The three reorts evaluated this hypothesis and came to dramatically different conclusions. 'Some Observations on the Link Between Learning Disabilities and Juvenile Delinquency' by Zimmerman, et al. maintained that there were no significant differences between learning disabled (LD) and non learning-disabled (NLD) teenagers in frequency of ever having engaged in various types of delinquent activities. Broder, et al. concluded in 'Further Observations on the Link Between Learning Disabilities and Juvenile Delinquency' that LD youths had significantly lower mean frequency of general delinquent behavior than their NLD counterparts once differences in age, social status, and other characteristics had been controlled statistically. In 'A Causal Analysis of the Relationship Between Learning Disabilities and Juvenile Delinquency.' Dunivant reached the conclusion that LD youths were significantly more likely to have engaged in delinquent behavior than NLD teenagers. This paper contends that the last conclusion is the correct one. Additional analyses of the data used by Zimmerman, et al. clearly demonstrate that LD boys engaged in significantly more delinquent behavior than their NLD peers. The only reason that the results of Zimmerman, et al. differed from those of Dunivant was that Zimmerman, et al. performed their analyses separately for NLD and LD samples, whereas Dunivant based his analysis on the combined samples. In the second report, Border, et al. made an error in interpreting the output of the regression analyses. When correctly interpreted, the results of Broder, et al. show that LD boys had significantly higher general self-reported delinquency scores than their NLD counterparts. Five references and statistical tables are provided.