Behavioral interventions to reduce the level of disruptive behaviors have been based on the problems they caused to acquire adaptive repertoires. Reducing techniques mainly used were extinction, differential reinforcement, time out, response cost, and so on. In this work, an alternative way of a children freetime group activities. The experiment was implemented with ten children between 8-11 years old that showed a higth level of marginal indices (long term families unemployment, criminal and/or addictive behaviors among some members of their families, academic problems…) Five discruptive behaviors were elected (to stand up, to insult, to talk them, to shout and to unjustified complaint in player, to debate about some topic and to play). During baseline children showed a high and stabilized frequency of disruptive behaviors. Then, intervention was implemented. The alternative technique used was a say-do correspondence training (saying no- not doing- saying that he or she haven´t do). All the ten children answered that were not going to behave disruptively when they were asked, and some children were rewarded for fulfilling  it. The high frequency of disruptive behaviors showed by the children in the baseline decreased sharply when the correspondence training was implemented. These results were maintained even when the tretment was removed (reducing cosequences and fading questions and answers).            Resuts were analyzed by a within subjects analysis of variance, that confirmed the significant differences existing between baseline and intervention phases, and the absence of signiticant differences between intervention and maintainig phases. This paper added empirical evidence on the validity of correspondence training how a technique to use setting with the intervention are some important aspects to stand out. Additionally, we would say that these children have acquired some self-control behaviors that is an important repertoire to children with a high risk to develop criminal and/or addictive behaviors.Las conductas disruptivas generan dificultades en la adquisición de repertorios adaptativos y en la integración social de las personas que las emiten con frecuencia, lo que ha justificado el diseño de técnicas de intervención dirigidas a disminuirlas: extinción, reforzamiento diferencial, coste de respuesta, tiempo fuera, etc. En este trabajo se presenta una forma alternativa de intervención para la reducción de las numerosas conductas disruptivas que un grupo de 6 niñas y4 niños de una barriada marginal realizaban durante sus actividades de tiempo libre. Se aplicó un procedimiento denominado entrenamiento en correspondencia "Decir no -no Hacer- Decir que no se ha hecho" en el que los niños dicen que no van a comportarse disruptivamente, son reforzados por cumplirlo y finalmente se les formulan preguntas  sobre lo dicho, lo hecho y la relación entre ambas cosas. La intervención redujo con bastante rapidez la elevada frecuencia de conductas disruptivas que mostraban en la línea-base. Los logros de la intervención se mantuvieron tras la retirada de los elementos de la misma. Se discuten las ventajas del uso de las técnicas de correspondencia decir-hacer en los contextos educativos.