The validation study of the prediction method for indoor air pollution was carried out by comparing the results of emission models based on data obtained in a large and a small emission chamber, with actual measured concentrations. A new decorated room was studied as a case. Emissions of complicated objects and simple surface layer materials were studied respectively in the large and small chamber and emission models were developed. Those models were based on the assumptions regarding mass conservation of substances and the hypothesis that pollutants were well mixed. The emissions of formaldehyde and TVOC (total volatile organic compounds) in the studied room were predicted by the method. The predicted concentration trend of pollutants was in accordance with the measured trend when some air exchange (0.03 ACH, air change per hour) was taken into account. The normalized standard errors of formaldehyde and TVOC pollution prediction were respectively 2.8% and 1.6%. Modeling analysis shows that the contribution to total formaldehyde pollution of the studied room was: furniture > paint > floor; the contribution to total TVOC pollution was: paint > floor > furniture. The results lead to the conclusion that this prediction method can well describe the pollution trend, can assess the contribution of different sources, can guide the choice of building materials and is an effective tool for indoor air pollution assessment and control.