Correlational Designs

Correlational Designs

Non-experimental (or correlational) methods are used when a true experiment is not possible for logistical or ethical reasons.
Because non-experimental methods lack the control of experiments, we cannot establish cause and effect when we use them, but we can discover patterns.
Case Study - an in-depth study of one (or a small number) of individuals. Usually used when someone has a rare condition or set of experiences.
Survey or Questionnaire - collecting answers to a series of (usually closed-ended) questions from a group of people. Can be done by having the individual participants fill out a survey form, or by asking the questions and marking down responses
Focus group - a group interview where participants are able to and encouraged to interact with each other
Observational design - watching a person or group of people to see how they behave in a situation, with no intervention by the researcher
Quasi-experimental designs - look like experiments, but have at least one variable that looks like an independent variable but was measured, not manipulated. This is usually the case when you are interested in a variable that cannot be manipulated for either ethical or practical reasons. We cannot make cause and effect conclusions because we don't have the kind of control needed to do so.
Example: You cannot ethically assign some participants to experience childhood abuse. You cannot randomly assign people to be a particular age, sex, gender, race, have a particular household income, etc.
Practice: Non-Experimental Methods
Anna conducts a series of one-on-one interviews with nurses and finds that the nurses who report the greatest job satisfaction work with attending physicians who are perceived by the nurses as trusting their knowledge and experience. What can be concluded from this study?