ASU Ecn 221 Exam 3 Study Guide Questions
And Answers 100% Pass
Hypothesis testing - ANS can be used to determine whether a statement about the value of a
population parameter should or should not be rejected
null hypothesis - ANS a tentative assumption about a population parameter
alternative hypothesis - ANS the opposite of what is stated in the null hypothesis
what do hypothesis testing procedures use to test competing statements indicated in the null
and alternative hypothesis? - ANS data from a sample
type I error - ANS - rejecting the null hypothesis even when it's true
level of significance (alpha) - ANS the probability of making a type I error when the null
hypothesis is true as an equality
type II error - ANS failing to reject (aka accept) the null hypothesis even when it's false
p-value - ANS the probability computed using the test statistic, that measures the support
(or lack of support) provided by the sample for the null hypothesis
, ©EVERLY 2025 ALL RIGHTS RESERVED
rejection region - ANS where the value of the test statistic is if the p-value is less than or
equal to the level of significance
at what point do you reject the null hypothesis in regard to p-value? - ANS when the p-value
is less than or equal to the level of significance (alpha)
critical value for the test - ANS - the value of the test statistic that establishes the boundary
of the rejection region
- found on the z or t table
what is the rejection rule? - ANS - lower tail: reject the null hypothesis if t ≤ tα (alpha)
- upper tail: reject the null hypothesis if t ≥ tα (alpha)
when does the statistic have a z distribution, called a z statistic instead of a t statistic? -
ANS if the sample sizes are large (greater than 30) and the population variances are known
what happens when the sample sizes are at least 30 for each sample and the population
standard deviations are not known? - ANS it follows a t distribution
factor - ANS variable that the experiment has selected for investigation
treatment - ANS level of a factor
experimental units - ANS objects of interest in the experiment
completely randomized design - ANS an experimental design in which the treatments are
randomly assigned to the experimental units