The identification and characterization of low levels of long-chain-branching (LCB) in essentially linear polyethylenes has attracted significant interest in recent years. One experimental technique is nuclear magnetic resonance (NMR), which can detect LCB in essentially linear polyethylene homopolymers for LCB in the range 0.2-3 branches per 10 000 carbon atoms. Another approach has been the use of a rheological measurement in combination with a dilute-solution measurement (intrinsic viscosity or GPC). NMR is a direct method of LCB measurement, but it provides no information on branch length and has other limitations related to interference from short chain branches. Rheology provides a sensitive but indirect method of measurement, capitalizing on the strong effect of LCB (longer than M, the entanglement molecular weight) on translational mobility of the polymer chains and thus viscosity. The purpose of the present work is to provide an assessment of the two approaches of LCB determination, using a series of well-characterized, essentially linear polyethylenes. NMR was shown to work satisfactorily in characterizing LCB for a series of metallocene-catalyzed polyethylenes, but it failed to detect LCB in other cases, including a series of linear polyethylenes where LCB was introduced deliberately via peroxide modification. A rheology-based index for LCB characterization was shown to be preferable, due to its robustness and general applicability in all cases examined.