Using triangulation to evaluate findings from random-intercept cross-lagged panel models: An application with data on curiosity and creativity

In a recent study, researchers found cross-lagged effects between curiosity and creativity in an analysis with the random-intercept cross-lagged panel model (RI-CLPM) and concluded that curiosity and creativity mutually reinforce each other. However, it is known that the RI-CLPM can give biased results. Here, we used triangulation and analyzed the same data (N = 400) with additional models, including a latent change score model (LCSM) and multilevel regression analyses of person-mean centered scores. Only results from the original RI-CLPM were consistent with the conclusion of mutually reinforcing effects between curiosity and creativity while results from the other models contradicted this conclusion. Moreover, the model of spurious longitudinal associations (MoSLA) suggested that data might have been generated without any direct effects between curiosity and creativity. An aggregation of available evidence made us conclude that longitudinal associations between curiosity and creativity in the present data probably were spurious, possibly due to confounding by a trait common to curiosity and creativity and common auto-correlated state factors with effects on curiosity and creativity measured at the same occasion. The present study, and the available analytic script, can be used as a model/tutorial by researchers wishing to scrutinize results from the RI-CLPM.