An error correction encoding method is provided for encoding source digital data, having the form of a frame, wherein the data can be classified into N classes, where N is an integer at least equal to 2.The encoding method includes: a first step of recursive systematic convolutional encoding of the data of the class 1; an implementation of the following steps, for each n ranging from 1 to M, where M is a positive integer equal to or lower than N−1: nth mixing (108n+1) of a set formed by data of the class n+1 (102n+1), the systematic data and the parity data from a preceding encoding step; and (n+1)th recursive systematic convolutional encoding (110n+1) of data formed by the result of nth mixing. A decoding method is provided for decoding encoded data with the encoding method, using an associated encoding device and a decoding device.