Chapter 10 Overall Conclusion

With respect to SAM Dataset

PCA with Varimax roation revealed important principle components represented by the variables. This could effectively help with dimensionality reduction.

MCA able to capture the non linear relationship among the levels of the variables.

Between group pattern better explained by DiCA > BADA. DiCA was also able to correctly assign the gorups unlike BADA which catogorized all participants as normal memory. Accuracy for random effect was almost equal to fixed effect, in case of DiCA.

PLSC was good at figuring out the correlation between the memory types of participants and their preference for mental imagery.

MFA was able to find a pattern between the SAM OSIQ and BFI, i.e. find a pattern between memory types, their preference for mental imagery and personality traits amongst the participants.