Sociology 2206A/B Chapter Notes - Chapter 9: Selection Bias, Latin Square, Interrupted Time Series
Document Summary
When you modify something and compare it to the outcome without modification. Strongest for casual relationships, able to demonstrate causation. Dividing subjects into groups at the beginning of experimental research using random process, so the experimenter can treat the groups as equivalent. Science of the sophomore: refer to potentially limited external validity of students based on undergraduate samples (usually expira method) Starts w 2 groups, then splits them into 2 or more groups (dice, coin, count) Deception: when experimenter lies to subjects about the true nature of an expira or creates a false impression through their actions of the setting. Debrief: researcher gives true explanation of the expira to subjects after using deception. Experimental design: arranging parts of an expira and putting them together. Classical experimental design: has random assignment, control group, expira group and pre-tests and post tests for each group.