A reform of real estate taxes which is surely unavoidable in the future and introduction of a cadastral tax will cause the necessity of a mass appraisal of real estates, in other words - value assessment of all real estates. One of the basic methods of determining market value of real estate is valuation accomplished by means of a comparative approach. This approach assumes that market value of real estate is assessed through comparison to other similar real estates, for which transaction prices and characteristics distinguishing these real estates and having the essential influence on their value are known. One of the essential difficulties of comparative approach is the necessity to have information about real estates similar to the one to be appraised, which were subjects of trade on a given property market. Generally used method of object attributes determination on the basis of maps and reconnaissance requires participation of a man (an operator). In case of mass appraisal it would cause enormous expenditure of labour, resulting in huge costs. But to solve these problems Geographic Information Systems can be used. One of their classical applications are spatial analyses allowing determination of spatial characteristics of objects and relationships between them. In the paper communication acces sibility attribute was analysed in detail. The access to real estate was considered in two categories: movement with own transportation means and using public transport facilities. The first case is more important in rural areas, but also the second case is significant because of weak public communication network. The term movement with own transportation means refers to journey along the road network from the central point, around which concentrates the life of inhabitants, directly to real estate. Such analysis can be conducted independently for each real estate, by calculation of best route to the centre, or determination of zones of specified travel times around the centre and assigning particular real estates to them. Regarding mass transport facilities, one can ride only to the nearest stop, and then has to move on foot. Such partition into two stages significantly complicates the modeling. Examination of conditions of journey to real estate was conducted based on two data models: vector and raster. In the first case functions for geographical networks analyses were used. Two tools were applied: Best Route between two points and Service Area. In case of rasters the Cost Distance function was used, which calculates the least-accumulative-cost distance to the nearest centre for each pixel. Vector analyses were limited to movement along elements of road network, unfortunately also for pedestrian traffic. As for rasters, every variant of journey forced preparation of suitable data. When travel only along roads was considered, pixels outside roads were assigned .no data. value in reclassification process. In case of raster data, there is no need to have information concerning road network, recorded as linear objects. Information coming from land register on parcels under roads or land use type road will be sufficient. Raster data makes it also possible to conduct analyses regarding movement outside the road network, which can be applied in case of pedestrian traffic examination. After suitable preparation of a movement cost raster, it also appeared possible to integrate in one analysis getting to the stop on foot and further travelling using public transportation means. An advantage of raster data is also the possibility of taking into account the centre of elongated shape, for example situated along the main street, which is often the case in countryside. However, resulting raster data sets were finally converted into vector form, containing borders of zones, for easier assignment of real estates to them. Research described in the paper confirmed the usefulness of tools for vector network analyses and raster cost analyses for determination of real estates communication accessibility. The only problem can be, as in many other cases, the availability of suitable data.