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!
- - - - -

How to calculate the power of a family-based study and for case-control studies

Power sample size Family-based study Case-control genetics

  • Please log in to reply
No replies to this topic

#1 Andreia Marques

Andreia Marques


  • Active Members
  • Pip
  • 5 posts

Posted 10 March 2012 - 09:05 AM

I've been looking for softwares to calculate power for association studies evaluate if I'm using an adequate sample size. In my particular case, I'm investigating the association between a specific gene and a disease using 2 approaches:

1 - Case-control study where I compare the genotypes of cases (patients) versus controls (healthy subjects). I've a sample of 550 cases and 700 controls. I want to know the power for an odds ratio of 1.5, considering an alpha level of 0.05 and a frequency of the risk allele = 0.50. Using softwares like EpiInfo, Gpower and some online calculators like Sampsize I got a power of 94% (I wonder if I'm doing this well...)

2 - Familiar study involving trios (with 2 parents and 1 affected children). This one is complicated since I still couldn't find the adequate analysis to compute power... I have 250 complete trios (750 samples) + 150 incomplete trios (300 samples). I want to know the power considering an odds ratio of 1.5, alpha level of 0.05 and a frequency of the risk allele = 0.50. How many families do I need to achieve 95% of power?

I'm very confused...I'd appreciate if someone could help me, give me some tips or tell me the adequate software or statistical tests for my specifi cases. And sorry, my knowledge about statistics is far from the ideal, sorry If I sound too silly...

Also tagged with one or more of these keywords: Power, sample size, Family-based study, Case-control, genetics

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

©1999-2013 Protocol Online, All rights reserved.