Genetically modified foods
1, In order to determine if genetically modified foods are equally safe as traditionally farmed crops, 1,000 people will be chosen at random and will be asked to join the trial (Walpole, 1982). If 500 or more of those individuals consuming genetically modified foods in their daily diet suffer from a new medical disorder within one year of the testing period, genetically modified foods will be considered unsafe or less safe than the traditionally farmed crops (Patel et al., 2005).
Null hypothesis: Genetically modified foods are equally safe as traditionally farmed crops.
Alternate hypothesis: Genetically modified foods are safer than traditionally farmed crops.
2. In order to test the two hypotheses, a binomial parameter for the probability of the success of the null hypothesis is p = ¼ and the probability of the success of the alternative hypothesis is p > ¼.
H0: p = 1/4
H1: p > 1/4
The statistic on which a decision will be based is X, the number of individuals that will be included in the test group who will consume genetically modified foods for a period of at least 6 months. The possible values of X will be divided into two groups: those that acquire new medical illnesses and those than remain healthy for an entire year. The null hypothesis will be tested at the 0.0409 level of significance.
3. The hypotheses will be tested using the two-tailed test, which splits the testing population into two equal parts and placing each tail of the distribution of the test statistic. The null hypothesis will be stated using the equality sign to specify a single value. The alternative hypothesis should allow genetically modified foods to be either inferior or superior to the conventional farmed crops.
4. The Chi-square test may not be applied to this test because the hypothesis involves discrete numbers of individuals that either contract or do not contract an illness. In addition, the population size to be tested is not too large, hence the two-tailed sided is optimal for application to this type of analysis. This test also involves generation of critical regions, which are not applicable if a chi-square test is employed. The critical regions are those values that are generated within the midline of the expected data.
Patel R, Torres RJ and Rosset P (2005): Genetic engineering in agriculture and corporate engineering in public debate: Risk, public relations, and public debate over genetically modified crops. Int. J. Occup. Environ. Health 11:428-436.
Walpole RE (1982): Introduction to statistics. New York: Macmillan Publishing, Co.521 pages.