Instructions: Read the following description and answer questions below.

"To investigate whether musical stimuli selected on the basis of these data represented the hypothesized emotional dimensions of happy, sad, and neutral valence, a principal component analysis (PCA) [Cattel, 1996] was carried out on the Pearson correlation matrix of participants' mean ratings for all stimuli. Three factors were requested in accordance with the hypothesis. Factors with an Eigenvalue of larger than one were accepted, following standard procedures [Coolican, 1994]. An orthogonal rotation was used (Varimax), resulting in uncorrelated factor scores, in order to maximize the statistical independence and interpretation of the expected emotional dimensions [Kaiser, 1958]. The PCA yielded three factors that could clearly be identified from the rotated component matrix and the scree plot. These factors were labeled as ''Happy'' (Factor 1: Eigenvalue 13.17), ''Neutral'' (Factor 2: Eigenvalue 9.21), and ''Sad'' (Factor 3: Eigenvalue 4.47) musical stimuli. The factor structure accounted for 45% of the variance. Items that did not load onto its expected factor and had low communality (<0.3) were excluded from the stimulus selection."

1. 2)If this other method was used, how might the factor loadings be expected differ compared to principal components analysis? (1 mark)
2. 3)To someone unfamiliar with factor analysis, interpret what it means for items to have a communality of <0.3? (2 marks)