Faced with the current trend of rising food insecurity and chronic diseases in urban areas, the adoption by farmers of plants with nutritional and medicinal properties in urban agriculture becomes one of the solutions to this challenge. Currently, chia (Salvia hispanica L.) is one of the crops attracting attention because of its nutritional, medicinal and cosmetic properties. Thus, this study pursues a double objective, namely to study the perceptions of farmers in City of Butembo on this crop and to evaluate the effectiveness of the application of logistic regression. Binary and multinomial logistic regression models were constructed based on data collected from a simple random sample of 120 farmers. This study shows that for a farmer, the knowledge of the chia plant, the consumption of the seeds, the knowledge of the nutritional and medicinal virtues of its seeds, the opinion to undertake this culture, the perception on the evolution of the demand at the market level and the adoption of this crop in urban agriculture vary according to socio-demographic characteristics. The study also shows that logistic regression provides better overall precision, acceptable error rates and moderate Cohen's Kappa coefficients.