Research Committee Page
Research Forum: Types of Quantitative Research
The purpose of any scientific inquiry is to add to a body of knowledge that helps explain, predict, or control events of interest in the domain. Although there are a myriad of specific research designs, each falls under one of four general categories: historical, descriptive, correlational, and experimental. Many factors can influence which type of design is most suitable for answering a particular research question.
Historical research studies events after the fact. It examines previous experience through documentation and is especially useful for studying the evolution of a particular characteristic over time. Historical research might, for example, study the influences that East Indian, African, European, and Cuban dance and rhythmic arts have had in the development of the art form now known as Flamenco. The historical method is most appropriate when some or all of the events of interest occurred in the past. Historical research is probably the only methodology available, for example, to study the influence of the French Revolution on the development of ballet. Confidence in the conclusions drawn when using the historical method is limited by the incomplete and tendentious nature of the data available to the historical researcher. Conclusions about causation are suggestive at best when using the historical method.
Descriptive research systematically documents current events, lasting products or other phenomena that can be measured directly by researchers today. Descriptive research can take the form of questionnaires, polls, surveys, or case studies. Measurements of various characteristics of ballet dancers abound in dance science and medicine literature. Physiological parameters such as body composition, aerobic capacity, strength, and flexibility have all been measured to create a profile displaying the unique qualities of the ballet dancer. Descriptive statistics (mean, median, range, variance, and standard deviation) are used to summarize and give order to the measurements made in descriptive research. While clarifying what exists is a useful first step, other research methodologies are needed to discover information that can promote positive changes for dancer.
Correlational research reflects a natural evolution from descriptive research methods. Correlational studies reveal systematic relationships between descriptive parameters (measurable features of a phenomenon under investigation). A correlation facilitates prediction of one parameter based on another. For example, a correlational study might show that ankle injuries are more frequent for dancers who rehearse and perform more often in high-heeled shoes. It might be tempting to infer from such a correlation that working in a heeled shoe causes ankle injuries, but conclusions about causation are not justified by correlational research. It might be, for example, that the injuries are actually caused by rehearsing and performing in cold and drafty studios or theatres with hard flooring and inadequate warm-ups. Correlations are restricted to prediction; the identification of causal variables requires an experimental analysis.
Experimental research reveals a cause and effect relationship by systematically manipulating one parameter (the independent variable) and observing the influence on another (the dependent variable). Experimenters might, for example, administer an abdominal strengthening program to dancers and observe its effect on the dancers' ability to maintain a stable torso in petit allegro or small jump combinations. To establish an experimental proof of causation, the manipulation of the independent variable must be done in a manner that makes any other explanation for the change in the dependent variable impossible or at least highly unlikely. A variety of experimental designs have been invented to accomplish this purpose and statistical tests have been created to assess the probability that something other than the measured independent variable may have caused the change in the dependent variable. Two major challenges in mounting an experimental analysis are the time and expense involved with such analyses.
The question to be answered and the resources available will dictate the appropriate type of research methodology. Careful application of an appropriate research design will ensure results that make useful contributions to body of knowledge in the domain of dance science and medicine.
Gay LR. Educational Research: Competencies for Analysis and Application. New York: MacMillan, 1987.
Kuhn JE, Greenfield ML, Wojtys,EM. A statistics primer: types of studies in the medical literature. Am J Sport Med. 1997;25(2):272-274.
Sax G. Foundations of Educational Research. New Jersey: Prentice-Hall, 1979.
Sidman M. Tactics of Scientific Research. New York: Basic Books, 1960.
Written by Virginia Wilmerding, Ph.D.
Virginia Wilmerding is at University of New Mexico, Albuquerque, New Mexico, USA.
This article originally appeared in the IADMS Newsletter, Spring (April) 2000.
This article may be reproduced in its entirety for educational purposes, provided acknowledgement is given to the "International Association for Dance Medicine and Science."
Copyright © 2011 International Association for Dance Medicine and Science (IADMS) and Virginia Wilmerding, Ph.D.