In-plane shear is considered as the main deformation mechanism during the forming of fabrics on double curved geometries. Non-Crimp Fabrics (NCFs) are more and more used in the industry thanks to their high mechanical performances. The uniaxial bias extension (UBE) test is commonly used for characterizing the in-plane shear behavior of fabrics. However, presence of slippages calls the reliability of this test into question for NCF material. These slippages lead to a macroscopic kinematic which does not respect the fundamental hypotheses of UBE test theory. The variety of NCF architectures is usually pointed while the lack of standardized experimental methods is seldom discussed. The first section of this paper presents a two-way approach to detect slippage on an NCF. This approach is based on two kinematical descriptions of the UBE test. The first one assumes a pure shear behavior whereas the second one assumes a simple shear behavior. These behaviors correspond respectively to the rotation of fibers and to the slippage of fibers from a macroscopic point of view. In the second section, the two-way approach is used to analyze experimental UBE tests. This investigation highlights the influence of the sample width on the deformation mode during a UBE test. More precisely, it is shown that increasing the sample width of NCF specimens improves the UBE test reliability.