Experimental and Quasiexperimental Method September 16, 2008
Learning Objectives • Review APA reference list • Gain some experience with experimental/quasi-experimental designs. • Learn how to use excel to analyze experimental data. • Continue with introduction to writing a scientific paper.
References Brown, W. M. , & Moore, C. (2002) Fluctuating asymmetry and romantic jealousy. Evolution and Human Behavior, 24, 113117. Buss, D. M. & Greiling, H. (1999). Adaptive individual differences. Journal of Personality, 67(2), 209 -243. Buss, D. M. (1991). Evolutionary personality psychology. Annual Review of Psychology, 42, 459 -491. Buss, D. M. (1997). Evolutionary foundations of personality. In Hogan, R. , (Ed. ), Handbook of personality psychology (pp. 317 -344). London: Academic Press. Buss, D. M. (2000). The dangerous passion: Why jealousy is as necessary as love and sex. New York: Free Press.
First Data Collection • This time you will be the participants. • We are going to run a quasi-experiment • For now you do not need to know what are the conditions of the quasi-experiment.
Instructions • Go to http: //www. thewritingpot. com/stroop/
Instructions • Go to http: //www. thewritingpot. com/stroop/ • There are going to be a series of colored words.
Instructions • Go to http: //www. thewritingpot. com/stroop/ • There are going to be a series of colored words. • Your task is to hit the key which corresponds to the color of the word not what it says.
Instructions • Go to http: //www. thewritingpot. com/stroop/ • There are going to be a series of colored words. • Your task is to hit the key which corresponds to the color of the word not what it says. • If you are ready go ahead and click on try experiment and then hit space bar.
Stroop Effect Continued • When you are done write down click on view results. • Write down the average time for words which were different for the color and the amount of time it took for words which matched the color. • Now repeat this entire process two more times.
Microsoft Excel • Open a new database. • Write the following variable names: congruent 1, congruent 2, congruent 3, averagecongruent, noncongruent 1, noncongruent 2, noncongruent 3, average congruent. • Enter your scores from the experiment.
Calculating a mean in Excel • In the cell under average congruent type =average( • Type a 2: c 2) or highlight the cells which contain your congruent data. – In both cases you should have something like the following : =average(a 2: c 2) • Do the same for non congruent variables
Were you strooped?
Were you strooped? • First you might want to plot mean stroop effects. – To do this highlight the Average stroop scores for congruent and non-congruent • In this case the ctrl button is helpful. • First highlight one press control and highlight the other. – Now click on insert graph – Pick the bar graph – Hit enter • If your non-congruent stroop average score was higher than the average congruent score you were strooped
Here are my scores 1400 1200 1000 800 Series 1 600 400 200 0 Congruent Stroop Noncongruent Stroop
Did your performance improve? • How would we know? • We could plot the learning curve for each condition. • First highlight your congruent 1, congruent 2, and congruent 3 scores. • Now make a similar graph as before.
My Improvement 940 935 930 925 Series 1 920 915 910 1 2 3
Was the Class Strooped? • First we have to enter data. • We already have the variable names in our database. • I’m going to call each of you one at a time. – I want everyone to enter their classmate’s scores. – I also want them to record each participant’s sex
The Dataset Congruent 1 Congruent 2 Congruent 3 Average. Co ngruent Noncongru ent 1 Noncongre nt 2 Nongongre nt 3 Average Congruent Sex
Let’s Do a T-Test • On toolbar on top click “Tools” dropdown box will appear – Choose “Data Analysis” then. . . • Click “t-test: 2 samples assuming equal variance. ” • Highlight data: You should see a box with room to enter numbers for “Variable 1 range” and “Variable 2 range. ” • Hit Enter
Are males more easily strooped? : Our quasi-experiment • concept is the same – comparing means • analysis is a bit more complicated • What do we need to do?
Are males more easily strooped? • concept is the same – comparing means • analysis is a bit more complicated • What do we need to do? – Hint- What are our independent variables
Data Manipulation • How do we calculate this difference score? – First type “=(“ in cell J 2 • This creates a new variable. • If you want you could name it stroop score. – Highlight the noncongruent average • (cell H 2 if you followed my example) – Follow that with a – sign – Then Highlight the congruent average • (cell d 2) – The final product should read =(H 2 -D 2)
Data Manipulation – Now highlight cell J 2 • There will be a little box on the border of that cell • Click and drag it down to the end of the data set. • This will give you the stroop score for each person. – Positive numbers mean they were strooped negative or scores of zero means they were not.
Now the Dependant Variable • We have to sort the data set • We need all the males in one section and all the female data on the other. • Highlight all the data by clicking on the top left hand corner of the spreadsheet. • Click on data • Click on sort • Sort by the column for the variable sex.
Let’s Do a T-Test • On toolbar on top click “Tools” dropdown box will appear – Choose “Data Analysis” then. . . • Click “t-test: 2 samples assuming equal variance. ” • Highlight data: You should see a box with room to enter numbers for “Variable 1 range” and “Variable 2 range. ” – This type highlight the difference scores for one of the sexes for “variable 1 range” and highlight the difference scores for the other sex “variable 2 range” • Hit Enter
Assignment • I want you to write up a methods and results section for the quasi-experiment we just conducted. • See pages 367 and 368 in the Leary book.