Q1. Suppose of an outcome variable anal sex and independant variable access to pornography - i find the proportion of anal sex is higher in those having access to pornography as compared to not having access to porn but the p value is greater than 0.05 . we interpret it as following
1. There is association but not a statistically significant one
Q2. Null hypothesis : No association between Anal sex and Pornography access
Alternate hypothesis : Anal sex associated with Pornography
The p value turns out to be greater than 0.05
As per this concept irrespective of proportion of outcome in independant variable ( even if proportion of anal sex is higher in those accessing pornograph ) there is no association as p value is greater than 0.05
Kindly resolve this concept
Code the answers in a way that you can compute a normal correlation and see if it is significant
http://stats.stackexchange.com/questions/103801/is-it-meaningful-to-calculate-pearson-or-spearman-correlation-between-two-boolea
Dear Ashish,
P value tells you if your results are chance occurrence or not. If your results are more thn 0.05 chances are that the difference that you have got in the variables are due to chance occurrence and if the observations are repeated in other populations the difference may not hold true. That is what it means. When we do statistical analysis such as this, we are in a way looking at the distance between two population curves rather than particular values, here is where the difference comes in what it actually is and what it seems to be.