In: Statistics and Probability
Fifty percent of registered voters in a congressional district are registered Democrats. The Republican candidate takes a poll to assess his chances in a two-candidate race. He polls 1200 potential voters and finds that 621 plan to vote for the Republican candidate. Does the Republican candidate have a chance to win? Use a=0.05 PLEASE SHOW CLASSICAL & P-VALUE APPROACHES & TI84 METHODS FOR FINDING CV, P-VALUE & TEST STAT.
We use one proportion z test to test the claim that the republican candidate have a chance to win.
GIVEN:
Total number of potential voters
Voters who plan to vote for the republican candidate
HYPOTHESIS:
(That is, the republican candidate have a significant chance to win.)
(That is, the republican candidate does not have a significant chance to win.)
LEVEL OF SIGNIFICANCE:
TEST STATISTIC:
Since this is quite large sample we can use the central limit theorem to say that the distribution of proportions is approximately normal. Thus the test statistic is,
which follows standard normal distribution.
where
is the sample proportion
is the hypothesized value 0.5
CALCULATION:
Sample proportion
Now
Thus the calculated z test statistic value is .
CRITICAL VALUE:
The one tailed z critical value at is .
DECISION RULE:
.
INFERENCE:
Since the calculated z statistic value (1.215) is less than the z critical value (1.645), we fail to reject null hypothesis and conclude that the republican candidate have a significant chance to win since the proportion of voters who vote democrat will be in the minority ().
P VALUE:
The p value is given by,
From the z table, the p value is the value corresponding to 1.2 row and 0.02 column.
DECISION RULE FOR P VALUE:
INFERENCE:
Since the calculated P-value (0.8888) is greater than the significance level (0.05), we fail to reject null hypothesis and conclude that the republican candidate have a significant chance to win since the proportion of voters who vote democrat will be in the minority ().
Z TABLE: