M4 – Class 14 Notes (12 December 2024) PDF

ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION Mischel’s famous Marshmallow experiment ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION The marshmallow test is one of the most famous pieces of social-science research: Put a marshmallow in front of a child, tell her that she c...

ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION Mischel’s famous Marshmallow experiment ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION The marshmallow test is one of the most famous pieces of social-science research: Put a marshmallow in front of a child, tell her that she can have a second one if she can go 15 minutes without eating the first one, and then leave the room. Whether she’s patient enough to double her payout is supposedly indicative of a willpower that will pay dividends down the line, at school and eventually at work. Passing the test is, to many, a promising signal of future success. Mischel et al. (1972) Journal of Personality and Social Psychology ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION Mischel et al. (1972) Journal of Personality and Social Psychology ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION https://www.goodcapitalinvestmentgroup.com/learn/would-you-pass-the-marshmallow-test ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION But, could there be another explanation for the long-term outcomes? https://www.goodcapitalinvestmentgroup.com/learn/would-you-pass-the-marshmallow-test ENSURING GOOD SCIENCE CORRELATION-VS-CAUSATION Time Parents’ affluence Ability to delay gratification at age 4 Academic success in high school Watts et al. (2018) Psychological Science ENSURING GOOD SCIENCE CORRELATION VS. CAUSATION https://www.theatlantic.com/family/archive/2018/06/marshmallow-test/561779/ ENSURING GOOD SCIENCE CORRELATION VS. CAUSATION The original results were based on studies that included fewer than 90 children—all enrolled in a preschool on Stanford’s campus. In restaging the experiment, Watts and his colleagues thus adjusted the experimental design in important ways: The researchers used a sample that was much larger—more than 900 children—and also more representative of the general population in terms of race, ethnicity, and parents’ education. The researchers also, when analyzing their test’s results, controlled for certain factors—such as the income of a child’s household—that might explain children’s ability to delay grati cation and their long-term success. https://www.theatlantic.com/family/archive/2018/06/marshmallow-test/561779/ fi ENSURING GOOD SCIENCE CORRELATION VS. CAUSATION To establish whether two variables are causally related, that is, whether a change in the independent variable X results in a change in the dependent variable Y, you must establish: Time order: The cause must have occurred before the effect Co-variation (statistical association): Changes in the value of the independent variable must be accompanied by changes in the value of the dependent variable Rationale: There must be a logical and compelling explanation for why these two variables are related Non-spuriousness: It must be established that the independent variable X, and only X, was the cause of changes in the dependent variable Y; rival explanations must be ruled out. ENSURING GOOD SCIENCE LURKING & CONFOUNDING VARIABLES ENSURING GOOD SCIENCE LURKING VS. CONFOUNDING ▪ Confounding: when two or more explanatory variables are not separated and so it is not clear how much each explanatory variable contributes in prediction of response variable (Confounding variable is included in the study) ENSURING GOOD SCIENCE LURKING VS. CONFOUNDING ▪ Confounding: when two or more explanatory variables are not separated and so it is not clear how much each explanatory variable contributes in prediction of response variable (Confounding variable is included in the study) ▪ A lurking variable is a variable that is not among the explanatory or response variables in a study and yet may influence the interpretation of relationships among those variables (lurking variable is not included in the study) ENSURING GOOD SCIENCE LURKING VS. CONFOUNDING ▪ Two variables are confounded when their effects on a response variable cannot be distinguished from each other. The confounded variables may be either explanatory variables or lurking variables. ▪ Example: Studies have found that religious people live longer than nonreligious people. ▪ Religious people also take better care of themselves and are less likely to smoke or be overweight. ENSURING GOOD SCIENCE LURKING VS. CONFOUNDING https://statisticsbyjim.com/basics/lurking-variable/ ENSURING GOOD SCIENCE THE 3 R’S Repeatability (same team, same experimental setup): The measurement can be obtained with stated precision by the same team using the same measurement procedure, the same measuring system, under the same operating conditions, in the same location on multiple trials. Replicability (different team): Obtaining consistent results across studies aimed at answering the same scientific question, each of which has obtained its own data. The measurement can be obtained with stated precision by a different team using the same measurement procedure, the same measuring system, under the same operating conditions, in the same or a different location on multiple trials. Reproducibility (different team): Obtaining consistent results using the same input data; computational steps, methods, and code; and conditions of analysis. The measurement can be obtained with stated precision by a different team, a different measuring system, in a different location on multiple trials. https://doi.org/10.1116/1.5093621 ENSURING GOOD SCIENCE The reproducibility crisis in science - https://www.youtube.com/watch?v=NGFO0kdbZmk ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS https://diversity.social/unconscious-bias/ ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS BIAS: Cause to feel or show inclination or prejudice for or against someone or something You collect data to make a decision But then, things don’t turn out the way you expect What went wrong? You might be dealing with some type of bias which distorts how you perceive the world These biases are universal and are a part of how everyone processes information ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS Common forms of bias when using data 1. Confirmation Bias What the human being is best at doing is interpreting all new information so that their prior conclusions remain intact. — Warren Buffett Confirmation Bias: The tendency to believe that you are right and to disregard things that conflict with your ideas (Kahneman, 2011) ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS Image source: https://thinkingispower.com/guide-to-the-most-common-cognitive-biases-and-heuristics/#h-representativeness-heuristic ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS https://www.youtube.com/watch?v=Kho5KvPBDSw ENSURING GOOD SCIENCE NEED FOR THE SCIENTIFIC METHOD: THE ISSUE OF BIAS Common forms of bias when using data 2. Survivorship Bias Survivorship bias or survival bias is the logical error of concentrating on entities that passed a selection process while overlooking those that did not. This can lead to incorrect conclusions because of incomplete data. Abraham Wald This hypothetical pattern of damage of aircraft returning from airstrike attacks in World War II