Landscape composition plays an important, but poorly understood, role in the population dynamics of agricultural pest species with broad host ranges including both crops and weeds. One such pest, the generalist plant bug Lygus hesperus Knight (Hemiptera: Miridae), is a key cotton pest that feeds on various hosts differing in quality in California's San Joaquin Valley (USA). We investigated the effects of 15 common crops and uncultivated agricultural land on L. hesperus populations, by correlating the densities of L. hesperus in focal cotton fields with the areas of the 16 crops in surrounding rings. Insect counts were provided by private pest-control advisors, and spatial data were obtained from Kern County records. We first calculated Spearman's partial correlation coefficients on an annual basis for each crop separately, and then performed a meta-analysis of these correlations across years to describe the overall effect of a particular crop on L. hesperus after the effects of the 15 other crops are removed. Consistent with studies conducted in other areas, L. hesperus density was positively correlated with safflower, and negatively with cotton. Lygus hesperus density was also correlated with several other crops that are often not considered in pest management, including grape, oat, and onion (positive correlations), and almond, pistachio, and potato (negative correlations). Lygus hesperus density was also found to be negatively correlated with alfalfa and positively correlated with uncultivated habitats, a relationship that receives mixed support in the literature. Several other crops tested were not significantly correlated with L. hesperus densities in focal cotton fields, suggesting a neutral role for them in L. hesperus dynamics. The improved understanding of the effects of a greater variety of crops on L. hesperus population dynamics will be useful in the design of agricultural landscapes for enhanced management of this important polyphagous pest.