Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

What test should I use?


  • Please log in to reply
4 replies to this topic

#1 biology_06er

biology_06er

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 55 posts
0
Neutral

Posted 26 May 2012 - 07:54 PM

Hi there,

I recently did an ELISA and have no idea what type of stats test to use?...I haven't done a single stats paper/stats related anything in over 6years..so am struggling to analyze my current results.

I basically did the standard elisa,,,,adding various amounts of protein to coated wells...and i did triplicate wells for each concentration...so i guess what i would like to know if there is a a stat. sig difference between the highly concentrated and less concentrated samples...

anyhelp will be greatly appreciated..preferably in a language i can understand!

cheers,
b_06er

#2 Felipillo

Felipillo

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 54 posts
4
Neutral

Posted 30 May 2012 - 05:47 PM

You should try repeated mesures AOVA
Chance favors the prepared mind
Louis Pasteur.

#3 newborn

newborn

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 71 posts
2
Neutral

Posted 30 May 2012 - 06:12 PM

ANOVA test

#4 biology_06er

biology_06er

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 55 posts
0
Neutral

Posted 05 June 2012 - 06:38 PM

Hi,

Thanks for the replies...I tried doing the RM anova but it came out as sig..but clearly I had no binding..soo what am I doing wrong..essentially the -ve control values were the same as by test readings..the curve shows a very very slighly increase from lowest conc. to highest conc. If i wanted to test if there was a diff. between highest/lowest comp..seems an Anova isn't working :(

#5 Felipillo

Felipillo

    Enthusiast

  • Active Members
  • PipPipPipPipPip
  • 54 posts
4
Neutral

Posted 06 June 2012 - 07:13 PM

May be you need a post hoc test, like Dunnet for comparing every group against the control, or the tukey test to compare every group of your samples.

you could try to post this question to http://stats.stackexchange.com with some sample data

Edited by Felipillo, 06 June 2012 - 07:14 PM.

Chance favors the prepared mind
Louis Pasteur.




Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.