Environmental datasets often consist of numerous features analyzed in many investigated samples. Therefore, the evaluation of those datasets is difficult. Chemometric methods like the factor analysis are useful tools to handle big datasets. In this paper, we discussed the relation between the geogenic background (noise) and anthropogenic pollution (source) for the suitability of environmental datasets for factor analytical methods. Thus, computed test datasets with different sources, diverse maximum of the sources, and various geogenic backgrounds were generated. Afterward, the maximum of the source was decreased stepwise, a factor analysis was computed, and the corresponding results were investigated in respect of the credibility. The major impacts on the evaluation of a feature are the mean value of the noise and the standard deviation of the noise. With the help of these two parameters, a pollution index can be calculated. The maximum of the source has to exceed that index in order to be usefully evaluable with the factor analyses. The evaluation of the results of the factor analysis would become increasingly complicated if the variability of a dataset decreases due to reduced maximum values or geogenic/anthropogenic sources which correspond to increasing environmental quality.