When one changes the significance level of a hypothesis test from 0.01 to 0.05, which of the following will happen? Check all that apply. A. It becomes easier to prove that the null hypothesis is true. B. The chance of committing a Type I error changes from 0.01 to 0.05. C. It becomes harder to prove that the null hypothesis is true. D. The test becomes more stringent to reject the null hypothesis (i. e. it becomes harder to reject the null hypothesis). E. The test becomes less stringent to reject the null hypothesis (i. e., it becomes easier to reject the null hypothesis). F. The chance of committing a Type II error changes from 0.01 to 0.05. G. The chance that the null hypothesis is true changes from 0.01 to 0.05.

Question

Heidy Berger

Social Studies

5 January, 10:52

When one changes the significance level of a hypothesis test from 0.01 to 0.05, which of the following will happen? Check all that apply. A. It becomes easier to prove that the null hypothesis is true. B. The chance of committing a Type I error changes from 0.01 to 0.05. C. It becomes harder to prove that the null hypothesis is true. D. The test becomes more stringent to reject the null hypothesis (i. e. it becomes harder to reject the null hypothesis). E. The test becomes less stringent to reject the null hypothesis (i. e., it becomes easier to reject the null hypothesis). F. The chance of committing a Type II error changes from 0.01 to 0.05. G. The chance that the null hypothesis is true changes from 0.01 to 0.05.

+4

Answers (1)

Know the Answer?

Jane Bentley · Answer 1

B) The chance of committing a Type I error changes from 0.01 to 0.05., E) The test becomes less stringent to reject the null hypothesis (i. e., it becomes easier to reject the null hypothesis), therefore. C) It becomes harder to prove that the null hypothesis is true and G) The chance that the null hypothesis is true changes from 0.01 to 0.05 are all correct answers

Explanation:

The alfa error or type I, refers to the probability error of rejecting the null hypothesis, when it is true, it is the chance of mistake when affirming that an association exists between two variables tested as a cause of an effect on something. i. e: H1: fast food is responsible for diabetes (this is working hypothesis), H0: red hair is responsible for diabetes (this is the null hypothesis). The beta error or type II is related to the size of the sample, it is the chance of accepting something (the null hypothesis) when it is false, depends mostly on having enough measures (or persons under study) so your hypothesis can be proven and be a real representation of the population under study. The statistical significance, namely the p value, can be narrow (p=0.01) or wide (p=0.05), it can be easily understand if we explain it in terms of percentage: you can have 99% (p=0.01) of certainty to affirm that the null hypothesis (the one that you do not believe is true, in the example, red hair as cause of diabetes) is actually wrong or 95% (p=0.05) of certainty to affirm that the null hypothesis (again, the one that you do not believe is true) is actually wrong.