i need specific answer

Suppose we conducted a study and examined the relationship between Number of Years in School and Number of Offenses Committed. We found the correlation value (r) to be .72, but it was not significant. Despite having such a high value, why might this relationship not be statistically significant? Suppose we conducted a second study and examined the same variables. This time, the correlation value (r) was .32, but it was significant. Why, despite being much lower (i.e. weaker), might this relationship be statistically significant? Which would you trust as a policy maker (the weaker significant relationship or the stronger nonsignificant relationship)? Explain your position.