PSY 3377 Lecture Notes - Lecture 5: Effect Size, Reproducibility, Speech Segmentation

• School: University of Ottawa
• Department: Psychology
• Course: PSY 3377
• Professor: Sylvain Gagnon

Document Summary

Reasons for not reproducing data (replication) that were previously produced in another study: bias in samples, not enough power (participants), etc. P-value: the chance of achieving this is . 05 (5%) Effect size: when significant, you don"t need a large sample to find a high effect size: if small effect size, increase sample and effect size will increase (usually) We start processing the external stream of information (interpret, perceive, etc. ) if they reach a certain threshold or if we detect a change in the threshold. Attention is important = when you perceive and detect something, and attention is deployed toward that source of sensory info, the information will be processed better. To detect the minimal threshold, you must attend to the info because if we don"t attend to. We cannot define perception by only the sensations; attention must also be taken into it, we won"t notice it account. No simple association between perception and attention.