Question

What does p value tell you about significance?

I'm trying to interpret an OLS output and the table has a legend that says

*p<.1

**p<.05

***p<.01

if i'm trying to interpret the significance of the coefficients in the table in essay format, what would i say for *, **, and ***?

for example, how would i finish this statement: "(Variable A) is significant at ___ because its p value is less than (.1, .05, or .01)."

Answer 1

Significance codes:

0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Significance codes are used to compare p values for the coefficient. Significance codes are level of significance alpha. If pvalue is less than alpha, we reject the null hypothesis. Here pvalue is compared with all given significance codes (alpha) and decided about the degree of significance of coefficient.

For example,

if pvalue <0.001, the coefficient is highly significant. In this case p value is followed by ‘***’

If p value is <0.01, the coefficient is moderately significant. . In this case p value is followed by ‘**’.

If p value <0.05, the coefficient is significant. . In this case p value is followed by ‘*’.

If p value <0.1, the coefficient is less significant. . In this case p value is followed by ‘.’

If p value <1, the coefficient is not significant. . In this case p value is followed by nothing.

Asteric (*) sign provides a quick and easy comparison between p value and alpha. Only looking at number of asterisks one can tell about significance of coefficient.

The significance codes tells about percentage of certainty about the coefficient has an impact on the dependent variable.

For example, a significance level of 0.01 indicates that there is less than a 0.01% chance that the coefficient can be equal to 0 i.e coefficient is insignificant. In other words, there is 99.99% chance that it is significant.

statement: "(Variable A) is significant at 0.1 because its p value is less than 0.1”.

or "(Variable A) is significant at 0.05 because its p value is less than 0.05”. and so on.