Multidimensional scaling - concept, context and coverage

Jayasankar, J (2023) Multidimensional scaling - concept, context and coverage. In: Training Manual on Advanced Analytical Tools for Social Science Research Vol.2. CMFRI Training Manual Series No. 29/2023 . ICAR- Central Marine Fisheries Research Institute, Kochi, pp. 31-43.

Text Winter School on Social Science Research_2023_Multidimensional scaling - concept, context and coverage_J Jayasankar_Vol 2.pdf Restricted to Registered users only Download (934kB)

Abstract

Social scientists are often accosted with datasets that are inherently qualitative, yet owing to computational compulsions getting converted onto scaled or interval model. Such a set of experiments that are based on ranking of multiple samples on multiple scales will always be built on preference extremes with areas of maneuver spread in between. The most sought-after inference from such a kind of experiment would be to group or discriminate sampled units based on their proximity or diversity as perceived in palpable dimensions viz two or max three, which could get mapped onto such spaces. These kinds of datasets require the best of both worlds that drive multivariate inference viz reduction of dimensionality as well as measuring proximity/ divergence. The first ensures better visualization while the latter dwells with the similarity or dissimilarity. Among a slew of tools that perform these cardinal functions in the realms of multivariate analysis, Multi-Dimensional Scaling (MDS) is an important one, worth to be understood, evaluating, and deploying in most common social science studies

Actions (login required)

View Item