Are social contact influenced by housing design? It is well known from sev-eral studies that social interaction between neighbours is more intense insingle family dwellings than in blocks of flats.But it is not easy to single outhousing design as the most significant explanatory variable because of thedifferencies in household structure etc that are usually found when compar-ing the two types of housing. We have studied a large residential block -mainly by mailed questionnaries-where half of the dwellings have a designsimilar to terrace houses with access from broad balconies, and the otherhalf are ardinary confirms that the households in the terrace-like houseshave more contacts with one another and are better known by the other ten-ants. There were no differencies between the tow dwelling types when itcomes to household structure, dwelling size or other kinds of social inter-action, such as with relatives friends, that could explain the diffrent con-tacts with neighbours.The large residential block that we investigated is built some twenty yearsago around a square yard. Two years ago this countyard was covered by aglass roof, incroporating all windows facing the courtyard. Most of thepresent residents lived in the block before this happend, and our study wasmade a year after. We have found changes in the residents contact withneighbours as well as changes in evaluation and use of dwellings, balconiesan private patios. All of these changes can be related to the glass cover. Therebuilding process in itself can generally be expected to have a positive ef-fect on contacts between neighbours. Our study gives some evidence of this.Still, type of dwelling-terrace house or ordinary flat- has had the greatestinfluence on the social interaction between the residents in the block.