PSY 311 1st Edition Lecture 9Outline of Last Lecture I. Confidence intervals Outline of Current Lecture I. One way design II. Assignment of participants III. Posttest designIV. Pretest posttest design V. Factorial design VI. Nomenclature VII. Assignment of participants VIII. Main effects IX. Combined effects a. Additivityb. Non-additivity Current Lecture : Experimental Design I. One way design • One independent variable– But can have two or more levelsE.g., Does the amount of coffee consumed affect braking speed?– Design with 2 levels: 1 vs. 2 cups– Design with 5 levels: 1 vs. 2 vs. 3 vs. 4 vs. 5 cupsII. Assignment of participants - Randomized (between-subjects)- Matched-Subjects- Repeated Measures (within-subjects)III. Posttest design - Data gathered after exposure to the independent variableo E.g., measure braking time AFTER drinking the coffeeo Note: Even if there was a control condition (i.e., where 0 cups of coffee are consumed), it is still considered a posttest design because of the comparison to the experimental conditionsIV. Pretest posttest design • Also measure dependent variable prior to exposure of independent variable– E.g., measure braking time before drinking coffee, and then again after drinking coffee• Advantages:– Make sure participants are equated on the dependent variable– Can measure the amount of change which reduces the need of a control condition where no cups of coffee were consumed– More power!!!• DisadvantagesPretest sensitization• If you think it’s going to be a problem, don’t use pretest – posttest• If you don’t think it’s going to be a problem but are not sure, you can run half the participants in post-test only and compare results (but can be expensive and a waste of time)V. Factorial Design • Two or more independent variables tested simultaneously– E.g., Does the amount of coffee consumed and the amount of sleep affect braking speed?• In this context, the independent variable is called a FactorAmount of Coffee (Factor A)1 Cup2 CupsAmount of Sleep (Factor B)3 HoursA1B1 A2B17 HoursA1B2 A2B2VI. Nomenclature - a X b design– a is the number of levels in factor A– b is the number of levels in factor B- And so on…..– a X b X c– a X b X c X d– …..Thus , • 2 X 2 design has two levels of factor A and two levels of factor B• 3 X 2 X 4 design has three levels of factor A, two levels of factor B, and 4 levels of factor CVII. Assignment of participants • Randomized group factorial design– Each participant is assigned randomly to one of the experimental conditions• Repeated measures factorial design– All participants in every condition• Mixed factorial designs– A mixture of between and within subjects– E.g., This study takes place over two days. On both days, half of the participants drink 1 cup of coffee while the other half drink 2 cups of coffee. However, on one night the participants are limited to 3 hours of sleep and on the other night they are limited to 7 hours of sleepVIII. Main effects • Main effects refer to the effect of a single factor while IGNORING the other factorsE.g., • Is there an effect of the amount of coffee on braking times?– To answer this you ignore the role of sleep by combining the 3 hour group with the 7 hour group• Is there an effect of the amount of sleep on braking times?– To answer this you ignore the role of coffee by combining the 1 cup group with the 2 cup group• Effects can be quantifiedE.g., – The additional cup of coffee might, on average, slow down braking time by 300 ms– Four hours less sleep might, on average, slow down braking time by 250 msIX. Combined effects - What happens when you consider all factors simultaneously (which is the point of the factorial design)?a. Additivity - If the combined effect is the sum of the individual effect, we call this additive– E.g., – The additional cup of coffee might, on average, slow down braking time by300 ms– Four hours less sleep might, on average, slow down braking time by 250 ms– Then participants who have the additional cup of coffee AND experience four hours less sleep might, on average, slow down braking time by 300 + 250 = 550 msb. Non-additivity - On the other hand, participants who have the additional cup of coffee AND experience four hours less sleep might, on average, slow down braking time by:– 1000 ms– Or 20 ms– Or 610 ms– Or anything besides 550- We call this non-additivity and it occurs when the combined effect is not equal to the sum of the main effects– There could be enhancement – There could be suppression- An Interaction is when the combination of factor leads to non-additivity.- This is the same as saying that the effect of one factor is different depending on the level of the other factor– E.g., the effect of coffee is different depending on whether you slept 3 hours of 7