Here is the problem. I conducted a small pilot research on multicultural differences in the perception of giftedness. I asked my students, half Arabs half Jewish, to score the level of giftedness for 5 people: a politician, a sportsman. An artist etc… each figure received a grade 0-100 according to its giftedness in the eye of the student.

Yet, Each group received a different list, for example, a Hebrew famous poet (Bialik) for the Jewish versus a famous Arabic poet matched for the Palestinians (Jubran).

Now I want to compare means, and see if the difference in rating are significant,

The Independent variable is nationality; the dependent is the level of giftedness rated by each students to each figure in the list.

In the SPSS Data fileI built, each figure had a different name and was rated by a different group.
In order to conduct a paired sample t- test I need to match them somehow.
However, how?