Dataset:  esp_studies1.csv

Sources: L. Storm, P.E. Tressoldi, L. Di Risio (2010). "Meta-Analysis of
Free-Response Studies, 1992:2008: Assessing the Noise Reduction Model in
Parapsychology," Psychological Bulletin, Vol. 136, #4, pp. 471-485.

J.N. Rouder, R.D. Morey, J.M. Province (2013). "A Bayes Factor Meta-Analysis
of Recent Extrasensory Perception Experiments: Comment on Storm, Tressoldi,
and Di Risio (2010)," Psychological Bulletin, Vol. 139, #1, pp. 241-247.

L. Storm, P.E. Tressoldi, J. Utts (2013). "Testing the Storm et al. (2010)
Meta-Analysis Using Bayesiand and Frequentist Approaches: Reply to Rouder
et al. (2013)," Psychological Bulletin, Vol. 139, #1, pp. 248-254.

Description: Results from 67 ESP experiments in 3 categories:
1) Ganzfeld   2) Nonganzfeld with noise reduction   3) Standard free response
Each trial is classified by: # trials, # Successes, z-score and 2 effect size
measures. Studies 7 and 11 have different numbers than original paper.
Z computed as follows (not all match authors' given z-values).
p0 = 1/#choices   p_hat = #Success/#Trials = y/n

p_hat >= p0   Z = (y - np0 - 0.5) / sqrt(n*p0*(1-p0))
p_hat < p0   Z = (y - np0 + 0.5) / sqrt(n*p0*(1-p0))

Effect Size 1: Z/sqrt(n)
Effect Size 2 (Cohen's h): 2*(arcsin(sqrt(p_hat)) - arcsin(sqrt(p0)))

Variable Names:
Study  
Group  
InFinal   (1 if used in Reply to Rouder, et al by Storm,Tressoldi,Utts)
Rouder1   (1 if used in Rouder, et al Comment on original paper)
Year
numtrials
numhits
k        (# choices, usually 4)
pi_0
pi_hat
Z_obs
eff_sz1
eff_sz2