Completely agree, initially, with the answer of Béatrice Marianne Ewalds-Kvist ;well, that way, SOMETHING COULD BE OBTAINED; Despite this, I am VERY SCEPTIC about the possibility of obtaining some kind of correlation between one variable at the Nominal level and another at the Ordinal level or any other (not even the "fi": phi type)
If I understood correctly, you have (a) a qualitative variable with two categories (musculoskeletal disorder yes / no) and (b) an ordinal qualitative variable (with 5 categories of stress, from less to more).
So you can use the Chi square test, and consider the result that the linear-by-linear association test gives you.
Alternatively, you can consider the results on the stress scale as a continuous variable and assess the association with the presence or absence of musculoskeletal disorder by comparing the means with the Student’s T test.
Both tests can be easily performed with the SPSS. Attached are some screenshots showing how to find them.
Good luck and success with your research. Greetings from Catalonia.