level on a planetary scale. The other falls coincide with the El Chichon volcanic eruption in Mexico in 1982 and the volcanic eruption on the Caribbean island of Montserrat in 1996.

But even if we didn’t have an explanation for those reversals, looking at the change from one year to the next isn’t especially helpful when we’re thinking about climate change.  That’s a bit like saying that Phoenix, Arizona is snowier than Buffalo, New York since on December 6, 1998 it snowed in Phoenix but not in Buffalo.

What we’re really interested in is long term trends.  Think about it from a farmer’s perspective. If you have a drought in one year, it sucks, but you keep farming. If you have a drought for 10 years, you’ll probably lose the farm.  That is, one unusual year doesn’t make a place bad for farming, but a long series of them might.  This is what’s happened to Australian farmers.

That’s how scientists look at these climate change trends.   Year to year, these indicators go up and down or change at different rates and looking at just one of them doesn’t give us the whole picture.  As I tried to make clear in my previous posts, it’s the combination of indicators and their trends, not any single kind of data, that convinces us climate change is happening. If there were just a few years where this index increased, or all the increases were offset by decreases in other years, scientists like me wouldn’t be spending their Saturday afternoons trying to convince their relatives that the world is in big trouble if we don’t do something about climate change.

—–

I was trying to explain this concept of long term trends versus short term fluctuations to another aunt of mine who helps people invest their money. She got it right away and gave me a great analogy from the stock market.  Unfortunately, I understand too little about the stock market to reproduce it here, but I’ll try to get her to explain it to me again for your benefit.