The current research sought to develop the Implicit Relational Assessment Procedure (IRAP) as a measure of bullying attitudes amongst Secondary School and University Students in South East Ireland. The research assessed whether IRAP performance differed between University and Secondary School Students; and investigated the impact of picture versus word stimuli on IRAP performance. It also examined whether an educational intervention video affected participant responding on implicit measures by presenting the IRAP at pre and post-intervention. Explicit measures were presented at pre-intervention only and compared across studies. Implicit measures were presented at pre and post-intervention and compared across groups, gender, and IRAP stimuli (words versus pictures). In Study 1, 30 University Students and 30 Secondary School Students were exposed to (i) a word-based IRAP designed to assess attitudes towards toxic (e.g. Just go die/Rot in hell) and innocuous phrases (Go on ya fool/Don’t be daft) pertaining to bullying; (ii) explicit measures including the Bullying Prevalence Questionnaire (BPQ), the Revised Pro-Victim Scale (RPV-S), the Bullying Attitudes Questionnaire Modified (BAQ-MM) and the Cyberbullying Survey (CS) and (iii) an educational intervention video about the negative and lasting effects of bullying. IRAP trial-type analysis for Study 1 revealed statistically significant effects on the Toxic- Abusive and Innocuous-Harmless trial-types. Results revealed no statistically significant differences between data for groups, gender, or between pre and post-intervention responses on the IRAP. Using Pearson’s product-moment correlation coefficient, statistically significant correlations were found between the Pro-Social subscale of the BPQ and Toxic-Harmless and Innocuous-Abusive IRAP trial types. In Study 2, 30 University Students were exposed to a picture-based IRAP with images pertaining to cyberbullying and the same intervention and explicit measures as in Study 1. Again, participants were exposed to the explicit measures at pre-intervention, and to the IRAP at pre and post-intervention. Trial-type analysis for Study 2 revealed statistically significant effects on the Toxic-Abusive and Innocuous-Harmless trialtypes. Results revealed no statistically significant differences between participants’ pre and post-intervention scores on the IRAP or explicit measures; and no correlations between implicit and explicit measures. Further analysis using a 2x2x4 mixed repeated measures ANOVA found no statistically significant differences between University Students’ responses on a word-based IRAP in Study 1 versus a picture-based IRAP in Study 2. Overall, participant responding on the IRAP showed a statistically significant effect for the Toxic- Abusive and Innocuous-Harmless trial-types. Findings are discussed with reference to the research literature.