The corn stem borer, Sesamia cretica Lederer, and the beet army worm, Spodoptera exigua Hubner, are important corn pests in Khuzestan province which cause severe damage in corn fields every year. Sampling is one of important principles of integrated pest management program, especially for population density monitoring. Determination of the best sample universe is critical to perform correct, fast and low cost sampling of pest population with acceptable precision level. In this study, the best sample universe to relative estimate of the pest population in corn fields, with weedy and non-weedy conditions, were studied during two agricultural seasons (2016-2017). Two corn fields, one with weedy condition and another without weedy condition (each field area was 10000 m2), were used in this study in Susa district. Samplings were weekly done from three different parts of corn plant, down, middle and top parts. The best sample universe was determined using comparisons of regression correlations of total density and recorded population from each sampled parts by goodness of fit tests and comparisons of correlation coefficient values. The best sample universe was different according to the pest species. Results indicated that the best sample universe for the corn stem borer was middle part of corn plant. For the beet army worm, the down part of the plant can be used as sample universe for monitoring of the pest larval population. Weed condition was not significant effect on the sample universe. Results of the research is critical to develop integrated pest management programs of corn pests.