A type of research design in which patterns of correlations are analyzed.
Descriptive Studies Descriptive studies are used to describe general characteristics of the distribution of a disease, particularly in relation to person, place and time (1). Descriptive studies can be further classified into correlational (ecological) studies, case series and cross-sectional surveys.
Correlational (Ecological) Studies Correlational studies use data from entire populations to compare disease frequencies between different groups.
For example, mean per capita income could be compared to death due to coronary heart disease for several countries.
Correlations studies have a number of limitations. Since there are many populations characteristics that could have been chosen for examination, they often tell as much about the beliefs of the investigators (i.e. that income and coronary heart disease are somehow linked) as they do about the disease under study. They also provide no information about individuals within a population. We don't know whether those individuals within a population who have higher incomes were actually those who died of coronary artery disease, only that on average those countries with greatest average yearly income also have the highest rates of coronary artery disease (ecological fallacy).
In addition, there may be other differences between the populations that are associated with the exposure under study that may play a more important role in the disease. For example, if those with a higher income were more likely to eat diets high in fat and/or smoke more these would be more likely causes for increased coronary artery disease mortality. In this case income would merely be a marker for dietary and smoking habits.
Therefore, although correlational studies are useful for generating hypotheses they have many limitations and can be very misleading at times.
With a comparative descriptive design, the researcher describes two or more groups of participants. For example, a researcher administers a questionnaire to three groups of teachers about their classroom practices. The researcher chooses the three schools because the schools vary in terms of the amount of professional development that they provide to teachers.
A correlational research design is used to describe the statistical association between two or more variables. For example, a researcher measures the student-teacher ratio in each classroom in a school district and measures the average student achievement on the state assessment in each of these same classrooms. Next the researcher uses statistical techniques to measure whether the student-teacher ratio and student achievement in the school district are connected numerically; for example, when the student-teacher ratio changes in value, so does student achievement. The researcher can then use the student-teacher ratio to predict student achievement, a technique called regression analysis. When there is more than one predictor variable, the technique of multiple regression analysis produces a multiple correlation that is used for prediction.
Research designs can be divided in to two broad categories::
* Non-experimental research – In this category the researcher observes the phenomena as they occur naturally and does not intervene in any way. Types of designs in this category include:
o Descriptive research
o Correlational research
Learning research can be difficult because it is not exactly black and white, which would make it easier to comprehend. Designs are not as rigidly defined as the book indicates. A study can use a combination of designs (most sophisticated studies do). As noted above, a longitudinal descriptive study of developmental outcomes is undoubtedly going to also be a correlational study and will likely conduct predictive correlational analyses with their data.