Childhood obesity has followed, during the last two decades, an ascending trend. Insulin resistance (IR) is central to the pathophysiology of obesity. So far, several methods of assessing IR have been proposed. We aimed to evaluate critically some of the simplest methods used to assess IR in the pediatric population. We studied retrospectively the records of children evaluated for obesity in the “Louis Ţurcanu” Emergency Hospital for Children Timisoara, over a period of 10 years. The study population consisted of 342 children. Anthropometric and metabolic variables were analyzed, and the following indices of IR were assessed: impaired glucose tolerance (IGT), Homeostatic Model of Assessment-IR (HOMA-IR), Homeostatic Model of Assessment- b (HOMAβ), Quantitative Insulin Sensitivity Check Index (QUICKI) as well as the TG/HDLc ratio. Data was expressed as frequencies, means ± standard deviations or median ± interquartile interval for or a 95% confidence interval. The t-test for independent groups or the Mann-Whitney test to assess differences of IR indices across weight, gender and pubertal categories. HOMA-IR diagnosed the most children with IR, at the opposite pole we found QUICKI. IGT was a rare finding. It is necessary to reconsider how we assess the carbohydrate metabolism in children. Of the methods we evaluated, HOMA-IR is the optimal method for assessing IR children.