Why is the Nisbett and Wilson replicated Halo Effect experiment valid assuming that it is statistically significant? Discuss if the results would be found using different samples, designs, procedures, or measures. In particular, discuss the strengths and weaknesses of an alternative design in a future replication. What future research would be warranted by the results in exact or conceptual replications? What 'real world' implications may the results have, such as for interventions, policies?