Expression of Prejudice: Are people more likely to express prejudice when they have established their "moral credentials" (i.e., when their past behavior has established that they are nonprejudiced persons)? Present Friday Oct 28

Monin and Miller conducted three experiments and found that yes, people are more likely to express prejudice when they have shown that they are nonprejudiced. A more recent study showed that White adults who know that they are going to write about a controversial racial issue like Affirmative Action choose first to write about their glowing friendships with people of color. Your task is to focus on one of these two designs and specific hypotheses and see if these findings replicate with Hope students who will  complete a questionnaire.

 I will give you two different articles testing the idea that once a person shows his/her moral credentials (shows that he/she is not prejudiced), his/her real prejudiced attitudes come out. One study is an abstract/summary of a research project. The other study is a research paper by Monin and Miller. Your lab group must decide which of the two you want to replicate with Hope students or family/friends.

Does “establishing moral credentials” cause the “release of true sexist attitudes?”

I you go with the Monin and Miller study:   Finding only shown for men (even with 20 men and 20 women as subjects)

Monin, B. & Miller, D. T. (2001). Moral credentials and the expression of prejudice. Journal of Personality and Social Psychology, 81, 33-43.

Monin and Miller conducted three experiments (focus on Study 1) and found that yes, people are more likely to express prejudice when they have shown that they are nonprejudiced. Your task is to see if these findings replicate with Hope students who answer your surveys in 10 minutes or less.

Ways to strengthen this study: First, the moral credential manipulation involved answering/protesting to 5 blatantly sexist statements (experimental condition) or 5 less sexist statements (control condition, see materials Inman gives you). Five statements was not a strong manipulation. I recommend that you add 5 more statements to the experimental (blatant) and control (less blatant) control conditions. Second, subjects start guessing hypotheses, so add a filler survey (see attached). Third, one could challenge the construct validity of the prejudice measure (is this job suited for men or women). It's one item involves construction (a male-dominated profession). I recommend think about the construct validity of the “expressing true sexist attitude measure.” Monin argues that saying a man is better for a male-dominated job reflects sexist attitudes. Do you agree? If not, you can think of other ways to assess prejudice against women (e.g., use Glick & Fiske's benevolent and/or hostile sexism scales--see attached). A prior Social Psych Lab had participants respond to hiring decisions of a traditionally male job (construction), traditionally female job (nursing), and gender neutral job (professors) and reasoned that sexist attitudes should emerge in the first two hiring decisions. I attach sample manipulations and measures for you. If you want the full Monin article, let me know. I can get you the whole copy.

If you go with the Hispanic/Affirmative Action Study: By Eden King (SPSP Abstract that is attached) Run White subjects only

You’ll need to develop a procedure that takes 10 minutes and that makes the participant feel free to give his/her honest opinion. This might be over telephone or by survey. If you opt to create a dependent variables that only has two values (person described positive/negative experience), then you will need many participants. If you assess White people’s feelings about friends/strangers, forewarn them/don’t forwarn, and then have a yes/no dependent measure, you’ll need at minimum 10 to 15 people per cell. 2 x 2 cell is 4 cells. 4 x 15 is 60 participants, 40 minimum. (see the abstract for the design of this study). You don’t have to have the person actually write/talk about Affirmative Action (just have them believe they WILL be doing so). I would recommend making a dependent variable that has a continuum of 1 to 9. Perhaps have them rate the degree to which their written/described relationship with that person is negative to positive. How many Hispanic friends do they have? Rate the depth of friendship, depth of contact they have with Hispanics. Having a dependent variable that is on a continuous scale is desirable. (avoiding a dichotomous measure like “yes/no”)

Things to do/Tasks:

  1. Meet as a group (in class) and identify each person's strengths and tasks.
  2. Discuss the project idea and specify the hypotheses (everyone's input here)
  3. Design the method of the study (manipulation, materials), send materials to Inman

Discuss the way you'll present the info, what you'll say, what you'll do, debriefing, ethics. This complete information is given to Inman on the Pre-Lab Worksheet

  1. Submit your ideas to Inman in writing, (Pre-Lab Worksheet--see Web, then wait to get approval & materials from Inman

Send materials to Inman by email. If you use the existing materials exactly as they are, let her know. She has them on file.

  1. Collect your data
  2. Analyze/Code the data
  3. Enter the data into a computer file (following Inman's instructions)
  4. Learn of the results with Inman and how you'll present them graphically. If you are shaky on stats and methods, I recommend that you make every effort to attend the meeting with Inman. If you miss this meeting, it's YOUR responsibility to get the info from lab mates who attended.
  5. Prepare a 8-minute (10-slide or so) powerpoint/overhead presentation (see website) decide on speaker roles
  6. EVERYONE completes the Lab worksheet about your study (see web)--individual writing here
  7. EVERYONE has some speaking role in the class presentation.