The impact of the so-called "twist & curl"-effect on the production process of liquid packaging paperboard at the company SIG Combibloc has proven to be rather problematic in terms of quality assurance. Consequently, it is the aim of the current study to characterize the hitherto unknown causes of this processing problem via statistical data analysis of the recorded process data. One method that is well suited to characterize the causes of the "twist & curl"-effect is applying statistical process data analysis procedures. It was shown that, using the IBM SPSS statistical analysis software, the process induced formation of this detrimental effect can be made visible after the production process, if the quantity and quality of the collected data is sufficient. Based on the results of the regression analyses, it was established that the influence of the paperboard’s moisture content is one of the most significant influences on the development of the effect. But in many cases, the effect will be caused by a combination of several factors, all of which may adversely affect the distortion of plastic composites. In one of the three cases that were investigated, the inferior quality of the paperboard with regard to insufficient moisture content, paperboard weight and tensile strength is suspected as the cause for the development of the "twist & curl"-effect. In the other two cases, the problems in the production process were likely caused by process variations due to adjustments of process parameters during the production of new materials and thicknesses as well as breaks of the board roll caused by shutdowns of the coating plant. The most significant impact in the binary logistic regression analysis for each case study was proven to be the increased thickness of the inner coating, referred to as „Grammatur Innen 1“. In the polymers used for this layer (EAA, PA6), differences in the crystallization temperature ranges and the crystallization behaviour, as compared to the most commonly used homopolymer (LDPE), were determined. The impacts of such differences in shrinkage are in some cases amplified through temporary increases in layer thickness or higher cooling and melt temperatures. The resulting increases in degrees of crystallinity lead to an increase in hindered shrinkage which ultimately results in warping effects in the compounds. In addition, the observed results of the differential scanning calorimetry measurements showed a higher degree of crystallinity in the inner layer and a slightly lower degree in the outer layer. This fact is probably due to a short-term change in the processing temperatures and strengthens the effects of the differences in the crystallization behaviour of the varying polymers. In summary it can be said that with adequate data collection the process-induced formation of the undesired effect can be described by applying a statistical process data analysis using the software SPSS.