Forest management can be used to optimize stand structure, and it is an effective measure to regulate forest productivity and species diversity. The influence of selective harvest, which is an important method of forest management influencing the structure and dynamics of forest stands, does, however, not achieve consistent results. Accordingly, there is an urgent need to use more comprehensive data to study selective harvest and the change in forest stands after harvest. Following the protocols of the Center for Tropical Forest Science forest dynamic plot, we constructed a 42 hm² plot in a mixed broadleaved Korean pine forest, located in Jiaohe, Jilin Province. We selected 19 ha of the plot to manage in the winter of 2011. The study was based on the managed plot, describing the harvest events using simple numeric variables and analyzing the change in the forest structure before and after the harvest event. At the same time, using the data of the second inventory of the plot after 5 years, we compared the mortality and recruitment between the managed plot and the control plot, which had the same habitat conditions at the stand and species levels. A linear mixed-effects model was employed to explore the effects of harvest on the growth of individuals. The results showed that the harvest intensity of the managed plot was 5.4%, as computed by the basal area. The species most influenced by the harvest included Acer mono, Ulmus laciniata, Juglans mandshurica, Carpinus cordata, Fraxinus mandshurica, and Tilia amurensis. Canopy tree species were harvested almost exclusively, whereas only a very few shrubs and sub-canopy trees were harvested. The harvest events did not noticeably change the species composition and diameter at breast height (DBH) distribution. During the 5 years, the stand density was lower than that of 2010 in both plots. The mortality of the managed plot was lower, whereas the recruitment status was not better, than those of the control plot. The basal area increment per year of the managed plot was larger than that of the control plot, indicating that the thinning resulting from harvesting promoted the growth of trees. We entered the predictors representing harvest intensity into the model and found that the DBH was the most significant variable for the growth analysis, followed by the asymmetric competition factor. The predictor representing harvest showed a significant effect only for the growth model of T. amurensis. In general, low harvest intensity had little effect on the structure and dynamics of the community, whereas the radial growth of different species showed various responses to selective harvest.