In this paper we aim at estimating the unemployment duration in Botosani County in order tostudy the impact of individual characteristics (gender, age, place of residence, unemploymentbenefit, etc.) on the length of unemployment spells. We use Cox regression model to measure theeffects of gender, age, residential environment, etc. on the hazard rate of leaving unemploymentandKaplan-Meier estimator to compare survival probabilities among different categories ofunemployed persons. The study is carried out on a sample of 200 unemployment spellsregisteredwith the Employment Agency of Botoşani County from January 2012 to December 2015. Theresults reveal that place of residence, unemployment benefit and unemployed category have asignificant impact on unemployment spells.