Purpose: This study aimed to investigate the time course of meaning activation of the 2nd morpheme processing of compound words during Chinese spoken word recognition using eye tracking technique with the printed-word paradigm. Method: In the printed-word paradigm, participants were instructed to listen to a spoken target word (e.g., [Chinese characters omitted], /da4fang1/, generous) while presented with a visual display composed of 3 words: a morphemic competitor (e.g., [Chinese characters omitted], /yuan2xing2/, circle), which was semantically related to the 2nd morpheme (e.g., [Chinese characters omitted], /fang1/, square) of the spoken target word; a whole-word competitor (e.g., [Chinese characters omitted], /lin4se4/, stingy), which was semantically related to the spoken target word at the whole-word level; and a distractor, which was semantically related to neither the morpheme or the whole target word. Participants were asked to respond whether the spoken target word was on the visual display or not, and their eye movements were recorded. Results: The logit mixed-model analysis showed both the morphemic competitor and the whole-word competitor effects. Both the morphemic and whole-word competitors attracted more fixations than the distractor. More importantly, the 2nd-morphemic competitor effect occurred at a relatively later time window (i.e., 1000-1500 ms) compared with the whole-word competitor effect (i.e., 200-1000 ms). Conclusion: Findings in this study suggest that semantic information of both the 2nd morpheme and the whole word of a compound was activated in spoken word recognition and that the meaning activation of the 2nd morpheme followed the activation of the whole word.