Area correction in direct shear test will not alter the friction angle, hence many tend to ignore it. However, correction to the area is important to obtain accurate shear stresses, especially while testing saturated soils. If you ignore the area correction, the area in which the soil is in contact with the box is unaccounted for and leads to the estimation of apparent shear strength, which is the area correction times the actual shear strength.

Thank you for the very good quesiton. Actually, we should cosider this problem on the area correction. But, in fact it is somewhat difficult to counter.  It is worth doing such reseach or reviseing the traditional shear strength device.

Area correction should be done in accordance with IS: 2720 (Part 13) - 1977 Reaffirmed 2010. Although phi value is not significantly affected, the stress-strain plots will be wrong.

The correction of the shear plane should always be made, of course, when it is possible. For example, in shear test in situ made on incoherent materials (sand, gravel, rock fill, and the like) it is important to make corrections in the area, because displacement is expressed, until approaching the shear strength. In this way the contact surface is reduced. and the calculation of the stress is done taking into account the reduction in surface contact. That can be easily programmed to obtain diagrams shear stress - shear strain.

It is also advise able to perform a large scale direct shear test

Thanks all for suggestion and recommandation