Null Hypothesis, Astrology and Cargo Sciences

Recently, I came across this particular article in ToI.  After reading the article, I chanced upon the comments section.  I was quite surprised to see that most of the readers have accused Dr. Venkatraman Ramakrishnan of being an idiot, moron, a terrorist, not respecting Indian culture and quite a few have gone berserk with demands to withdraw Nobel Prize.  A couple of comments show surprise by the fact that he doesn’t believe in astrology inspite of being a Nobel Laureate in Signs.    Fine, I made up the last line. 😛

Anyway the main question is how do you test for the credibility of something like astrology, existence of God and other abstract things which is militantly supported by more than a billion people?  It may be true that astrology lacks scientific evidence but there maybe, just maybe an iota of truth in it.  Maybe it is not yet good enough to be called Science.  What if the astrological “principles” are sort of pointing in the correct direction but we are not able to grasp or make a breakthrough?  The problem is that if a person dare says a word against astrology, the entire populace of India seems to find trolling an enjoyable idea.  It is difficult to argue with people who weave stories about how astrology predicted one particular event in the life of one particular person.  This is where the beautiful concept of Null Hypothesis can make the difference.

The main idea of the Null Hypothesis is that you assume a certain stance unless proven otherwise.  Or to put in more compact terms, the default position.   To demonstrate the idea of Null Hypothesis, let me take an example of a random coin toss experiment.  I assume that the default position is that the coin is unbiased, that is, there is as much chance of getting heads as there is of getting tails.  So, I toss the coin 10 times.  Now the expected number of heads that I should get is 5, assuming independent trials.  But what if I get 7 heads and 3 tails?  Can I declare “confidently” that the coin is biased?  Not really.  What if I toss the coin 100 times and I get 70 heads?  Just maybe.  What if I toss the coin 1000 times and I get heads 700 times?  Now it is more apparent that the coin is biased, but is it biased (0.65,0.35) or (0.70, 0.30)?  I do not know that yet.  I need to conduct more experiments.  Of course this is just the property of Law of Large Numbers.  The question is how many trials do I require before being say “95% confident” that the coin is biased?  There is a bit of mathematics involved in it, which I will not go into the details of.   So let me assume that in I need 70 trials out of 100 to turn up heads to declare that with 95% confidence the coin is biased.  This is termed as “Rejecting the Null Hypothesis with 95% confidence”.

Now that the tutorial on Null hypothesis is over, let’s see what can be done regarding astrology.  The main problem with astrology is that there are so many contentions regarding the rules that it becomes difficult to have a comprehensive test.  So, a manual regarding the rules of astrology needs to be prepared such that there are no evident contradictions.  If the argument is that astrology is too vast a subject to be put down into a manual, let them take a small part of it, say zodiac sign and one dominant characteristic of the person belonging to that sign.  Now we need to prove that “most” of the times the Zodiac signs and the trait in question are predictable.  So, it becomes a binary random variable akin to the coin toss experiment – heads if the person has the same traits as the one predicted by the Zodiac sign, tails otherwise.  One obvious problem is what the Null hypothesis should be.  If my thesis is that the theory of zodiac sign is truly random, we take (0.5,0.5) as the null hypothesis.  If I believe that it predicts the behaviour 60% of the times, my null hypothesis would be (0.6,0.4).

Either way, it should be an interesting experiment.  I wonder about the technical feasibility of such a project.  Well, if not the entire nation, maybe at a smaller scale inside the insti campus.

Any views? 🙂

Advertisements

25 thoughts on “Null Hypothesis, Astrology and Cargo Sciences

  1. Mrs Beryl says:

    The author is putting Deutschebank pseud.

  2. Prakash says:

    I wonder how many such thesis have contributed to this screwed up world

  3. Ola! Halfsigma,
    I take your point, The mean SAT score in mathematics is 481. The founders of a nationwide SAT preparation course claim that graduates of the course score higher, on average, than the national mean. Suppose that the founders of the course want to carry out a hypothesis test to see if their claim has merit.

    What is the null hypothesis and the alternative hypothesis that they would use?
    I’ll be back to read more next time

    • Hello,

      Sorry for the late reply. Hadn’t seen the to-be-approved comment box for a long time. I would appreciate it if you can use your actual email id, otherwise it gets classified as spam.

      Anyway, coming to your question, the founders of that course need to make a claim – something of the sort, “The subscribers of our correspondence course, score more than the national mean by so many standard deviations.”

      Then you repeat the same experiment that I have described before.

  4. The problem with checking the validity of a claim, that an astrology chart predicts something, in itself is very subjective. Here is a link to an astrology debunking video http://www.youtube.com/watch?v=3Dp2Zqk8vHw.
    It clearly shows that the wordings in which a prediction is made, is so general that many people find things applicable to them in the readings. It is just a subjective bias that people tend to look at such astrology readings, and see the ones that correspond to them and tend to forget the ones that don’t.
    The coin toss expt. on the other hand is very objective, in the sense that one can clearly identify head or tails, while the accuracy of astrology is fuzzy and subjective.

    • I had actually seen that video before. The experiment I am describing is this. Devise a test where people are asked to describe themselves. Whether they are soft-spoken or not. Introverted or an extrovert. Stuff like that. Along with it ask for their birthday. We have a separate manual which specifies the dominant trait of a particular zodiac sign.

      In this way, we can isolate the person and the bias which accompanies while reading a book on astrology.

  5. Vinay Hegde says:

    ‘It may be true that astrology lacks scientific evidence but there maybe, just maybe an iota of truth in it. Maybe it is not yet good enough to be called Science.’
    For a moment I thought you actually meant this. Thank goodness you didn’t.

