Posted by Anonymous on Tuesday, March 26, 2013 at 10:04pm.
A researcher is interested in examining whether knowledge of a previous criminal record will affect the juryfs decision to acquit or convict the defendant. The researcher shows 15 participants a 4hour trial in which a woman is accused of passing bad checks. Participants were divided into three groups based on what they were told about the defendantfs criminal past: criminal record group (defendant had a criminal past), clean record (no criminal past), or no information group (no information regarding criminal past). Participantfs then rated the defendant from 1 (completely innocent) to 10 (completely guilty). The data are presented below. Perform the appropriate statistical test (using statistical hypothesis testing) and any necessary post hoc tests to determine whether prior criminal history affects juries.
n=5 ‡”X2 = 589 T1= 40 T2= 20 T3=5

stats  MathGuru, Wednesday, March 27, 2013 at 8:11pm
Here are a few hints:
Check out a oneway ANOVA for this kind of problem. Do the calculations to find the Fratio to compare to the critical or cutoff value from the Fdistribution table (used for ANOVA tests). Find the critical or cutoff value at .05 level of significance using the table to reject the null hypothesis (which would state that all population means are equal). If the null is rejected in favor of the alternate hypothesis (which would state that all population means are not equal), then you can conclude a difference.
One of the most common post hoc tests include Tukey's HSD test. There may be others you can use as well.