10.44 Two-sided permutation P-value Refer to the previous exercise and the sampling distribution mentioned there.

What is the permutation P-value for the two-sided test with alternative hypothesis that the mean improvement scores are different under therapies 1 and 2? (Hint: As for any regular test, the two-sided P-value is the the sum of the tail probabilities of test statistic values as extreme as the observed one. Extreme differences are those with an absolute value as large as or larger than the absolute value of the observed difference.)