In order to conduct a power analysis, you have to decide on two things. First, you must develop
your specific hypotheses. Second, you should know how you are going to analyze your data.
After you have completed these two tasks, you need to consult a power table such as the
one provided below. Unless you have a good rationale to do otherwise, then you should use an
alpha level of .05 and a medium effect size.

The reference for this power table is: Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.
Most of you should be able to use this table for your proposals. If you want additional information, then
consult:  Wahlsten, D. (1991). Sample size to detect a planned contrast and a one degree-of-freedom
interaction effect. Psychological Bulletin, 110, 587-595.

The following is the first part of a method section for a proposal. If your design is somewhat
complicated or you believe the reader will have a difficult time figuring it out, then you should
include a "Summary of Design" paragraph.
 

Culture     10 Method               Summary of Design                            The experimental design can be characterized as a 2 (culture) x 5 (feedback condition)                  between participants design. Participants identified as individualists or collectivists will be                  randomly assigned to one of five conditions where feedback information will be manipulated.                  The five conditions are: Opinion-same, opinion-different, ability-lower, ability-same, and                  ability-higher.                Participants                           Power analysis. Cohen (1992) has suggested that experiments should strive to obtain a                  power of .80. Using power tables from Wahlstein (1991), a sample size of 30 per condition will                   yield a range of power over .90 with 10 groups. This effect size estimation is based on                   planned contrasts and will yield a small to moderate effect size.                              Participants will be approximately 300 undergraduates from the University of Kentucky.                 Approximately 60% of the sample will be female and approximately 50% of the sample will be                 identified as collectivists. All participants will receive course credit for their participation and                 will be treated in accordance with the ethical standards of the American Psychological                  Association (American Psychological Association, 1992).