Home | Community | Message Board


This site includes paid links. Please support our sponsors.


Welcome to the Shroomery Message Board! You are experiencing a small sample of what the site has to offer. Please login or register to post messages and view our exclusive members-only content. You'll gain access to additional forums, file attachments, board customizations, encrypted private messages, and much more!

Shop: PhytoExtractum Maeng Da Thai Kratom Leaf Powder   Unfolding Nature Unfolding Nature: Being in the Implicate Order

Jump to first unread post Pages: 1
Offlinehoodbran
Dosser
Male User Gallery


Registered: 06/01/08
Posts: 1,570
Loc: Phloston Paradise
Last seen: 19 hours, 18 minutes
statistics
    #15366125 - 11/13/11 06:47 PM (12 years, 4 months ago)

Hi All, if I have a data file:



Would it make sense to exclude the cases that have experienced no difference in bc_before and bc_after?

The participants in this data file rated on a scale from 4-24 and thus to include them would skew the results if they were not excluded.

Does this logic sound right?


--------------------
Not all drugs are good, Some are great.




Extras: Filter Print Post Top
InvisibleBrainstem
_@_y
Male User Gallery


Registered: 07/31/10
Posts: 1,969
Loc: In my shell Flag
Re: statistics [Re: hoodbran]
    #15366629 - 11/13/11 08:46 PM (12 years, 4 months ago)

I don't really know, but wouldn't doing that give a disproportionate result ?, by only using a third of the results I mean. This would then lead to a conclusion based on less than 50% of the cases. Like I say I don't know.

Extras: Filter Print Post Top
OfflineKickleM
Wanderer
 User Gallery

Registered: 12/16/06
Posts: 17,953
Last seen: 2 days, 6 hours
Re: statistics [Re: Brainstem]
    #15366824 - 11/13/11 09:21 PM (12 years, 4 months ago)

really hard to say just from raw data and no background info. for instance I don't understand this: "The participants in this data file rated on a scale from 4-24 and thus to include them would skew the results"

Naturally changing your numbers, including the number of participants, is going to change your statistics. But what is the detriment to including them? If you are only removing the absence of change in order to find significant change, it's likely a bad practice.

Example: I have a new pharmaceutical I am trying out and I give it to 50 participants. Of these 50 participants, 20 report no change in presenting symptoms. That means the pharmaceutical was only effective on 3/5 of those tested. If I remove those who experienced no change, suddenly it looks like the pharmaceutical was 100% effective and this simply isn't true and is very misleading.


--------------------
Why shouldn't the truth be stranger than fiction?
Fiction, after all, has to make sense. -- Mark Twain

Extras: Filter Print Post Top
InvisibleBrainstem
_@_y
Male User Gallery


Registered: 07/31/10
Posts: 1,969
Loc: In my shell Flag
Re: statistics [Re: Kickle]
    #15366851 - 11/13/11 09:27 PM (12 years, 4 months ago)

I was tempted to ask what bc was, now I'm wondering if the bc before was 18, and the bc after was 18, how the bc be 0 ?, bc alone would indicate the difference between before and after, but saying bc before and after would be like saying difference before and after based there being a bc column. :willynilly:

Extras: Filter Print Post Top
Invisiblejohnm214
Male User Gallery
Folding@home Statistics
Registered: 05/31/07
Posts: 17,582
Loc: Americas
Re: statistics [Re: hoodbran]
    #15366863 - 11/13/11 09:29 PM (12 years, 4 months ago)

As has allready been suggegsted, without defining what you want to measure, or what question you want to answer, it is impossible to advise what set of data may or may not be appropriate to examine or exclude.

Extras: Filter Print Post Top
Offlinehoodbran
Dosser
Male User Gallery


Registered: 06/01/08
Posts: 1,570
Loc: Phloston Paradise
Last seen: 19 hours, 18 minutes
Re: statistics [Re: Kickle]
    #15367776 - 11/14/11 01:50 AM (12 years, 4 months ago)

Thanks.  I found this helpful:

Quote:

Kickle said:
really hard to say just from raw data and no background info. for instance I don't understand this: "The participants in this data file rated on a scale from 4-24 and thus to include them would skew the results"

Naturally changing your numbers, including the number of participants, is going to change your statistics. But what is the detriment to including them? If you are only removing the absence of change in order to find significant change, it's likely a bad practice.

Example: I have a new pharmaceutical I am trying out and I give it to 50 participants. Of these 50 participants, 20 report no change in presenting symptoms. That means the pharmaceutical was only effective on 3/5 of those tested. If I remove those who experienced no change, suddenly it looks like the pharmaceutical was 100% effective and this simply isn't true and is very misleading.




--------------------
Not all drugs are good, Some are great.




Extras: Filter Print Post Top
Jump to top Pages: 1

Shop: PhytoExtractum Maeng Da Thai Kratom Leaf Powder   Unfolding Nature Unfolding Nature: Being in the Implicate Order


Similar ThreadsPosterViewsRepliesLast post
* Man/Boy Love
( 1 2 3 4 ... 25 26 )
zahudulallah 27,893 505 05/15/05 06:03 AM
by Alan Stone
* Pharmaceutical Inquisition??? barfightlard 1,911 6 05/18/05 02:29 AM
by Nocforweb
* 2012 Statistical Proof ! lucid 626 6 12/28/03 05:26 PM
by lucid
* body modification (pics included!)
( 1 2 all )
SpecialEd 1,930 22 02/09/04 03:18 PM
by anarchyhollow
* What exists, from my perspective (includes psychic phenomena- please read!) RedNucleus 937 18 07/05/05 01:30 PM
by RedNucleus
* Islam is Not For Anyone!
( 1 2 3 all )
LSDempire 3,197 41 08/14/05 10:50 AM
by Redstorm
* The total nonsense of Homeopathy
( 1 2 3 4 5 6 all )
Swami 8,021 119 04/10/09 10:32 PM
by spyder
* Legislating Moral Values?
( 1 2 3 4 all )
SeussA 4,825 73 02/19/04 09:55 AM
by Seuss

Extra information
You cannot start new topics / You cannot reply to topics
HTML is disabled / BBCode is enabled
Moderator: Middleman, DividedQuantum
386 topic views. 0 members, 13 guests and 1 web crawlers are browsing this forum.
[ Show Images Only | Sort by Score | Print Topic ]
Search this thread:

Copyright 1997-2024 Mind Media. Some rights reserved.

Generated in 0.022 seconds spending 0.004 seconds on 12 queries.