Which of the following statements about principal components is true?

The statement that the proportion of variance explained by an additional principal component never decreases is accurate. Each principal component is designed to account for as much of the remaining variance as possible after accounting for previous components. Therefore, when a new principal component is added, it captures some amount of variance from the dataset, which can only maintain or increase the total proportion of variance explained, but never decrease it.

As you add more components to the model, they continue to contribute at least a non-negative amount to the total variance captured, making it impossible for the proportion of variance explained by these components to drop. This fundamental property reflects how principal component analysis (PCA) is built—it successively explains variance and optimally captures the dimensions of the underlying data structure.

The other options suggest concepts that either misunderstand how principal components function regarding variance or mischaracterize tools used for PCA analysis. The cumulative proportion of variance explained will not decrease and a scree plot can indeed be very effective in illustrating the explained variance to help decide how many components to retain.