PSYC217 Chapter 11 - Complex Research Design IV's with Two+ Levels - Simplest experimental design > IV has only two levels (also called groups or conditions). - A researcher might want three or more levels of the IV for several reasons > frequently interested in comparing more than two groups. " By adding a third condition researchers can have an additional control condition that could potentially provide greater clarity to results. · Levels: operationalization of IV in an experiment (often referred as condition or groups) - Only two levels of the IV may not provide enough information about the relationship between the IV and DV. " Might not be able to see if there is a true relationship between variables > need to examine different levels of the IV to find the effects on the DV. - Variables are sometimes related in a curvilinear fashion -> direction of relationship changes across values of the variable " An experiment with only two conditions cannot detect curvilinear relationships like an inverted-U. · If a curvilinear relationship is predicted, at least 3 levels must be used. - Many such curvilinear relationships exist in psychology. · E.g., relationship between fearful arousal and attitude change · Increasing the amount of fear aroused by a persuasive message increases attitude change to a moderate level of fear, but further increases in fear actually reduce attitude change. Factorial Designs - Simple experiments have one IV (e.g., reward) with two levels (e.g., no reward versus vs. $40). . In addition to adding additional levels of an IV (e.g., $10 and $20 rewards), you can also add additional IVs to an experiment (e.g., task difficulty). - Adding more independent variables to an experiment brings it closer to real-world conditions, in which many influences (IVs) interact with one another to produce some behaviour or effect. - Factorial design: an experiment with more than one independent variable (also called a factor), with each factor having at least two levels (i.e., two conditions). · Design has more than one IV (also known as a factor), and with each independent variable (IV) also having more than one level. · Simplest factorial design has 2 IVs, each with 2 levels. 1 Typically describe a factorial design like this: " Number of levels of first IV x Number of levels of second IV · For two IVs, each with 2 levels would be described as a "2 × 2 factorial design" Example: effect of body type on food consumption. · Participants told study was about movie-watching experience and offered snacks while watching the movies. " All participants were in pairs, but the other "participant" was a confederate. · IV1: amount of snack food the confederate selected (30 candies or 2 candies), · IV2: confederate's body size (also had 2 levels: thin vs. obese) o Researchers had a special suit that increased confederate's natural size from size 0 to size 16, adding approximately 80 pounds to her frame.
· DV was the number of candies the real