Work sampling is a very effective tool for measuring and monitoring the performances of construction crews. Work measurement was conducted at 3 different sites with their unique qualities. Productivity was measured for crews comprising of one mason (skilled worker) and one helper (unskilled worker)-1M (Mason)+1H (Helper). Enough observations were made to ensure a 95% confidence limit and ± 5% accuracy of sampling results. Block work and internal wall plastering were chosen for work study. These works were divided into their elements. Sampling was done by categorizing work elements as Direct Work (DW), Indirect Work (IW) and Not Working or Idle (NW). The regression analysis on the work sampling data shows that observed quantity is having very low correlation with DW content and satisfactory correlation with working content (DW+IW). Results of sampling show that shares of DW, IW and NW are 56%, 27% and 17% for observed works respectively. Labor Utilization Factor (LUF) given by the sum of direct work and half of indirect work is around 69% for all works. The values of LUFs for all sites are in good range and help in monitoring trends. The direct work content is almost twice as indirect contributory work for both works in all sites. 1M+1H crew composition is found to be optimum and economical compared to 2M+1H and 1M+2H combinations. The major cause of delays in all sites is found to be waiting for materials and equipment.