Correlated T-Test Homework^ã
 

    Please read this article, which you can find in Blackboard under Course Documents, Articles. I have simulated data based on a model derived from the statistics reported in the article. You will being using these data in your correlated t test homework assignment. Here is a brief description of the procedure you are to assume was used to collect the data. An electrode is attached to each participant to record e's electrodermal activity (EDA), a measure of emotionality (skin conductance, sweating). While relaxing in a reclining chair, each participant is asked to imagine something emotionally neutral ("a time you were walking to class, feeling neither good nor bad, just neutral"). EDA is measured during this time. Then each participant is asked to imagine sexual infidelity on the part of e's lover ("Imagine you find out that your partner is having sexual intercourse with this other person,"). EDA is measured during this time too. Finally, each participant is asked to imagine emotional infidelity ("Imagine that your partner is falling in love and forming an emotional attachment to that person."). EDA is measured at this time too. To control for order effects, we have half of the participants imagine sexual infidelity first and emotional infidelity second, and for the other half of the participants we have them imagine emotional infidelity first and sexual infidelity second. For each participant we compute two scores: EDA_SEX is the amount by which the participant's EDA changed, relative to the neutral condition, when imagining sexual infidelity, and EDA_EMOT, the amount by which the participant's EDA changed when imagining emotional infidelity.

    Positive scores indicate that the participant was more aroused (physiologically, presumably due to being upset) during the infidelity imagining than during the neutral imagining. Negative scores indicate that the participant was more relaxed while imagining infidelity than while imagining something neutral. Such negative scores might result if the participant was still nervous about the experiment while the baseline (neutral condition) measurements were being taken, but calmed down later in the experiment (at which time the infidelity imagining took place).

    I have provided you with simulated data for both men and women. These data are deposited in my SimData directory on the core server. You can jump directly to your data file by clicking below on the link with your name. It will have a name of the form 'EDAnn,' where 'nn' is the identification number I use to identify your data files. When you look at the data file you will find that each participant has one line of data. The first score for each participant is the code for the variable GENDER, '0' for female, '1' for male (you should use SPSS Value Labels to assign the verbal labels to the numeric codes). The second score for each participant is EDA_SEX, and the third is EDA_EMOT.

    For each of the two genders (men and women - you will need to split the file by gender) use a correlated t test to test the null hypothesis that the change in EDA caused by imagining sexual infidelity is the same as that caused by imagining emotional infidelity. Use a .05 criterion for statistical significance. For each test write an APA-style summary statement, including the computed value of t, the degrees of freedom, Hedges g, the exact p value, and a 95% confidence interval for the unstandardized difference in means. The statement should identify who the participants are, what the independent variable is, what the dependent variable is, whether the effect is statistically significant or not, and what the direction of the effect is -- for example (using different variables), "Male participants' belligerence scores were significantly greater when they were watching a rugby game (M = 25.92, SD = 44.54) than when they were listening to classical music (M = 6.12, SD = 47.76), g = 0.43, t(24) =2.17, p = .040.  A 95% confidence interval for the mean difference runs from 0.98 to 38.62."

    Deliver your report to me in a Word document named "Yourlastname_Corr_T.doc."   Please do include the underscore character after your last name in the file name.  At the top of the document, following your name, type your two summary statements, one for the women and one for the men.  After that (at the bottom of the Word document) paste in your SPSS output.   Do not send an SPSS spo file.

    At the bottom of the Word document paste a screen shot of the G-EqN-Run.sav file after you have run the associated syntax file to compute g.  Here is how to do this:

• Enter into G-EqN-Run.sav the means and standard deviations for the women.
• On the next row enter in the means and standard deviations for the men.
• Run the syntax file.
• Resize the window to about the size shown below.
• Hold down the Alt key while you hit the Print Screen Key.
• Place the cursor at the bottom of the Word document.
• Hold down the Ctrl key while you hit the V key.
• Zap, the screen shot is pasted into the Word document, like this:

    The subject of your email should be "PSYC 2101:  Correlated t Assignment."

    This assignment is due at 5 PM on Friday the 6^th of June, 2008.

    If you are using the Internet Explorer, to download the file you should point at your name and then RIGHT CLICK on the mouse. You will get a drop down menu. From that menu select "Save Target As" and then point at the drive/folder where you wish to store the data file.

• Use your browser's BACK key to return to the page you previously visited or use one of the following links.

•  Visit Karl's SPSS Lessons Page -- learn or relearn how to do t tests with SPSS.

•  Visit Karl's Index Page -- search for other information on Karl's pages

•  Visit the PSYC 2101 base page

•  Related research by David DeSteno -- critical of Buss' conclusions
  

Contact Information for the Webmaster,
Dr. Karl L. Wuensch


This page most recently revised on 4. June 2008

^ã Copyright 2008, Karl L. Wuensch - All rights reserved.