Explore BrainMass
Share

# Hypothesis: Null and Alternative

1: How do you decide your null and alternative hypotheses?
2: How do you as a researcher decide the confidence level? What factors do you consider to make that decision?
3: How do you decide on the type of test? For example, should you use z-test, t-test, or non-parametric test that we will study later on [such as x^2 test, F-test]?

#### Solution Preview

(1) The claim made by the researcher with respect to a problem or issue in question forms the alternative hypothesis (Ha). The opposite of this claim forms the null hypothesis (Ho). Thus, The null hypothesis is chosen to be the opposite of the claim, that is, opposite of what the researcher hopes or believes is true. The claim forms the alternative hypothesis.

For example, if a quality control inspector believes that the mean diameter of a population of steel bolts produced by a machine is greater than 12.00 mm, then Ha: u > 12.00, while Ho: u ? 12.00.
Mathematically, the null hypothesis must contain the ?=? sign. The other statement, which does not contain the ?=? sign forms the alternative hypothesis.

(2) What confidence level a researcher may choose depends upon how critical the managerial ...

\$2.19