Argues that only positive results in a reading-level (RL) match and negative results in a chronological-age (CA) match are interpretable; negative results in a CA match and positive results in a RL match are not. Argues that using the RL control is only a first step in research into reading disability. (RS)

Discusses conflicting interpretations of results from contrasts of poor and normal readers matched either for chronological age (CA) or reading level (RL). Argues that the outcomes from RL-matched designs or from RL- and CA-matched designs in combination should be interpreted within a broad theoretical and empirical context. (RS)