Posts by Bart Janssen
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Hard News: Word of the Year 2014: The Vote, in reply to B Jones,
I think it just has to be a big word or concept for 2014. Ebola’s a good example – it’s been around in the 70s but took off, I hate to say virally, following the outbreak this year in West Africa. What makes it a WOTY contender for me is that it’s moved out of its original meaning and drifted into metaphor – there were a handful of examples of media commentators using it to mean something big and nasty, eg “it’s the Ebola of politics”.
My problem with Ebola is that it has killed only a few thousand this year. By contrast AIDS is killing 10s of thousands a year and nobody quite knows how many ISIS/ISIL has killed. For me there is a huge disconnect between the fear uncertainty and doubt spread around ebola and the actual damage it is doing.
But I guess that is sometimes the thing about words - sometimes they gather more momentum than you might expect.
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
Bart thought that it was probably scientific already to the weak extent that any economic claim can be
More importantly without actually reading the methods and being knowledgeable enough in that field I was willing to trust those who were in the field rather than some mere politician who didn't like what was being concluded.
Steve has found experts in the field who have concluded the methodology was bollocks, something he suspected from his expertise and experience in that kind of statistical analysis.
Which is great.
The thing for me is, from outside a field, I am really cautious about finding reasons to distrust conclusions from people in a field. The whole climate change debate has been filled with non-experts mistrusting the experts without ever actually testing their methodology (because in most cases they weren't equipped with the background knowledge to test it).
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
If economics is not scientific enough
I’m pretty hard on economists in general. Particularly those that get jobs as mouthpieces for banks. But I do think there is genuine quality science done in the economics field. I think your questions about this statement from the OECD are fine. But if you are going to go down the path of calling them unscientific then I really think you need to do the work to actually find the original references and do a thorough analysis of the methodology being used.
If after that you conclude their findings are scientifically flawed, go for it with your criticism. But at the moment it feels like you are taking pot shots at this work without any basis for it other than your reacons about cause and effect. It may turn out that you are right and their conclusions are flawed but until you show that then you are essentially doing just what Key and English are do to any scientific result they don’t like – attacking the messenger to cast doubt on the message. I don’t think you mean to do that but it certainly feels that way to me.
-
Speaker: Inequality: Too big to ignore, in reply to Katharine Moody,
and it seems to be totally divorced from the business/economics and/or public policy schools.
Not true. Business courses teach students in the first year all about ethics.
They are taught that ethics has no place in business and should be avoided at all costs.
-
Up Front: Adric and the Art of Asking, in reply to Emma Hart,
I am used to being unable to move, and I enjoy it.
Enjoying it too much during treatment will confuse them.
-
Speaker: Inequality: Too big to ignore, in reply to Matthew Poole,
But what I’m getting at is that it’s never suggested that they will make a change and then rescind it if it doesn’t work.
I think I understand how we ended up with this discussion. You’re confusing political and economic ideology with actual science :P
The issue is scientists are wrong ALL THE TIME. Generally we try and be wrong in private but sometimes ... Our response to being wrong is to say "we were wrong, lets try something else with the aim of being right".
Politicians are also frequently wrong. Their response is to blame the data or the person providing the data.
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
It’s quite possible to not make a prediction about what will be observed, but to just design an experiment that pits some variable against another, make observations, and then guess the hypothesis at some point that isn’t right at the beginning.
Sure. You could do that, if you were independently wealthy and never had to justify your work or apply for funding or publish you results. Or you were blindingly lucky.
But really we actually do make hypotheses.
To be fair we sometimes know that our hypothesis is incomplete or even likely to be wrong. The key is not that the hypothesis be right but that it defines an experiment that is possible and will test the hypothesis.
Note also that hypothesis driven science is not the only type of science, just the only type I do.
It is possible to do discovery science, or exploration science. But even then there is the underlying hypothesis that if you look this way or with that tool you will observe things that are interesting. Things that will enable you to develop hypotheses ...
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
I’m really curious whether, in practice, you make your hypotheses in advance of your experiments.
Fuck yeah! it costs way way too much money, time and emotional energy to just wing it. We spend literally hours with whiteboards drawing models. Hours and hours. We spend ages with journal papers figuring out what other peoples experiments mean for our next experiment. All to get a good hypothesis.
It is ALL about the hypothesis.
ETA: Actually, to be a bit more general, I’m very interested in every bit of light you can shed about how hypotheses come to you.
Ok because I'm in avoidance mode (paper writing)
It really does work the way it is taught in high school.
We* make observations ourselves. We read the literature, which is all about the observations made by other researchers in the field** (and sometimes out of the field).
We then draw models, paper, whiteboards, napkins, shower door, whatever is handy. The models try and incorporate all the observations. If we have enough observations and put together a good model the next experiment is obvious. That is, the model makes a prediction that something will happen when you perturb the system. An arrow leading from A to B suggest if you break A then B will do something, or stop doing something.
Then it's simply (snort) about figuring out how to break A and only A.
* We is a team, at the moment our group is three and a half plus two students plus a collaborator plus anyone else around the place who will help us if we get stuck.**field in this sense means our research subject
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
My aim is more to question how it is that we can’t adopt an attitude in which methodical experimentation is acceptable.
Ok I accept that.
So can you accept the in effect there have already been multiple experiments? Each time a government makes changes to policy that affect both inequality and economic growth an experiment is being performed.
But two problems exist:
1) multiple variables change simultaneously, even with the best will in the world it would be nigh on impossible to design a single variable experiment that examined this question.
2) For the most part no government actually tries to reduce economic growth.So none of the experiments are great, but there have been quite a large number over the last 5 or so decade over all the countries the OECD compares.
-
Speaker: Inequality: Too big to ignore, in reply to BenWilson,
I could very reliably see that it was the opening of the hand that caused the rock to fall to the ground, not the other way around, just from the fact that causation can’t flow backwards in time.
Which is great for a relatively simple system such as a rock and your hand, in the presence of a single large gravitational source, in a non-viscous fluid.
But macroeconomics? No way any experiment is simple enough to eliminate any doubt. Many sepaarte experiments with overlapping conditions can strongly support a hypothesis (that is exactly what the OECD and their economists are saying BTW). But nailing cause and effect level proof, no way.