Tests of Proportions (1 of 4)
A manufacturer is interested in whether people can tell the difference between
a new formulation of a soft drink and the original formulation. The new formulation
is cheaper to produce so if people cannot tell the difference, the new formulation
will be manufactured. A sample of 100 people is taken. Each person is given a taste
of both formulations and asked to identify the original. Sixty-two percent of the
subjects correctly identified the new formulation. Is this proportion significantly
different from 50%?
- The first step in hypothesis testing is
to specify the null hypothesis and an alternative
hypothesis. In testing proportions, the null hypothesis is that π,
the proportion in the population, is equal to
some specific value. In this example, the null hypothesis is
that π
= 0.5. The alternate hypothesis is π ≠ 0.5.
- The second step is to choose a significance
level. Assume the 0.05 level is chosen.
- The third step is to compute the difference
between the sample proportion (p) and the value of π specified
in the null hypothesis. In this example, p - π = 0.62 - 0.5 = 0.12.