Background: Immunization is the process by which a person is made immune or resistant to an infectious disease, typically by the administration of a vaccine. Childhood immunization is one of the most significant health indicators of healthy children. Ethiopia has extremely low immunization coverage in which nearly 1 million children are unvaccinated, and about 16% of children mortality has been attributed to vaccine-preventable diseases in the country. Thus, the aim of this study was to investigate the spatio-temporal distribution and associated factors of children’s immunization among the aged 12–23 months. Methodology: The data were drawn from the 2000-2016 Ethiopia Demographic and Health Survey (EDHS). A two-stage stratified sampling technique was used and a total of 6,767 children were identified and included in this analysis. The study was evaluated on different statistical models such as proportional odds (POM), Partial Proportional odds (PPOM), Generalized Ordered logit (GOM), Adjacent-categories logit model and spatial ordinal logistic regression. Models among the comparable models select the best model by using AIC.Results: The overall full immunization coverage among children aged 12-23 months in 2000 was (14.6%), in 2005 was (22.7%) in 2011 was (25%) and in 2016 was (39.4%). The spatial distribution of children immunization status was none-random in the country within each EDHS year. The PPOM has smallest AIC has values 10383.830 from this PPOM was the best model for this data. And the Score Test for the Proportional Odds Assumption has Chi-square statistic 186.8130 with 43 df and p-value of 0.0001 it is statistically significant it shows that each cumulative logit is not parallel, and the proportional odds assumptions for the full-model is not appropriate. Conclusion: The partial proportional odds model shows that; region, Residence, Mother Education, Sex of household head, Religion, Wealth index, work status of mothers, Place of delivery, household Visited by health worker last 12 month, Antenatal care during pregnancy, EDHS year were significant effect for children’s immunization status. And in the spatial ordinal logistic regression model, the auto-covariance variable was significant and negative, which indicates that zones with a lower level of immunization status were surrounded by zones with higher levels of immunization status. Recommendation: Concerned body facilitated in high risk (hot spot) areas by giving special attention may be by giving education about immunization, building health services and employ health workers. Key words: immunization status, spatial ordinal logistic regression, partial proportional odds model, Ethiopia, Hot Spot.