Question

The International Journal of Applied Application Theory has a policy that they will only publish experimental results that are statistically significant when tested at a significance level of 0.05.

a. Could all of their published results be true? (Yes or No)

b. Could all of their published results be false? (Yes or No)

c. A researcher collected data that fit the criteria for a two-sided Z-test. (A Z-test is just a test of a normally distributed statistic after “normalizing” the statistic to have mean 0 and variance 1.) He set the significance level at 0.05. He ran 80 trials and got a z-value of 1.7. This gave a p-value of 0.0892, so he could not reject the null hypothesis. Convinced that his alternative hypothesis was correct he ran 80 more trials. The combined data from the 160 trials now had a z-value of 2.1. He wrote a paper carefully describing his experiments and submitted it to the journal in part (a). Will the journal publish his results? (Yes or No)

d. Explain your answer for part (c): Why or why not?

Answer 1

Answer:

Given

significance level=0.05

(a) No. As there is a significance level associated

(b) No. For the same reason stated above.

(c) given

significance level=0.05

p-value of 0.0892

He ran 80 trials and got a z-value of 1.7.

The combined data from the 160 trials now had a z-value of 2.1

answer: No. The journal would not publish his results.

d. since the  corresponding to the z score of 2.1, the p-value is 0.036, which is less than 0.05(level of significance).Therefore we have to reject the null hypothesis(H0) . And hence the journal would not publish his results.