Human beings are inquisitive creatures. Since the beginning of time, they have measured, categorized and attempted to understand the world around them. On a daily basis, you can hear these efforts on the evening news and read them in your favorite magazine. Furthermore, to meet this goal, scientist and intellectual explorers arm themselves with the science of statistics. Statistics are applied all in the name of learning something new, proving a point, or perhaps reluctantly admitting there is no connection between two phenomena. The concept of descriptive statistics, the application of descriptive statistics to my own study using a small data set, as well as an example of correlation will be examined.
For me, the best thing to do when overwhelmed by something is to organize it. In addition to the organization of data, it is desirable to share information about the data in a transparent and concise manner. According to Blessing and Forister (2013), when utilizing descriptive statistics to organize your data the “key concepts are measures of central tendency and measures of variability” (p. 194). These concepts are accomplished by using mean, median, mode, as well as, standard deviation, range and standard error. There value lies in the ability to give information about a set of data, without giving all of the data. There are various situations suited to the different elements of descriptive statistics and it is important to keep in mind the type of data you are using (ordinal, interval, ratio) when selecting the descriptive statistics for your study. In my research study regarding telehealth and 30 day readmission rates, descriptive statistics would be interesting to observe central tendencies for age of the patient at the time of readmission, type of cancer (liquid vs. solid tumors), and type of treatments (chemotherapy, radiation, surgery). Of course, with enough time and money, the variables you could study are endless.
The mean is the sum of the scores divided by the total number of scores and “can only be calculated for interval or ratio variables” (Blessing & Forister, 2013). In my research study looking at the role of telehealth (teleconferencing) in influences 30 day readmission rates, the mean would be helpful to know when looking at the average numbers of days are from discharge to readmission. For example, the mean of the following data points (15, 18, 22, 22, 25 and 30) is 22 days. In this scenario, a researcher could use this information to fine tune the intervention, or make recommendations about future studies. Using the same data points, the median would be 22 since there are even numbers of scores. Of interest, the median is useful when studying the “skewness” of a curve (Grove, 2007, p. 139). Finally, the mode is also 22 because it is listed the most in my data set. In a normal distribution, the mean, median, and mode may be similar (or the same) as seen in my example.