Research Design Principles

Research Problem

The research problem is an area of concern that the investigator is interested in exploring. It reflects a gap in knowledge about a specific social issue or problem discussed in the scientific literature. Often a research problem focuses on a situation that arises out of everyday life and that calls for a solution, an improvement, or some alteration; it is a problem that needs to be solved. Conceptualizing and describing the social problem lead the investigator to identify the most salient concepts or variables involved and guide the investigator's thinking about why and how the variables are connected to each other. Through this process of articulating and framing the research problem, investigators may discover that little is known about an issue or that there is conflicting information about it within the scientific literature. Depending on what is known about the research problem, the types of specific research questions to be asked and the purpose and objective of the research study could range from basic exploratory research to control experiments.

Research Questions and Hypotheses

Research questions involve examining a relationship between two or more variables. The relationship between variables can be expressed as either positive or negative correlation or causal. An example of a research question about a correlation would be, Are neighborhood characteristics of crime and low-income related to health status?

Causal relationships describe the cause-and-effect relationships between variables. Three criteria are used to assess causal relationships: (1) correlation between the variables, (2) temporal precedence or sequence, and (3) no alternative explanation. First, there must be a significant positive or negative correlation between the two variables. That is, as the value of one variable changes, the value of the other changes as well. Second, there is a defined temporal sequence of the variables. The variable that is considered to be the cause (the independent variable) must precede the effect variable (the dependent variable) in time. The temporal relationship is often diagrammed as A → B or expressed as, if A then B. Third, the cause-and-effect relationship between the variables cannot be plausibly explained by the presence of another variable. This is called the third variable problem, which is reduced or eliminated by controlling specific methodological procedures used to implement a particular research design. These controls attempt to combat specific threats to internal validity that can negatively affect the inferences made from a study. Examples of research questions examining a causal relationship are, Does cognitive behavioral therapy reduce trauma symptoms among survivors of domestic violence? and Does neighborhood cohesion reduce community crime?

...