Reaction kinetics and composition of 4-vinylbenzenesulfonic acid sodium salt (VB)-acrylamide (Aam) copolymerization in 0.1M NaCl solution are investigated. Data obtained by the automatic continuous monitoring of copolymerization system, up to 80% conversion, are analyzed by an ''error-in-variables method'' developed for obtaining the reactivity ratios by on-line monitoring. Monomer reactivity ratios are found as r Aam ¼ 0.085 6 0.020, r VB ¼ 2.0 6 0.33. Although the terminal model describes the composition data well, it is seen to be inconsistent with the reaction rates. This discrepancy is attributed to implicit penultimate effects and using the recently developed calculation method, effective radical reactivity ratios are found as s VB ¼ 0.26 and s Aam ¼ 0.027, and both composition and rate data fit the implicit penultimate model extremely well. On-line monitored data showed that in the reactions where the VB was completely consumed, the subsequent Aam homopolymerization was very rapid; thus, the reaction showed definitely two rate regimes, before and after VB depletion. Acrylamide take up rate also showed these two rate regimes. We conclude that low conversion results can be misleading and reactions must be monitored up to a high conversion for a robust control of composition and reaction kinetics.