A constructive replication is defined as, "a replication which tests the same hypothesis as the original experiment, but also adds new conditions to the original conditions to assess the specific variables that might change the previously observed relationships." For your research proposal, you need to propose a constructive replication of a study related to one of the
variables listed on the handout (e.g. aggression, stereotype threat, etc.).

Your proposed experiment must be a constructive replication in the sense that it takes an existing finding and adds new conditions to the research to produce an interaction. For instance, you may wish to test for conditions under which a previously established effect might be weaker or stronger, or conditions under which it reverses. Or, you may want to rule out an alternative explanation for the previous effect. In every case, the prediction must be derived from the theory under consideration. Also, it needs to be a 2 x 2 design.

For this discussion forum, you should brainstorm about two potential hypotheses for a 2x2 design.

1. State a hypothesis for a possible 2 x 2 design incorporating at least 1 (or more) of the independent variables noted on the research proposal assignment (See Week 3 for the research proposal handout!). Be sure to explain your rationale for making the specific hypothesis. Also, identify the dependent variable, and explain how you are going to operationalize your variables! Remember, a good hypothesis must be falsifiable! Additionally, do not to use participant variables (e.g. gender), and finally, be sure to include an APA­formatted reference of the source(s) you cited/used to develop your hypothesis.

As an example using one variable from the handout, a common variable seen in the stereotype threat literature is related to the idea of "threat" or assessment of ability. For example, in a typical "threat" condition, participants who are members of a stereotyped group are typically told that a difficult test is going to be diagnostic of their ability in a stereotype­related domain. In the "non­threat" condition, participants are told that the test is not diagnostic of anything. This manipulation often impacts performance on the test such that minorities who feel threatened in the "diagnostic" group have impaired performance on the test. That is one variable from the handout that could be paired with another variable (with 2 levels), such that a hypothesis could be made regarding the new combination. An example of a 2x2 design using that variable plus a second participant variable would be to use threat and ethnicity. For example, Steele and Aronson (in 1995) looked at threat vs non­threat for African­Americans vs.

European­Americans, and they did find an interaction (as they hypothesized). Another example would be to take the threat variable (threat vs. no threat), and manipulate difficulty of task (easy test vs. hard test) using only participants who are members of a minority group for which a stereotype related to ability exists. One might make a hypothesis about the relationship between these two variables and predict an interaction. For example, I might predict that minorities who are threatened on the difficult task will show more impairment than those who are threatened on the easy task, whereas I might predict no difference between the threat groups on the easy task. I might explain that in terms of anxiety­­that the difficult task might arouse more anxiety about confirming the stereotype, and thereby interfere with performance more. There are other explanations I could use as well, but that is just one example of a constructive replication using a 2x2 design.

2. State a second hypothesis for a possible 2 x 2 design incorporating at least 1 (or more) of the variables noted on the research proposal assignment (See Week 3 for the handout!). It can involve the same variable from #1 or a different one from the handout. Be sure to explain your rationale for making the specific hypothesis. Also, identify the dependent variable, and explain how you are going to operationalize your variables! Additionally, do not to use participant variables, and finally, be sure to include an APA­formatted reference of the source(s) you cited/used to develop your hypothesis.

Discussion Forum Guidelines (IMPORTANT):

Rules for Discussion Postings: Postings must be made on at least two separate days to allow all students time to read and respond to others (so, in other words, you can't do all 3 postings on the same day). Original postings must be made by 11:59 PM on Saturday. The following points are what is looked for in your original postings to the discussion forum and your replies to others' postings:

Original Response to Posted Discussion Topic :

1. Relation of specific information from journal articles related to your topic. You need to incorporate and explicitly reference at least 2 specific points from one or more scholarly JOURNAL articles related to your topic. Be sure to use APA­formatted citations!

2. Discussion at a critical level, not just recitation of facts. Be sure to justify your hypotheses in a rational and logical way based on previous research findings.

3. Posting is made in time for others to read and respond. This means the original posting should be made by Saturday night in order to give classmates time to read it and respond... ­

Replies to Others' Postings (2 points each):

1. Replies are posted to at least 2 other discussion postings.

2. Provide specific and detailed feedback to at least two classmates on 1 or more proposed hypotheses. Did the hypothesis make sense? Were the operationalizations clearly defined? Were the variables being manipulated properly? Was it a 2x2 design with two manipulated IVs? Is the proposed research ethical? Was the rationale for the hypothesis clearly explained?

3. Postings are made in time for others to read and respond.

4. Be respectful of other students and their ideas!

Be sure to label each hypothesis (e.g. #1, 2). Also, don't forget to incorporate two specific points from one or more scholarly articles using APA­formatted, in­text citations.