Statistics represent one of the highly significant elements of research in modern times with respect to the manner in which it sorts data into quantifiable forms. Some people confuse inferential and descriptive statistics, which makes it difficult for them to choose the best alternative to employ in their research. However, a close look at the two forms of statistics shows the difference as it is evident in their names (Spriensma, 2016). Descriptive statistics depict the data collected, whereas inferential statistics deduce or enable the researcher to make a conclusion anchored in the gathered information.
I was at one time undertaking a research study regarding teenage pregnancy in a given high school with the help of descriptive statistics. In this assignment, I was investigating the prevalence of teenage pregnancies in the learning institution for five years. Using descriptive statistics, I was just summarizing the gathered data and identifying the pattern in variations. In the research, I found that in the course of five years, the highest proportion of teenage pregnancies occurred among the students in their third year. There was no need of demonstrating that in the sixth year, the third year learners were still the ones with the highest rate of teenage pregnancies. Such conclusions, in addition to forecasts, are made in inferential statistics but not in descriptive statistics (Kalish & Thevenow-Harrison, 2014).
The approach of describing or making conclusions applies to the data or the gathered information of the researcher. Just like in the earlier instance concerning teenage pregnancies, descriptive statistics are just limited to the population being studied. In simple terms, the data gathered in the given high school was only applicable to that learning institution. Were it in inferential statistics, the chosen institution could have acted as only a sample of the target population.
For instance, if a researcher is seeking to establish the rate of teenage pregnancies in New York schools, one high school can thus be a sample to represent all the learning institutions in New York City (target population). Nonetheless, this normally takes into consideration a margin of error as a single sample is not sufficient to reflect the entire population (Spriensma, 2016).
Statistics, particularly inferential, is greatly significant in contemporary industries, mostly because it offers details that have the possibility of assisting people in arriving at future decisions. For instance, applying inferential statistics on the level of population growth in a given city may act as a foundation for a company to choose whether it ought to open a plant in that region or not. The reality that such a company uses numbers to make its decision boosts the accuracy of the research, in addition to the understandability of the information (Murphy, Myors, & Wolach, 2014). To enhance precision, researchers also consider different aspects that may influence the population and convert it into numerical data. In this manner, the probability of inaccuracy is reduced, and a summarized analysis of the study is realized.
Descriptive statistics just explain the study without allowing forecasts and conclusions. On the contrary, inferential statistics enable the researcher to make conclusions and forecasts of the variations that might take place concerning the subject of the study. Descriptive statistics normally act in a given area that holds the entire target population, whereas inferential statistics use a sample of the population, as the population under the study is typically too big to carry out the research on.
Kalish, C. W., & Thevenow-Harrison, J. T. (2014). Descriptive and inferential problems of induction: Toward a common framework. Psychology of Learning and Motivation, 61, 1-39.
Murphy, K. R., Myors, B., & Wolach, A. (2014). Statistical power analysis: A simple and general model for traditional and modern hypothesis tests. Abingdon, United Kingdom: Routledge.
Spriensma, A. (2016). Descriptive versus inferential statistics. Journal of Endourology, 30(1), 1-4.