Hey together,

in context of my master thesis I found a significant correlation between emotion recognition ability and annual income (= only ordinal scaled variable).

After controlling for age, hierarchical position and education level due to an ordinal regression analysis, emotion recognition ability was no longer significantly linked to the criterion annual income. Instead, age, some classes of hierarchical position and education level revealed highly significant.      

Further significant intercorrelations were found between Annual income with age .54**, Hierarchical position .45** Education level .44**, and Emotion recognition ability .34*

Hierarchal position correlated with age .36**

How would you guys interpret these results? Do you see some theoretical explanations for that findings?

Do you know further statistical analysis that shows a better overview about the coherences of variables despite the ordinal level of data?   

Many thanks in advance!

Patrick 

Note.

* p < .05, ** p < .01

Hierarchical position is coded 0 = first line manager, 2 = middle manager, 3 = senior manager;

c Educational level is coded 1 = no degree, 2 = secondary school, 3 = high school, 4 = job training, 5 = advanced job training, 6 = bachelor’s degree, 7 = master’s degree;

d Annual income in Euro is coded 1 = 20,000, 2 = between 20,000 and 30,000, 3 = between 30,000 and 40,000, etc. until 10 = more than 100,000; 

For all variables except annual income N = 59; Annual income N = 57