The importance of chiral separations has increased over the last years as enantiomers display different pharmacological effects. Therefore the pharmaceutical industry needs access to enantioselective stationary phases at an early stage of drug development. Today a great number of chiral stationary phases (CSP) are commercially available. Unfortunately it is difficult to predict experimental conditions and which chiral stationary phase to use for a specific compound to obtain the best enantioselectivity. This report will present an attempt to predict the relationship between molecular structure and retention behaviour of a series of carboxylic acids on tartaric acid based CSPs for a better insight into the nature of chromatographic interactions. A training set containing fifteen compounds was used to find correlations between the molecular structure of the carboxylic acids and the responses k´ and a obtained when studied on three tartaric acid based chiral stationary phases. An experimental design, where the compounds were run with normal phase HPLC, six different mobile phases containing iso-hexane, one of the modifiers 2-propanol, ethylacetate or tert-butylmethylether and the additive 0.1% formic acid were made. By using 120 molecular descriptors as variables and the responses achieved from the design, multivariate data analysis was used to evaluate the correlation between the analytes and the CSPs. The multivariate data analysis showed that there is a correlation between the molecular structure and the retention behaviour on all three CSPs, but due the fact that the models had very low Q2 values no conclusions about the correlation could be made. For improved models, the k´ values should be evenly distributed over the interval studied. This will most probably be accomplished by extending the study with more compounds and by lowering the modifier content in the mobile phases. Abbreviations CSP Chiral stationary phase DATD N,N´-Diallyl-L-tartardiamide DMB 3,5-Dimethylbenzoyl DATD DTBB 3,5-Di-tert-butylbenzoyl DATD DTBMOB 3,5-Di-tert-butyl-4-methoxybenzoyl DATD EtOAc Ethylacetate FA Formic acid HPLC High performance liquid chromatography IH iso-Hexane IPA 2-Propanol PCA Principal component analysis PLS Partial least square TBME tert-Butylmethylether TFA Trifluoroacetic acid QSRR Quantitative structure-retention relationship MLR Multiple linear regression a Separation factor k´ Capacity ratio (capacity factor) L Column length N Plate number, i.e. the number of theoretical plates in a column R Gas constant (1.986 calK mol ; 8.314 JK mol ) R Resolution factor T Temperature (K) t Retention time w Baseline peak width