    • Actually I do mean it. I can’t not disprove astrology. The only option I have is to statistically eliminate the possibility.

      If not something which may sound as arbitrary as astrology, let’s take Homeopathy. Maybe, Homeopathy does cure some diseases. I can’t just dismiss it, can I? Or something like the existence of God.

      • “Maybe, Homeopathy does cure some diseases. I can’t just dismiss it, can I? ”
        …dude? Are you kidding? Do you know how it’s supposed to work? Infinite dilution making things stronger? That’s not reason enough to reject it out of hand?

  6. Vinay Hegde says:

    Well, I guess Ashwin was referring to the Forer Effect [http://en.wikipedia.org/wiki/Barnum_effect]. Anyway, the test wouldn’t be comparable to coin toss experiment because of the inherent subjective nature of language, especially adjectives, which people use to describe their traits. When you say ‘soft-spoken’, how much soft-spoken is really soft-spoken and compared to what standard? For example, when you say ‘loud’, how much loud is loud? 60dB? 100dB? Again, what is the standard? 90dB? Who decides this and on what basis? Similarly, sensitive, attentive, charming, intelligent, and so on. All these still remain very much based on individual perceptions and therein lies the problem.

    About Homeopathy, I won’t comment on it because I do not know enough to do so. About astrology, God, etc. I take that you are advocating the ‘absence of proof is not proof of absence’ line of argument which is a fine thing by itself. But what I wanted to say was, being educated in science, as we both are, we should accept anything (or at least try to) only _after_ we have sufficient proof that it is true. The truth being determined by objective laws which are rational and scientific. I am not advocating non-following of astrology or unbelief in God, which will remain, in spite of all discussion, personal choices. I am against forcing one’s subjective views on others, which anyway wasn’t the point of your post, so I’ll stop. 😛

    Yes, I agree that ‘statistical elimination’, as you call it, is a step forward but I am merely trying to point out the difficulty in carrying it out reliably.

    • You are right about “absence of proof is not proof of absence”

      Using Null hypothesis you can’t prove a theory, just statistically reject it. I agree conducting an objective experiment with respect to astrology is difficult, that is why I suggested the narrowing down of the possibility.

      If Zodiac signs and dominant traits are not objective enough, how about something of the sort, “If Saturn comes in your zodiac sign, bad things will happen”. I agree that it is difficult to measure what “bad” is. Let’s say Saturn came in the Zodiac sign Leo. So we ask those who were born under the sign Leo whether anything that could be considered “bad” happened to them. Let them be the judge of what “bad” entails.

      I am not saying it is easy to conduct this experiment. What I am saying is that you can statistically reject at the least one small part of astrology statistically, given you can conduct an objective-enough experiment.

  7. Vasuki says:

    How astrology works: Take a bunch of adjectives and make subsets out of them with no contradicting adjective in the same subset. One can show that with almost sure probability, some adjectives in each subset would associate with a person of a certain zodiac sign (or so they would want them to be associated with that adjective) 😛
    But I would surely be interested in an experiment like this!

  8. kashthealien says:

    Check out double blind test. It has been successfully used to disprove reputed “astrologers” in Pune. It has also been used against Homeopathy.

    How astrology actually works: Tell a person what exactly he wants to hear (eg the person is brave) and he will pick up every single situation in this world that he has been in to prove that he has been brave so as to reinforce the statement, completely ignoring the situations he hasn’t been brave. If a person wants to disprove however, he will just think of the situations he has been cowardly. A double blind test eliminates both these biases, and also other information that the astrologer may get from looking at the patient’s body language, language, attitude etc.

    Another thing people crave for is, help in desperate situations. Any promise of help is accepted without questioning in such situations and astrologers happen to provide basic common sense answers to such problems.

    • Sorry for the late reply. The internet connection here is screwed up.

      The double-blind test is nothing but an implementation of the null hypothesis.

      Right now, I agree on you regarding astrology. My question is even though most of it is hokum, what if there is actually some part of it which is true. How do you test for its authenticity.

      Check out the wiki page on null hypothesis. They explain how they use it for the standard double-blind test which is standard in all medical testing.

  9. Michele King says:

    Hi There Halfsigma,
    Thanks for your thoughts, What is the consequence when the null hypothesis is rejected? Also, when do you accept the null hypothesis and what is its consequence?
    Thx.

  10. @Nikhil Punnoose

    It is not necessary that I dismiss it because it doesn’t make sense now. Wave-Particle duality when first put created a similar sort of ruckus. It was accepted finally, wasn’t it?
    In fact, personally I do not believe in Homeopathy, but I can’t be naive enough to dismiss it either just because it seems to be whacko. A statement of the form, “You ARE saying that I am both a wave and a particle? Isn’t that notion enough to dismiss it?” is not a valid counter-argument.

    This article has nothing to do with whether Homeopathy cures diseases or not. It is about how you can statistically reject a hypothesis, something of the sort, “Homeopathy can cure common cold”.

  11. Roslyn Ochoa says:

    Hello Halfsigma,
    This question may be a little off-topic, What is the consequence when the null hypothesis is rejected? Also, when do you accept the null hypothesis and what is its consequence?
    Thanks

  12. Hi Halfsigma,
    Thanks you for your post, what are four factors that influence the likelihood of rejecting the null hypothesis, along with the direction of their influence (more likely or less likely to reject H0)? i’m just confused on this.
    Wishes

  13. Vinay Hegde says:

    Hello Samaadhi,

    This question may be a little off-topic, what are the consequences of rejecting a null hypothesis is rejected? Also, when do you accept the null hypothesis and what are the consequences of its acceptance?

    Thanks,
    Vi

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

%d bloggers like this: