X

Download Power lecture (Department of Zoology) PowerPoint Presentation

SlidesFinder-Advertising-Design.jpg

Login   OR  Register
X


Iframe embed code :



Presentation url :

Home / Forest & Animals / Forest & Animals Presentations / Power lecture (Department of Zoology) PowerPoint Presentation

Power lecture (Department of Zoology) PowerPoint Presentation

Ppt Presentation Embed Code   Zoom Ppt Presentation

PowerPoint is the world's most popular presentation software which can let you create professional Power lecture (Department of Zoology) powerpoint presentation easily and in no time. This helps you give your presentation on Power lecture (Department of Zoology) in a conference, a school lecture, a business proposal, in a webinar and business and professional representations.

The uploader spent his/her valuable time to create this Power lecture (Department of Zoology) powerpoint presentation slides, to share his/her useful content with the world. This ppt presentation uploaded by onlinesearch in Forest & Animals ppt presentation category is available for free download,and can be used according to your industries like finance, marketing, education, health and many more.

About This Presentation

Power lecture (Department of Zoology) Presentation Transcript

Slide 1 - Statistical Power
Slide 2 - Ho : Treatments A and B the same HA: Treatments A and B different
Slide 3 - Points on this side, only 5% chance from distribution A. Area = 5% Critical value at alpha=0.05 Frequency A A could be control treatment B could be manipulated treatment
Slide 4 - A B If null hypothesis true, A and B are identical Probability that any value of B will be not significantly different from A = 95% Probability that any value of B is significantly different than A = 5%
Slide 5 - A B If null hypothesis true, A and B are identical Probability that any value of B will be not significantly different from A = 95% Probability that any value of B is significantly different than A = 5% = likelihood of type 1 error
Slide 6 - What you say: Reality
Slide 7 - A B If null hypothesis false, two distributions are different Probability that any value of B will be not significantly different from A = beta = likelihood of type 2 error Probability that any value of B is significantly different than A = 1- beta = power
Slide 8 - A B Effect size Effect size = difference in means SD
Slide 9 - A B 1. Power increases as effect size increases Beta = likelihood of type 2 error Power Effect size
Slide 10 - A B 2. Power increases as alpha increases Beta = likelihood of type 2 error Power
Slide 11 - A B 3. Power increases as sample size increases Low n
Slide 12 - A B 3. Power increases as sample size increases High n
Slide 13 - Power Effect size Alpha Sample size
Slide 14 - Types of power analysis: A priori: Useful for setting up a large experiment with some pilot data Posteriori: Useful for deciding how powerful your conclusion is (definitely? Or possibly). In manuscript writing, peer reviews, etc.
Slide 15 - Example : Fox hunting in the UK (posteriori)
Slide 16 - Hunt banned (one year only) in 2001 because of foot-and-mouth disease. Can examine whether the fox population increased in areas where it used to be hunted (in this year). Baker et al. found no effect (p=0.474, alpha=0.05, n=157), but Aebischer et al. raised questions about power. Baker et al. 2002. Nature 419: 34 Aebischer et al. 2003. Nature 423: 400
Slide 17 - 157 plots where the fox population monitored. Alpha = 0.05 Effect size if hunting affected fox populations: 13%
Slide 18 - 157 plots where the fox population monitored. Alpha = 0.05 Effect size if hunting affected fox populations: 13% Power = 0.95 !
Slide 19 - Class exercise: Means and SD of parasite load (p>0.05): Daphnia magna 5.9 ± 2 (n = 3) Daphnia pulex 4.9 ± 2 (n = 3) (1) Did the researcher have “enough” power (>0.80)? (2) Suggest a better sample size. (3) Why is n=3 rarely adequate as a sample size?
Slide 20 - How many samples? PCBs in salmon from Burrard inlet and Alaska In an initial survey (3 individuals each), we find the following information (mean, standard deviation) Burrard – 120.5 ± 75.9 ppb Alaska – 75.2 ± 71.9 ppb The two error bars overlap, but that’s still a big difference and we only took 3 samples The difference could be “hidden” the sizes of the errors This would be reduced by increased samples, but how many should we take?
Slide 21 - How many samples? Our difference between (q) is ~40, therefore if our confidence limits (SE) were <20ppb, we should have a difference between populations, Burrard – 120.5 ± 75.9 ppb Alaska – 75.2 ± 71.9 ppb How many samples do we therefore need??
Slide 22 - Re-arrange the equation… So we should take 56 samples to be reasonably sure of a significant difference
Slide 23 - Don’t get silly..