In social sciences, most variables that researchers deal with are not directly measurable because of the nature of the variable, which is abstract and not tangible for direct observation. For example, you cannot measure intelligence such as a physicist measuring the temperature. You must change the variable that you want to measure into something measurable and we this process is called ‘operationalization’ (Bhandari, 2020). Continuing with the example of intelligence, a psychologist & scientist should operationalize it in order to measure the concept of intelligence, and considering how to measure intelligence concretely, s/he should use a suitable device or method if available, and if not, s/he should try to develop a suitable method to be able to measure. Thus, if a scientist wants to measure the intelligence which is an abstract concept that cannot be directly measured, the first step is to convert the notion into something concrete and measurable and then the measurement can be made as wished. Today, there are intelligence scales that are already available to scientists and have proven validity and reliability. The scientist we exampled above, can use the valid and reliable intelligence tests to measure intelligence. Operationalization is one of the first and most critical stages of doing research in social sciences and especially psychology.
Construct validity is about how accurately the independent and dependent variables in the study can represent the hypothetical variables. If a test or scale claims construct validity, it means that the aimed and desired variables are measured precisely. Operationalization and operational definitions of the variables are important for ensuring the construct validity. For example, you can create a study to measure whether a specially designed intervention program will reduce drug addiction among young adults. Construct validity is a measure of whether your research is actually measuring drug addiction or not. Construct validity is an essential and important, yet some of the well-known tests’ construct validity is a debate in the psychology literature so selecting a measurement tool for a research is critical.
One of the measurement tools used often is surveys. Surveys can be defined simply as asking people their opinions on certain subjects. There are two main types of survey questions: Open-ended questions and closed-ended questions. Open-ended questions can be answered anyway the participant wants to answer. There are no clear lines or types of answers that can be given, rather the answers can be given are immense. An example for open-ended questions would be “Which course is the best in Gelisim University?”. Open-ended questions would be rich and have depth (pros), but it is hard to code and compare them (cons). On the other hand, closed-ended questions have 2 branches as forced-choice questions and likert-scale rating questions. In forced-choice question type, the researcher determines some specific and precise answers for each question in advance and the respondent only allowed to choose among those pre-selected answers. For example, “Which course is the best in Gelisim University? A) Developmental Psychology, B) Introduction to Management, C) Advanced English, D) Fundamentals of Law”. The upside of a forced-choice question is that it is extremely easy to code and analyze, but the downside of it is the possibility that not all alternatives are among the choices. Last but not least, likert-style rating questions are frequently used in the field of psychology. The answering method is based on a scale which the participants can rate their opinions on that scale. For example, a sample could be like this:
“Females chew more gum than males.”
1 2 3 4 5
Strongly Somewhat Neither Agree Somewhat Strongly
Disagree Disagree Nor Disagree Agree Agree
The positive side of using likert-scale rating questions is that they do not need to be coded because participants are already rating their own opinions on a scale. However, there are some negative sides of using this type of measurement. For example, participants may more willing to give socially desirable answers which will affect the data in an unwanted way, or participants may answer the questions without thinking on it, or may be mislead by the wording of the question itself.
Research Assistant Dilara Nihal Çarıkçı
Istanbul Gelisim University Psychology Department
REFERENCES
Bhandari, P. (2020, June 26). Operationalization: A guide with examples, pros and cons. Retrieved April 28, 2021, from
https://www.scribbr.com/dissertation/operationalization/
