The rationale behind this study was that the low norm may not have been low enough. Therefore, we presented an even lower norm (.14 cents) and ran participants in a 2x2 (norm no-norm vs. speed delay) ANOVA.
Under the full data model (no adjustment for passing the comprehension check), we found that participants adjusted their contributions to match the low norm; participants who saw a low norm donated less (24.463, SD = 16.32) than if they hadn't been told a norm (31.225, SD = 16.379; F(1, 290)=12.607, *p*<.001).
We also replicated the speed effect with participants deciding quickly donating more (29.798, SD = 16.351) than if they waited to decide (25.889, SD = 16.335; F(1,290)=4.216, *p*<.041).
However, once again, the sought after interaction was nowhere near statistical significance (*p* > .957).