All scales and tests undergo a rigorous scientific tool development procedure to ensure that the most reliable, valid, and stable tools are used for psychometric assessment. Every psychological variable is construed differently depending on its theory, practice, and existing research literature. Hence, the specific methods to construct a scale also vary and details for each of them can be found in the research papers and test manuals. However, we discuss below some broad approaches we adopt while developing the tests and scales that have been used on this portal:
The key principles of the traditional psychometric approach include reliability, validity, standardization, norming, and scaling. The steps followed in this approach include:
The traditional approach to psychometric test development allows for the construction of reliable, valid, objective tools of measurement in Psychology. Tools developed using this approach provide measures that are quick to use for researchers and easy to understand for the public. They also generate accurate and comparable data that can be easily used for predictive analysis of real-life behaviours. It also allows for cross-cultural and comparable research. This approach uses Classical Test Theory for tool development.
The psycholexical approach (Goldberg, 1982) suggests that the most important traits and concepts for speakers of any language are encoded in the terminology of their language. To pick up a simple example, Eskimos, who live in the snow, have more words to describe snow than English speakers or Marathi speakers. Similar phenomena must occur for a description of people, their personalities, emotions, values, moralities, etc.
Psycholexical methods have been widely used to determine trait structures, taxonomy development, to make grounded investigations in languages, and lastly, for scale construction. Popularly used Big Five Inventory is based on psycholexical findings. This approach follows varied steps like extraction of a lexicon, culling of the lexicon, and large-scale data collection, in an iterative manner. Various language-specific tools on this portal are developed using this method.
Sometimes, when very little is known about a psychological construct or when the prevalence of a certain psychological construct is very localized to some regions, we use a mixed-methods approach to construct a scale. Typically, this involves conducting in-depth interviews, focused group discussions, Delphi group discussions, etc. to understand the nature of the construct. After a detailed qualitative analysis using thematic analysis or grounded theory approach, tools and scales are developed to measure the construct in an objective, reliable, and valid manner. For example, this method can be used to develop a scale to measure experience of the recent wildfires, floods, or droughts.
Psycholexical methods have been widely used to determine trait structures, taxonomy development, to make grounded investigations in languages, and lastly, for scale construction. Popularly used Big Five Inventory is based on psycholexical findings. This approach follows varied steps like extraction of a lexicon, culling of the lexicon, and large-scale data collection, in an iterative manner. Various language-specific tools on this portal are developed using this method.
One of the important objectives of M-OPC is to provide psychological tests available to people in Indian languages. This increases access and use of psychometric tests across the large geographical region. To meet this objective, translation and adaption of the tools is an important and critical step that is conducted with utmost care.
Tools are typically put through Brislin’s (1970) back translation procedure. The following steps are followed in this methodology:
If errors are found in comparison of the original and back-translated versions, the process is repeated with other bilingual experts unless concept equivalence is reached. After a validation study, the translated instrument is ready for use.