Research Methods in Psychology STATS 241 Lecture 3 DATE Non theory driven research B F Skinner Radical Behaviorism Basic Ideas 1 When you run into something interesting drop everything and study it 2 Apparatuses sometimes breakdown 3 Some ways of doing things are easier than others 4 Theories are not necessary and can get in the way of good research Seek only to describe the functional relationships between amongst variables As Fixed Ratio increases response rate increases A Straight Alley Runway Usually about 6 12 feet long START RUNWAY GOAL Skinner viewed theories as obstacles to good research Goals of Scientific Research Pages 40 49 of Text 1 Description events and their relationships are defined classified cataloged Examples DSM IV TR 2000 diagnostic statistical manual of mental disorders IV ed TR 2000 Pace of Life Study Levine 1990 Levine looking at the speed with which different cultures go through life Looked at clocks observation how fast do people walk how does this compare across cultures Data table does not describe or explain why just describes what is shown 2 Prediction correlation occurs when measures vary together co vary in a Examples GRE scores undergraduate GPA as predictors of success in a graduate school consistent way environment Ambady Rosenthal 1993 As gre scores go up does this show how well they will do in graduate school Ambady and Rosenthal looked at students in college classrooms and how they looked at teachers had those students watch a video of a teacher and had them do an evaluation Other students who actually had that teacher for the whole semester took evaluation and compared the evaluations of teachers Very high correlation between those who did not actually have her and those who had them for the whole semester Apparently it was easy to tell how the teacher was 3 Explanation causal inference requires three conditions co variation of events when one changes the other changes in a consistent way time order relationship one event always precedes the other Antecedent consequent elimination of plausible alternative causes eliminate all threats to internal validity True Experimental Design 4 Application research designed to solve a problem applied research Quasi experimental and Applied Behavior Analysis True Experimental Design true experiment involves the manipulation of one or more factors variables and measurement of the effects of this manipulation on behavior Variable a dimension that can take on different values age handedness time size etc intentionally varying to demonstrate if there are any changes there are constants as well that can be variables in a piece of research i e speed of light Variables In any experiment there are three types of variables a b variable that the E intentionally varies manipulates 1 Experimental Manipulations Independent Variable IV Intentionally varies and manipulates Individual Differences Variable IV cannot draw causal conclusions do not have a variable that you can manipulate instead you select subjects so that they vary in different ways a characteristic of the subjects treated as if it were an 2 Dependent Variable DV the variable that the E measures in order to determine whether the IV has had a systematic effect DV researched monitors to see if the independent variable has an effect 3 Secondary Variable a variable that is not a part of the hypothesis under study 2V as far as you know are not relevant to your hypothesis You can have more than one IV and more than one DV Types of Secondary Variables Confound secondary variable variable that has unintentionally co varied along with the IV causing a threat to internal validity You don t want a confound Extraneous or random secondary variable variable that is not directly related to the hypothesis and which the E does not attempt to control allows to vary at random Controlled secondary variable variable not related to the hypothesis but which E does control either by holding constant equating across groups or by matching Example of confound looking at mirror drawing cannot see hand directly but reflected in mirror 3 groups 1 cup of coffee 3 cups of coffee 6 cups of coffee Each group have a different time Researcher says as the number of coffee cups increase performance gets worse The problem is that the researched ran the experiment at different times of the day Could have been because of amounts of coffee could be because of the timing You do not know which one was an influence on the result of the research Could you fix it Yes but you have to be aware of it Extraneous all variables that your participants will be subject to A variable that you did not attempt to control allows it to vary at random You cannot control e g The amount of sleep if they took medication if they had a rough day if they are tired age Controlled secondary Not directly related to the hypothesis but the researcher does control Independent Groups Designs Random Matched Natural Independent Groups Designs the IV is varied or manipulated between sets of different subjects one set for each level of the IV Three Types of Independent Group Designs 1 Random Groups Design most often 2 Matched Groups Design least often 3 Natural Groups Design least often Random Groups Design 1 2 randomization of equal n per level The subjects are assigned to the levels of the IV randomly This is a two step process Draw a pool of subjects from population using some method of Randomly assign subjects to levels of the IV usually with a constraint Must use a formal randomization process i e a random number table in text dice coin toss Must be truly random Examples of way you should us a formal randomization process Time estimation example Two level IV with feedback and without Class of 50 students go to room 666 at 7 00 PM Time estimation multiple variables How good are people at time estimation Does feedback help one get better at estimating What else could have been a variable Punctuated people If people are always on time they are going to be more likely to be They should have given random people that walked into the room as feedback or no on time feedback Arbitrarily informal randomization decide first 25 that arrive placed in feedback last 25 that arrive placed in no feedback Results better time estimation in feedback group Real life subtle example Learning in Rats What does random assignment buy you Will on average in the long run have an equal representation of all levels of secondary variables in all groups The