PSYC 3100 1st Edition Lecture 6 Outline of Last Lecture I. Snyder, Tanke and Berscheid: Protocol II. Snyder, Tanke, and Berscheid III. Questions IV. Results V. The Case of CamillaOutline of Current Lecture I. Correlational Vs. Experimental Research II. Two Types of Studies A. Correlational B. Experimental III. Correlational Studies IV. Pearson r V. Correlational Studies VI. Casual Possibilities VII. How can one know VIII. An Experiment: What is it IX. Experiment X. Causal Inference XI. Confounding Variables XII. Trait-Confounded Experiment These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.XIII. Situation Confounded Experiment XIV. Confounds Problematical for Two Reasons XV. Situation Confounded Experiment XVI. Situation Confounded Experiment XVII. Situation Confounded Experiment XVIII. Avoiding Confounds XIX. ExamplesCurrent Lecture I. Correlational Vs. Experimental Research - Empirical Research is central to Psychology. II. Two Types of Studies A. Correlational - Correlational research is different in character than Experimental research.B. Experimental - Not every study constitutes an experiment. III. Correlational Studies - Concerned with co-variation-association-prediction. 1. Do People who drink a lot of beer have rounder bellies? 2. Do children who eat good breakfasts make better grades in school? 3. Do passionate people make poorer decisions than dispassionate people? - The extent to which variables relate to one another is considered a correlational study. Correlational studies are concerned with at least two variables and how they relate to each other, or determine a relationship. - Example: Number of cigarettes a person smokes per day. - Variables are outcomes and correlational studies can get very complex. - Co-variation can be (and regularly is) inferred informally. However, informal inferences can be (and regularly are) flawed. We sometimes fail to see correspondences that exist and also sometimes think we see correspondences that do not exist. We make assessments intuitively. We need a formal procedure because as humans, we are not always accurate.- Forces that can mask our perception or lead us to perceive illusory correspondences are expectation and desire (i.e., motivation). Under some conditions we tend to “see” what we expect and want to see.- Fortunately, there are formal statistical procedures that allow us to establish whether correlations (correspondences) actually exist.- The most common procedure is the Pearson Product Moment Coefficient, or the Pearson r. - The Pearson r determines the extent to which two variables are related to each other. (outcome) IV. Pearson r - Pearson r establishes the degree to which two variables (A and B) relate to one another.- Ranges in value from +1.0 to -1.0. Pearson r is a coefficient. +1 is a perfect correlation between two variables. -1 is a perfect correspondence between two variables. However, a -1 or +1 correlation is almost unheard of. It is more common to have a mid-range coefficient.- Absolute value tells you the strength of the relationship. The absolute value isvery important and it is the value of r taking away the sign. The closer to 1 the coefficient is, the stronger the relationship. The closer to 0 the coefficient is, the weaker the relationship. - Sign tells you about the character or nature of the relationship. A positive coefficient occurs when A goes down and B goes down. (go up and down together). A negative coefficient occurs when A goes up and B goes down or vice versa. (opposite directions)V. Correlational Studies - Allow us to predict, but are limited in their ability to allow us to understand. Correlational studies are great at predicting relationships, but limited as to understand them.- To understand, we have to know something about causality.- When considering correlational findings, we have to consider three causal possibilities. VI. Casual Possibilities - Possibility 1: Variable A (e.g., breakfast quality) could lead to Variable B (e.g.,grades) Variable A impacts Variable B.- Possibility 2: Variable B (grades) could lead to Variable A (breakfast quality). Variable B impacts Variable A.- Possibility 3: Variable C (e.g., parental involvement) could affect both Variable A (breakfast quality) and Variable B (grades). Could exist where thereis no causal relationship between A and B. The variable C would be simultaneously impacting A and B. EX: Quality of parenting or available time they have to work with their kids, or the parents may not have the opportunity to help with breakfast or studying. Another example could be self-esteem.- Notably, these possibilities are not mutually exclusive! A and B could both be true. All three possibilities could also be working together. - First establish that a relationship exists. Then consider which possibility exists and use experimental design to determine which possibilities to consider. VII. How can one know? - Scientists as a group seek not only to predict, but also to understand, which means they need a technique that will allow causal inference.- Fortunately, such a technique is available: The Experiment.VIII. An Experiment: What is it? - Investigator manipulates the value of some variable (independent variable).- Investigator measures the value of some other variable (dependent variable).- In order to have an experiment, you must manipulate a variable. An investigator actively manipulates (changes) the value of some variable and then observes the impact of manipulation on some outcome variable. The manipulation is the independent variable. IX. Experiment Normal Breakfast Great Breakfast- Manipulate breakfast quality and its impact on grades. The independent variable is the breakfast quality and the dependent variable is the grade. - What is it important to manipulate? Simply because it allows for control. - Breakfast quality is the only difference between condition A and condition B. X. Causal Inference - If an experiment is properly conducted, then one has potential for making a causal inference.- “Properly conducted” means that the experiment has no confounding variables.- The only thing extinguishing condition is the value of the independent variable. - Avoid confounding variables; extraneous variables. XI. Confounding Variables - Confound: A