Rather than pursue a "univariate" research approach to psychology, studying the effect that a single variable (such as "dominance") might have on another variable (such as "decision-making"), Cattell pioneered the use of multivariate experimental psychology (the analysis of several variables simultaneously). He believed that behavioral dimensions were too complex and interactive to fully understand variables in isolation. The classical univariate approach required bringing the individual into an artificial laboratory situation and measuring the effect of one particular variable on another - also known as the "bivariate" approach, while the multivariate approach allowed psychologists to study the whole person and their unique combination of traits within a natural environmental context. Multivariate experimental research designs and multivariate statistical analyses allowed for the study of "real-life" situations (e.g., depression, divorce, loss) that could not be manipulated in an artificial laboratory environment.  Cattell applied multivariate research methods across several intrapersonal psychological domains: the trait constructs (both normal and abnormal) of personality, motivational or dynamic traits, emotional and mood states, as well as the diverse array of cognitive abilities. In each of these domains, he considered there must be a finite number of basic, unitary dimensions that could be identified empirically. He drew a comparison between these fundamental, underlying (source) traits and the basic dimensions of the physical world that were discovered and presented, for example, in the periodic table of chemical elements.  In 1960, Cattell organized and convened an international symposium to increase communication and cooperation among researchers who were using multivariate statistics to study human behavior. This resulted in the foundation of the Society of Multivariate Experimental Psychology (SMEP) and its flagship journal, Multivariate Behavioral Research. He brought many researchers from Europe, Asia, Africa, Australia, and South America to work in his lab at the University of Illinois. Many of his books involving multivariate experimental research were written in collaboration with notable colleagues.

Answer this question "what was being used before the multivariate?" by extracting the answer from the text above.
He believed that behavioral dimensions were too complex and interactive to fully understand variables in isolation.