The estimation of the postmortem interval (PMI) from skeletal remains represents a challenging task in forensic science. PMI is often influenced by extrinsic factors (humidity, dryness, scavengers, etc.) and intrinsic factors (age, sex, pathology, way of life, medical treatments, etc.). Raman spectroscopy combined with multivariate data analysis represents a promising tool for forensic anthropologists. Despite all the advantages of the technique, Raman spectra of skeletal remains are influenced by these extrinsic and intrinsic factors, which impairs precision and reproducibility. Both parameters have to reach a high level of confidence when such spectroscopy is used as a way to predict PMI. As a consequence, advanced multivariate data analysis is necessary to quantify the effect of all factors to improve the estimation of the PMI. The objective of this work is to evaluate the effect of intrinsic and extrinsic factors on the Raman spectra of skeletal remains. We designed a protocol close to a real-world scenario. We used ANOVA-simultaneous component analysis (ASCA) to unmix and quantify the effect of 1 intrinsic (source body) and 1 extrinsic (burial time) factors on the Raman spectra. In our model, the burial time was found to generate the highest variability after the source body. ASCA showed that the variability due to the burial time has 2 mixed contributions. Seasonal variations are the first contribution. The second contribution is attributed to diagenesis. A decrease in the mineral bands and an increase in the organic bands are observed. The source body was also found to contribute to the variability in Raman spectra. ASCA showed that the source body induces variability related to the composition of bones. This quantification cannot be assessed by basic chemometrics methods such as PCA. The results of this study highlighted the need to use an advanced chemometric data analysis tool (like ASCA) combined with Raman spectroscopy to estimate the postmortem interval.