RESEARCH METHODS-CALCULATION OF SAMPLE SIZE- CROSS-SECTIONAL STUDIES
Where; Z = Standard normal variate or z-score (at 5% type I error (p less than 0.05)) which is always 1.96 and at 1% (p less than 0.01) and the value is 2.58. In most of the studies, the p-values is significantly considered to be less than 0.05 and therefore the z score is normally taken as 1.96. P = Expected proportion in population basing on previous studies or pilot studies. d = Absolute error or precision and this has to be decided by the researcher. Let’s take one example; During the first 3 months of COVID 19 pandemic related lockdown, the teenage pregnancy was reported to increase among school going girls in different part of Uganda. Some researchers therefore wanted to investigate the news to find out whether it was right. In that respect, let’s calculate the minimum sample size of girls that are required to generate enough information to approve or disprove the rumor. Taking the precision/absolute error d at 5% (d=0.05) and type I error at 5% (z=0.05), the most previous prevalence according to WHO in the country was 52% (p=0.52). Using the above formula, = 384 respondents Therefore, to generate enough evidence for the above study, one would have to collect data from a minimum of 384 respondents.
