Conversation The Forum
Helper icon Helpers: Chris2mates , LLstill , PrincessFruitBat


About us

Midsummer's Eve is a free online dating community - based around friendship, real meetups, real people, and real relationships. We've been online since 1999 and have twice won Radio 2's Web Site of the Day award. So why not join us for free and join in the discussion?

What is good science?

What is bad science?

1 2 3 4 5 Next   Last  

Jeff

Jeff  Male  East Sussex
15-Jan-2013 12:03 Message #4275585
Rather than other conversations getting bogged down in good science and bad science, how about this conversation to discuss this subject.

How can scientists, and we ourselves, try to eliminate personal bias when seeking truths outside ourselves
?
What are examples of bad practice in such investigations?

I'm not referring to good uses such as medicine and bad uses such as nuclear weapons, I'm referring to seeking truths that are factual and can be checked by other people, such as in physics and biology.

For example, there are various methods in which it needs only 2 people to prove/disprove astrological predictions without personal bias. Details on request.
Minnie-the-Minx

Minnie-the-Minx  Female  Hertfordshire
15-Jan-2013 13:08 Message #4275630
That rather assumes that all truths are factual.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 13:19 Message #4275638
No it doesn't. It's just the type of truth that I would like this conversation to be about.
This is intended to be about truths outside our heads, that other people can check, and often can proved to be true or false.
Other kinds of truth are discussed in other conversations.
Minnie-the-Minx

Minnie-the-Minx  Female  Hertfordshire
15-Jan-2013 13:27 Message #4275642
Almost all information is derivative, in its own way, unless you are looking at the actual raw numbers themselves. Even then, you have to be careful because even the way that data are collected introduces bias.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 13:32 Message #4275644
So what are some methods that can reduce or eliminate personal bias?
Jeff

Jeff  Male  East Sussex
15-Jan-2013 13:43 Message #4275651
GH,
It seems that you got that quote from www.modern-thinker.co.uk/1a%20-%20Reason%20and%20Intuition.htm by Ian Heath.

That article includes:-
"Insight is an inference that is validated by reason.
Intuition is an inference that is validated by the thinker’s belief systems.
...
In my view there are only two specific criteria of truth : insight and intuition."


I'm referring to the world outside our heads.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 14:13 Message #4275662
One's head/brain/mind/thoughts/... are necessary. But they are not sufficient.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 14:31 Message #4275671
GH,

Thanks for your thoughts. But they aren't in response to the subject which is about trying to eliminate personal bias in determining facts in sciences such as physics & biology.
Perhaps you can suggest some good and/or bad methods in science.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 14:55 Message #4275686
Hint: Double-blind experiments.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 15:02 Message #4275692
Further hint: Randomised.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 15:17 Message #4275703
"may i request these details please ??"

I will answer that request in detail. But first I would like us all to understand topics such as randomised double-blind methods, and whether we agree that they can eliminate bias. On request I'll explain that technique with examples, but someone else may like to do that first.

We can then apply suitable techniques to other areas such as astrology, homeopathy, mind reading, etc.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 15:50 Message #4275723
I'm not a genius. They are well-known techniques.

Randomised double-blind methods can also remove the bias from determining the difference between the effectiveness of drugs and placeboes, which cola tastes best, whether certain after-death experiences are real, and so on.

In www.midsummerseve.com/fora/thread.aspx?threadid=168445&page=3 ("The AFTER LIFE and other mysteries"), I would add to my suggested experiment of 11-Jun-12 at 17:01 a further randomisation, that the independent judges should see a mixture of pictures "seen" by near-dead people and other pictures on a random basis, so that the independent judges can themselves be checked for bias.
Notice that I am not dismissing off-hand such visions by nearly-dead people, but I am suggesting experiments to see whether they can be checked without bias.

If you were asked to test a few different fertilisers on plants, and your research was sponsored by the manufacturer of one of the fertilisers, how would you avoid personal bias?

