Operational definitions in scientific research and variable measurement

¶ … attributes of operational variables as used in empirical research. In addition, information is provided about experimental designs, probability, experimental bias, the placebo effect, and pseudoscience.

Identify a variable for scientific study that can be operationally defined. One of the most common variables that is operationally defined in psychology is intelligence. Intelligence is measured through the use of standardized tests, and reported according to the test outcomes on verbal and other scales. Intelligence is a difficult construct to pin down, so associating the construct of intelligence with the instrumentation that will be used to measure the variable known as intelligence provides a measure of control and agreement about just what is meant by intelligence in a particular study..

Create an operational definition for the variable. The operational definition of intelligence is that intelligence is an attribute and capacity of the human mind that is measured by intelligence tests. Using the Stanford Binet to measure intelligence, both mental age and an intelligence quotient are reported. Mental age refers to the chronological age of children at which an average IQ test score is in a particular range. Intelligence measurement is derived by dividing mental age by chronological age and multiplying by four.

Identify what aspects of the variable would be observable and measurable. A non-scientific definition of intelligence would refer to the ability to think abstractly, rationally, act purposefully, and interact with the environment in an effective manner.

• Lecture on non-experimental designs (ex post facto, correlational, survey method: Four criteria, and case study), role of chance, experimental bias, placebo effect, and pseudoscience.

Non-experimental designs (ex post facto, correlational, survey method: Four criteria, and case study). The four types of non-experimental research designs are: Descriptive, relationships (comparative relationships & correlational relationship), causal-comparative, and survey. In ex post facto research, the independent variable is not manipulated and subject "sampling" is based on naturally occurring groups -- this is a retrospective type of research, in that, it takes previously existing groups and follows them forward.

Role of chance. Probability is a core consideration in evidence-based research. Considerations about sample size and confidence intervals and inferential statistics have to do with probability -- specifically, the probability of assuming that a hypothesis is true when it is not, or that a hypothesis is not true when it is. Research design criteria and standards are intended to preclude any significant role of chance in a scientific experiment.

Experimental bias

Placebo effect. The placebo effect is a very real phenomenon that demonstrates the influence that the mind has over the physical body and its own mental states. In empirical studies using control groups, one group is typically assigned a placebo -- some substance or treatment that is so similar to the experimental condition (independent variable) that there is a high likelihood that the subjects will mistake it for the real substance or treatment. Any changes to the subjects in that group may then be attributed to the placebo effect, which informs the influence of the experimental condition (Independent variable).