Stoneroad: short explanation for p value is this: What are the chances that the results obtained in this study, which show improvement with Bavi- over placebo, happened from chance?
Researchers begin an experiment obviously wanting it to succeed, (ie that Bavi works). However, the statistical format is called "the null hypothesis". We are going to hypothesize there is no difference between placebo and Bavi (and hopefully disprove it). Was the null hypothesis, of no difference, proven or disproven here? Probably disproven. It appears that Bavi is the real deal showing improvement, more than doubling overall response rate to medication (ORR), and improving median progression-free survival (PFS) by 50%. Everyone spins these findings differently, but nobody can deny the first impression of a definite anti-cancer effect in Bavi. However, there were only 90 patients in each of the two treatment groups, and the more patients, the greater the reliability of the results, statistically. In this case most experienced researchers with statistical backgrounds would probably say "there appears to be a real difference between Bavi and placebo." Are we sure? How sure? If the pvalue is better than or equal to .05, for instance, that means: If we run 100 tests just like this one the chances of having the same signifantly better results is >95%. So 95 times out of 100 we would be right. Russian roulette with a 100 chamber gun gives you 5 or fewer chances of getting hit. Here's Wikipedia on the subject of pvalue: "In statistical significance testing, the p-value is the probability of obtaining a test statistic at least as extreme as the one that was actually observed, assuming that the null hypothesis is true. In this context, value a is considered more "extreme" than b if a is less likely to occur under the null. One often "rejects the null hypothesis" when the p-value is less than the significance level a (Greek alpha), which is often 0.05 or 0.01. When the null hypothesis is rejected, the result is said to be statistically significant."
Note: all the above is cloaked in negatives, and therefore difficult.
News 
Market Data 
Discover 
AI
