Recently, several studies have reported that the survival benefit of wedge resection might not be inferior to that of lobectomy in early-stage NSCLC patients, but there is no unified definition of the details or cutoff value. Patients with early-stage NSCLC with a tumour size ≤ 2.0 cm were chosen from the SEER database. The influence of confounding factors was minimized by 1:1 propensity score matching (PSM). Kaplan‒Meier curves and Cox proportional hazards models were used to evaluate the overall survival (OS) and lung cancer-specific survival (LCSS) of patients undergoing lobectomy and wedge resection. A total of 3891 patients with early-stage NSCLC with tumour size ≤ 2.0 cm were enrolled, of whom 2839 underwent lobectomy and 1052 underwent wedge resection. Both before and after PSM, lobectomy significantly improved OS and LCSS compared with wedge resection in the unstratified study population. In the tumour size ≤ 1 cm group, lobectomy had better OS and LCSS than wedge resection (P < 0.05) before PSM; after PSM, there was no significant difference in OS (P = 0.16) and LCSS (P = 0.17). In Grade I patients, before PSM, lobectomy was superior to wedge resection in LCSS (P = 0.038), while there was no significant difference in OS (P = 0.16); after PSM, there were no significant differences in either OS (P = 0.78) or LCSS (P = 0.11). For early-stage NSCLC patients with a tumour size ≤ 1 cm or with a tumour size ≤ 2 cm and with Grade I, there was no significant difference in survival between wedge resection and lobectomy.