I will answer your questions in due course. But I would prefer more people to contribute first.
Celt

Celt  Male  Ceredigion
15-Jan-2013 16:20 Message #4275743
Bad science is believing that an outside force will save us from ourselves, rather than biting the bullet and applying all of our talents to sort out our problems ourselves.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 16:33 Message #4275746
I hope that GH, or someone else, has a go at describing clearly to everybody the exact method you would use to test say 3 fertilisers, so that you cannot even be suspected of bias. Assume that they are all for indoor tomato plants, and you have a plentiful supply of those plants and of the fertilisers, and containers having various temperatures, humidities, lights etc which you can control on timers or however you wish. If you like, you can also use an assistant, pencil & paper, and a fair dice (showing 1 to 6 randomly)!

Then see if anybody on MSE can accuse you of being biased in determining which fertiliser is the best.

I'll now shut up for a while.

Minnie-the-Minx

Minnie-the-Minx  Female  Hertfordshire
15-Jan-2013 17:38 Message #4275791
Wil you mark us A+, if we get it right?
Darcyman

Darcyman  Male  East Yorkshire
15-Jan-2013 17:54 Message #4275804
When all you have is a hammer every problem looks like a nail.
Somebody pass me the hammer, come here Jeff. lol
brisinger

brisinger  Male  Lancashire
15-Jan-2013 17:59 Message #4275810
I'll have to read this thread to cheer me up after being in hospital today. I did notice, however, SM that you said that you trained in Biology. I'm curious on what branch did you specialize in?
Jeff

Jeff  Male  East Sussex
15-Jan-2013 18:07 Message #4275814
Everybody will mark it A+, and if you use it in your future thinking about science then they will say that you are a great scientist, GH.

If it's too difficult, you could try either of these problems to start:- (Again you have an assistant, a dice, pencil and paper.)

(a) You are an honest doctor specialising in various types of headaches in many patients, and a drugs company offers you a lot of money to test their new drug which claims to be much better than Ibuprofen. How do you test it in such a way that even if you find that the drug meets its claims, you cannot be accused of bias because the company is paying you so much?

(b) You are a market researcher and the Likea Cola company is paying you a lot of money to find out if more people prefer Likea Cola compared to Coca Cola. You take many samples of both onto the pavements, and using paper cups you give a sample of each cola to passers-by and note their preference. If the cups are labelled "Likea Cola" and "Coca Cola", then you can be accused of using a more pleasant tone of voice or facial expression or handing-over posture etc for the Likea Cola which might influence the taster. How do you avoid such possible accusations of bias?
brisinger

brisinger  Male  Lancashire
15-Jan-2013 18:20 Message #4275826
Oh no, religion isn't being dragged into it again is it. I thought it was about science and wisdom, not another destined for Room 101...
brisinger

brisinger  Male  Lancashire
15-Jan-2013 18:35 Message #4275840
"test say 3 fertilisers"

It's funny you should choose that, Jeff, but I've been involved in a very similar experiment with Oxford University called the Big Biochar Experiment.

You can find it at: http://www.bigbiocharexperiment.co.uk

It's only in it's preliminary stages as yet but they asked for volunteers to test Biochar against a number of different options to see which one would produce the best crops. I chose to get involved in it actually. It's been really interesting up to now and is well worth keeping up with.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 18:45 Message #4275857
Brisinger, you should get GH to help you. It might be the branch in biology that she specialised in.

Minnie, sorry for confusing you with GH. She's the nice one!
wonderoushen

wonderoushen  Female  Gwynedd
15-Jan-2013 18:50 Message #4275860
Who's the nice one Minnie or GH?
Bo-nio

Bo-nio  Female  Cheshire
15-Jan-2013 18:50 Message #4275861
Re scenarios, any robust research uses a 'double-blind' methodolaogy where the researchers handing out the drugs/ samples etc don't know themselves which is the placebo/ competitive product etc

To establish good/ bad research you need to know the methodolgy and subject that to scrutiny.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 18:55 Message #4275870
GH is the nice one. Minnie is like she was in the Beano.
Jeff

Jeff  Male  East Sussex
15-Jan-2013 19:06 Message #4275883
I'm hoping that someone will give simple clear instructions for a randomised double-blind experiment, such as one of my examples, that non-scientists can understand and appreciate, and then challenge anyone to say how it can be biased. My purpose is to show that science without personal bias is possible, even if there is big money trying to influence it.

(In the fertilisers example, the experiment can be simplified by assuming that they all look the same and require the same dose.)

1 2 3 4 5 Next   Last  


 Back to top

 Help with conversations