The effect of a new drug on blood pressure

A researcher wishes to study the effect of a new drug on blood pressure. Would you
recommend using a z-test, a t-test, or an A.N.O.V.A. for the analysis? Explain your answer.
What would your choice of test depend on? For the test you select, explain your design and
your comparison groups. Would the hypothesis be directional or non-directional? Would
the test be one-tailed or two-tailed? What would be the null and what would be the
alternative hypothesis?

The effect of a new drug on blood pressure

In this study, I would recommend use of ANOVA for the analysis. This will allow the
researcher to analyze differences between various means and other associated behaviors such as
the differences in their reactions, and any effects of the drugs on the sample size incorporated in
the study. This analysis is also preferred because, it help to provide statistical test in determining
whether the means of various groups is equal or not and generalizes t-test into more than two
group hence helping to reduce chance of committing type I error (Wilkinson, 1999). It is also one
of the best methods of comparing three or more means for statistical significance.
The test to be used is t-test. The choice of this test will depend on whether the null
hypothesis is supported. In case this is not supported then an alternative test will be used. This
will be used because, data may significantly be different from each other and the values of
scaling term in the test statistics may be known. The test will be carried out from various
people who will make up the sample. The sample will be then put under similar environment and
conditions to ensure that the study is accurate. All the patients in the sample will be suffering
from blood pressure. The medication will be administered at same time to ensure uniformity. The
patients will then be measured after a period of one month to determine if there is any variation
among the patients.

The hypothesis will be non-directional hence it will be two tailored. This is because, in
incase an independent variable is manipulated there is likely to have a change in the dependent
variable and this change cannot be predicted to be positive or negative. For instance, if the
amount of dosage is varied it is likely for the patient to either respond positive or negatively.
The null hypothesis will be ‘patients will respond in similar way upon administering of the
dosage’. Alternative hypothesis will be, “there will be different response among patients upon
administration of the dosage among the patients.



Wilkinson, L. (1999). “Statistical Methods in Psychology Journals; Guidelines and
Explanations”. American Psychologist 54 (8): 594–604.