The effects of two primary reading programs using a programed format (with and without audio-supplement) and a conventional format (the program format deprogramed) in a highly consistent sound-symbol system of reading at three primary grade levels were compared, using a pretest, post-test control group design. The degree of suitability of programed texts in reading instruction for primary children at various intelligence levels, at various age levels, and in different sex groups was also investigated. Treatment 1 followed a conventional approach in which commercial programed materials were deprogramed by stapling immediate feedback columns on the pages of the linear text so the learner could not use immediate feedback information. Treatment 2 used the same commercial series as Treatment 1 but in the prescribed programed way. Treatment 3 used the same series as Treatment 1 and 2 but with audio-reinforcement. An analysis of variance revealed Treatment 3 to be the most effective in grade 1 and Treatment 2 to be the most effective in grade 2; grade 3 findings were not statistically significant. Some statistically significant findings related to chronological age, mental age, and sex are also reported. (Author/CL)