It is the average of all MPE ‘s for a binary model or of all OMPE’s for an ordered categorical model.

AMPE<0 indicates that the probability of being in the chosen happiness category de-creases by AMPE units when

1) the corresponding metric correlate increases with one unit
2) the corresponding category of a categorical correlate is compared to the reference category.

AMPE>0 indicates that the probability of being in the chosen happiness category in-creases by AMPE units when

1) the corresponding metric correlate increases with one unit
2) the corresponding category of a categorical correlate is compared to the reference category.

It is the average of all MPE ‘s for a binary model or of all OMPE’s for an ordered categorical model.

AMPE<0 indicates that the probability of being in the chosen happiness category de-creases by AMPE units when

1) the corresponding metric correlate increases with one unit
2) the corresponding category of a categorical correlate is compared to the reference category.

AMPE>0 indicates that the probability of being in the chosen happiness category in-creases by AMPE units when

1) the corresponding metric correlate increases with one unit
2) the corresponding category of a categorical correlate is compared to the reference category.

Type: test statistic
Measurement level: Correlate: metric, Happiness: metric
Theoretical range: unlimited

Meaning:  
b > 0   A higher correlate level corresponds with a higher happiness rating on average.
B < 0   A higher correlate level corresponds with a lower happiness rating on average.
B = 0   Not any correlation with the relevant correlate.

Type: test statistic
Measurement level: Correlate: metric, Happiness: metric
Theoretical range: unlimited

Meaning:  
b > 0   A higher correlate level corresponds with a higher happiness rating on average.
B < 0   A higher correlate level corresponds with a lower happiness rating on average.
B = 0   Not any correlation with the relevant correlate.