http://www.talkstats.com/ Psychological statistics, quantitative psychology. Statistics in social sciences. Experimental design, data analysis. en Sat, 25 May 2013 17:33:14 GMT vBulletin 60 http://www.talkstats.com/images/misc/rss.png http://www.talkstats.com/ http://www.talkstats.com/showthread.php/43949-Within-subject-comparison?goto=newpost Sat, 25 May 2013 11:35:08 GMT I randomly exposed subjects (students) to one of two experimental conditions before each of four exams, with their exam performance as the dependent... I randomly exposed subjects (students) to one of two experimental conditions before each of four exams, with their exam performance as the dependent variable. T-tests comparing exam performance under Condition A vs Condition B on each of the exams are not significant, but there is a lot of variability.

I would like to do a within-subject analysis, but am unsure about how to proceed. My main issue is that, due to the random assignment, students were not exposed to the conditions equally. That is, some students might have been exposed to Condition A for all fours exams, others to Condition A three time and Condition B once, etc. An additional concern is that some students only completed three of the four exams.

I know that I will need to eliminate students who were only exposed to one condition from my final data set for a within-subject analysis, but what to do after that? Averaging the results under each condition for each student seems wrong, since the "average" would be just one exam result for some students (i.e., those exposed to one condition three times and the other just once). Should I rank order the test results for each student? That seems like a better strategy to me, but in that case, what test do I use?

Thanks in advance for any help offered! ]]> Psychology Statistics Phoenix123 http://www.talkstats.com/showthread.php/43949-Within-subject-comparison http://www.talkstats.com/showthread.php/43909-Negative-combach-s-alfa-in-evidence-based-questionnaire?goto=newpost Fri, 24 May 2013 19:27:51 GMT Hey everyone, For a small research I had to do I used the Rosenberg Self-esteem scale and tried to look for correlations with aggression using the... Hey everyone,

For a small research I had to do I used the Rosenberg Self-esteem scale and tried to look for correlations with aggression using the Buss Perry Aggression Scale.

Even though it proved very much insignificant I also found a negative crombach's alfa linked to the rosenberg scale.
I checked the coding which was correct. Does anyone have any idea what I'm doing wrong?

http://www.freeuploadsite.com/uploads/13694236301.zip

You can find the syntax and dataset in this file.

Cheers,

Gent.a ]]> Psychology Statistics GentlemanA http://www.talkstats.com/showthread.php/43909-Negative-combach-s-alfa-in-evidence-based-questionnaire http://www.talkstats.com/showthread.php/43905-Can-I-use-a-two-way-repeated-measures-ANOVA?goto=newpost Fri, 24 May 2013 18:39:33 GMT Hi,
I'm doing an acoustic experiment where I have the same 10 subjects listening to/undergoing 18 different conditions. I have 3 independent variables, each of which has different levels, eg. (A - 3 groups) (B- 3 groups) (C- 2 groups), and one dependent variable (d'). I'm trying to do a statistical analysis of this and I'm not sure whether I'm doing the right thing by doing a two way repeated measures ANOVA. Can I use this test for my data with 3 IV and 1 DV?

thanks a lot for your help,
mAcA ]]> Psychology Statistics mAcA http://www.talkstats.com/showthread.php/43905-Can-I-use-a-two-way-repeated-measures-ANOVA