Rod worth is measured by how it changes keff, that is by the Δk it produces as it relates to keff, or in other words, how much reactivity is removed when the control rod is inserted, or how much is added with the control rod is withdrawn
Individual rods have worth, and banks (sets) of rods have worth, and that will vary according to reactor conditions (hot vs cold, zero power vs hot full power), enrichment (including fissile Pu content), and burnable absorber (either in liquid form, e.g., boric acid in a PWR), or boron in the form of pyrex in special assemblies, or burnable absorber such as gadolinia, erbia, or ZrB2 in the fuel rods. In other words, it depends on reactivity distribution in the core.
One may refer to differential rod worth, i.e., how does the reactivity or Δk change with each step or fraction of control rod insertion/withdrawal, or integral rod worth, in which the reactivity effect is measured with a rod fully inserted or fully removed.
Rod worth is imported for shutdown margin, as for evaluation of transients such as reactivity insertion accidents, RIA.
With regard to control rod interference, it relates to the rate at which a control can insert reactivity into the core. Some commercial plants, both PWR and BWR, have experienced issues of control rod/blade interference. In some PWRs, assembly (guide tube) distortion has been an issue, while in BWRs, channel bow and bulge have caused slower insertion of control blades. Fuel management is a factor, as is fuel design.