    # Also in the Article

Sample Size Determination and Sampling Procedures
This protocol is extracted from research article:
Rate and Predictors of Neonatal Jaundice in Northwest Ethiopia: Prospective Cohort Study
J Multidiscip Healthc, Feb 22, 2021;

Procedure

The minimum required sample size was calculated using a single population proportion. We used a STATATM Version 14.0 statistical software to calculate the sample size by considering the following assumptions: level of significance (α) = 5%, Za/2 (value at 95% confidence interval =1.96), power of 80%, and the prevalence (P= 0.5 since there was no previous study). We could not apply sample size calculation for predictors since there was no previous study. Hence, the sample size was calculated using a single population proportion for this study. The calculated sample size was found to be 384, but since the estimated total populations in the study area during the study period were less than 10, 000 the following correction formula was used.

nf = ni/1+ni/N=384/1 + 384/1500 = 307

where

nf= final sample size

ni= initial sample size=384

N=total estimated neonates with their mother in the study area during data collection time=1500.

Then, after adding 10% non-response rate to get 338 samples (307 + (307/10) =338).

In Debre, Markos comprehensive, specialized Hospital, there were 1500 estimated number of neonates admitted to NICU during the study period. We determined the sampling interval (k) by dividing 1500 neonates as the study population (N) by sample size (n) as (k=N/n=1500/338=4). Then, we systematically selected 338 neonates starting from the second (2, 6, 10, 14, 18, etc.) neonates and then every fourth neonates who were admitted at the NICU of Debre Markos comprehensive, specialized hospital every day during the study period.

Note: The content above has been extracted from a research article, so it may not display correctly.

Q&A