# Jessica

less info
reputation
17
bio website location age member for 2 years, 1 month seen Oct 2 '12 at 18:28 profile views 28

# 37 Actions

 Feb13 awarded Yearling Oct2 comment Power analysis for binomial data when the null hypothesis is that $p = 0$ According to pwr.p.test, for a power of 0.5, you need at least 677 observations. But power = 0.5 is very low! Oct2 answered Power analysis for binomial data when the null hypothesis is that $p = 0$ Sep25 answered How to determine which variables are statistically significant in multiple regression? Sep25 asked Power analysis with weighted survey data Sep25 comment Conducting a power analysis on difference between two proportions In the pwr package, also pwr.2p.test and pwr.2p2n.test. Jun11 answered Practical significance, especially with percents: “standard” measure and threshold Jun9 comment Practical significance, especially with percents: “standard” measure and threshold In the context of $y = a+bx$, I think a good measure of practical significance is $Q = b\sigma_x/\sigma_y$ (is there a name for this?). It measures the change (in SD's) in $y$ for a 1 SD change in $x$. Thus, you could say something like: Q < 1/3: very low effect; Q < 1/2: low effect; Q 1/2 to 2: medium effect; Q > 2: high effect; Q > 3: very high effect. In your example, all I need to know is that $b = 0.0000001$ and the SD's of $y$ and $x$. Jun9 revised Practical significance, especially with percents: “standard” measure and threshold fixed typo Jun9 asked Practical significance, especially with percents: “standard” measure and threshold May8 comment Mixture model fixed effects Thank you everyone who has read my question and tried to answer. All of the answers gave me good ideas / starting points. Unfortunately, no answer gave a straightforward solution. I will therefore let StackExchange automatically assign the bounty based on the votes cast by the community. May8 comment Mixture model fixed effects Some good ideas. Thank you. May1 comment How to score Likert scales and test a hypothesis relating two variables? This is probably too much for the original poster, but others might find this interesting: en.wikipedia.org/wiki/Cultural_consensus_theory Apr30 awarded Commentator Apr30 comment Mixture model fixed effects Thanks! Done... Apr30 revised Mixture model fixed effects retag Apr30 awarded Promoter Apr30 revised Mixture model fixed effects Clarifications Apr27 comment Mixture model fixed effects @Aniko- Thank you! Any suggestion as to which package does this? (An R package, a MATLAB function, something else?) Apr27 answered Expected value for discrete (nominal) variable?