BackgroundDespite evidence demonstrating the effectiveness of the COVID-19 vaccine, vaccine hesitancy has emerged as a major challenge for vaccine uptake. The objective of this study was to classify latent typologies of vaccine hesitant adults in Ethiopia and identify predictors of the latent classes.MethodsWe employed a cross-sectional household survey among 1,112 individuals aged 18 and above who were partially vaccinated (one dose) or not vaccinated at the time of the survey. Data was collected in August 2022. We collected information on participant socio-demographics, COVID-19 knowledge, prevention practices, disease history, and vaccine hesitancy. Latent class analysis was used to classify individuals into categories of vaccine hesitancy. We conducted multinomial logistic regression to test the associations between latent typologies and different demographic and COVID-19 related characteristics of study participants.ResultsUsing latent class analysis we found a four-class solution for vaccine hesitancy typologies. The identified classes were strong vaccine acceptors (30%); vaccine acceptors with some concerns (7%); vaccine sceptics (13%); and vaccine rejectors (50%). In adjusted models with vaccine sceptics as the referent group, those with high COVID-19 vaccine knowledge were significantly more likely to belong to the strong vaccine acceptors class compared to those with low vaccine knowledge (adj. RRR: 17.36, 95% CI: 10.94–27.55). Better COVID-19 prevention practices were also significantly associated with belonging to the vaccine acceptors with some concerns class than the vaccine sceptics class (adj. RRR: 2.13, 95% CI: 1.09–4.16). Those who had one dose of the COVID-19 vaccine were significantly more likely to belong in the vaccine acceptors class than the vaccine sceptics class compared to those who had no dose (adj. RRR: 6.82, 95% CI: 3.06–15.21).ConclusionsHalf of the study participants were in the vaccine rejectors class. Individuals in the vaccine sceptics and rejector classes evidenced lower vaccine knowledge and worse COVD-19 prevention practices and were less likely to have been partially vaccinated. Future program interventions should focus on improving knowledge around the vaccine, decrease rumors and misconceptions, and target individuals who may be more amenable to changing their vaccination attitudes or behaviors like vaccine sceptics or acceptors with some concerns.