Notices
Results 1 to 8 of 8

Thread: hypothesis as part of scientific method

  1. #1 hypothesis as part of scientific method 
    Forum Freshman Crimson_Scribe's Avatar
    Join Date
    May 2005
    Location
    Alberta, Canada
    Posts
    39
    What exactly is the purpose of a hypothesis? Why do we need one when performing an experiment? Can’t it bias the study? Can’t it just be used as an artificial measuring stick of ‘proof’?


    Theatre is the laboratory of the human soul. – Peter Brook
    Reply With Quote  
     

  2.  
     

  3. #2  
    Forum Cosmic Wizard SkinWalker's Avatar
    Join Date
    Apr 2005
    Location
    Grand Prairie, TX
    Posts
    2,377
    A hypothesis is formed when an event or phenomenon is observed.

    For instance, say I come home from work and attempt to turn on my living room lamp by the switch near the door. The switch is moved but the lamp does not come on.

    My first thought is that bulb is burned out. I only suspect that this is the case, since the bulb didn't flash or burn out as I lifted the switch... but it's the most probable explanation. It could be the cord is unplugged from the wall, but this is unlikely since the cord plugs in behind the couch and there is no known mechanism to pull the plug. Had I recently swept behind the couch, then this hypothesis becomes very plausible, but for now, it remains an hypothesis that is possible but not very probable.

    Another hypothesis is that the power is out for the building, but a quick look at the VCR shows the clock operating, falsifying that hypothesis. So I settle on the bulb, go to retrieve one from the closet, and set out to remove the old one.

    But even when we think we've tested all hypotheses and have settled on the most likely one and have created a theory (mine is that the bulb has reached it's maximum life), we have to be open-minded for new information.

    Suppose as I go to remove the old bulb, I realize that it is a bit loose in the socket. Instead of removing it, I tighten it and the light comes on.

    The old hypothesis -the most probable one- is now falsified and I must revise my theory.

    Indeed, now I have an entirely new observation that gives rise to new questions. How did the bulb become too loose to make contact in the socket!

    Hypotheses are informed guesses about what you observe. It's best to come up with as many as you can, even when you have good reason to believe in a particular one. It's also important that each of these hypotheses be falsifiable, that is to say that they can possibly be proved or demonstrated to be wrong. For instance, an hypothesis that a ghost entered my home and loosened my bulb cannot readily be tested. Vibrations from passing trucks in the neighborhood can.


    Reply With Quote  
     

  4. #3  
    Forum Masters Degree invert_nexus's Avatar
    Join Date
    Apr 2005
    Posts
    638
    Damn fine explanation, Skinwalker.
    Can't really add much to that.
    So I won't.

    Now. Back to your debunkery...
    Ha! kidding. (Really. I kid.)
    Reply With Quote  
     

  5. #4  
    Forum Cosmic Wizard SkinWalker's Avatar
    Join Date
    Apr 2005
    Location
    Grand Prairie, TX
    Posts
    2,377
    Quote Originally Posted by invert_nexus
    (Really. I kid.)
    There's no evidence to support that.
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    .
    j/k
    Reply With Quote  
     

  6. #5  
    Forum Freshman craterchains's Avatar
    Join Date
    May 2005
    Location
    Tacoma, WA, usa
    Posts
    58
    This reminds me of the process of thought that brought me to realize that CS types of crater chains are not from any broken up bodies.

    Final highest probability would be "made by intelligence".
    It's not what you know or don't know, but what you know that isn't so that will hurt you. Will Rodgers 1938
    Reply With Quote  
     

  7. #6  
    Forum Cosmic Wizard SkinWalker's Avatar
    Join Date
    Apr 2005
    Location
    Grand Prairie, TX
    Posts
    2,377
    Only if you consider that you skipped the hypotheses that the bulb might be out, that fuse might be blown, that the power might be out, that the bulb might be loose, and even that the switch might be faulty in favor of an hypothesis that only includes the ghost.
    Reply With Quote  
     

  8. #7  
    Forum Freshman ericwernli's Avatar
    Join Date
    May 2005
    Posts
    13
    yes, it seems you've probably left out a lot of hypotheses. Remember to keep an open mind to other (natural) causes. Also remember disproof of one hypothesis does not prove a counter hypothesis.
    Reply With Quote  
     

  9. #8  
    New Member
    Join Date
    Mar 2005
    Posts
    2
    I really can't add much to Skinwalker's explanation. It is superb.

    Remember though too, that a hypothesis serves to help the researcher come up with a suitable experimental design, and thus focus the experiment. Experimental design is important as if you do not know how to go about your quest for data, your results might be completely useless. Take for example the experiment to test the effect of THC (the active ingredient on Marijuana) on mice. The null hypothesis was that increasing the AMOUNT of THC given had no effect. So the researchers proceeded to give these mice bigger and bigger injections of this solution. What they concluded was that it had a SEVERE effect in that some mice died, some just stopped functioning all together. Well think about it -- increasing the amount of the drug given per treatment only served to creat a bigger and bigger bubble on the mouse's back. If you were a mouse would you go on about your normal business, or would you stop everything and obsess over your new hump. Not only that, but too much of anything, even water, is lethal.

    Whereas, if the researchers had come up with a different hypothesis -- for example, that increasing CONCENTRATIONS of THC in equal doses, had no effect, they might have been able to get more believable results.

    Comming up with a hypothesis before you test does not sqew your results. In fact You need to formulate a hypothesis BEFORE you experiment. Comming up with a hypothesis after you have gathered data is lethal. You have already looked at the data -- this invalidates the test, unless you plan to collect a new set of data after you have looked at the initail set and formed a hypothesis.
    "Not getting what you want can sometimes be a wonderful stroke of luck!" -- The Dahlia Lama
    Reply With Quote  
     

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •