The idea of normal distribution curve is somehow theoretical in nature. It is derived from observations from which data collected produces repeated measures with constantly increasing or decreasing intervals and produces a central point which signifies a normal curve. The most likely event where a healthcare professional is likely to achieve a normal curve is when he or she collects data from a large group of samples where such samples have got the same interval. When the data collected from such samples is calculated and plotted on a graph, the result is likely to form a normal distribution curve. This is commonly referred to as means sampling distribution. In some cases, when the results of the research are plotted on a graph they may not provide the desired normal curve of distribution. In order to transform the data in to a normal distribution, curve taking of the logarithms can be used (Schneider, 2013, p. 1954).
According to Schneider (2013), it is difficult for a researcher to achieve a perfect normal curve of distribution from a practical research. However, when the mean, mode and the median of sample data collected is the same, a normal curve of distribution will be produced. The variability of the collected data would also influence the likelihood of producing a normal curve. Where similar samples are collected the resultant distribution curve is likely to be normal as opposed to data collected from different samples. Where the healthcare researcher obtains dissimilar data from collected samples, he may calculate the standard deviation in order to harmonize the results and similarly achieve a normal curve of distribution. The standard deviation helps to establish the extent of variability from the normal curve. Hence there is a chance of the researcher to produce a normal curve where the deviation is small.