# Interpreting regression coefficients with a multi-level categorical variable [duplicate]

How do I interpret the coefficients of a Regression with 1 continuous + 1 categorical predictor (with 4 levels - e.g., months)

Specifically, is the 1st coefficient equal to that of the 1st month or equal to the case in which there was no month??

Use this table to help explain:

                        Value    Std.Error    t-value       p-value
(Intercept)      0.2772475891 0.0113379058 24.4531568 8.535808e-106
I(year - 1950)   0.0009540568 0.0002015887  4.7326911  2.504407e-06
factor(season)2  0.0166704599 0.0151419786  1.1009433  2.711624e-01
factor(season)3  0.0769897290 0.0151419786  5.0845224  4.329953e-07
factor(season)4 -0.0096468223 0.0151419786 -0.6370913  5.241981e-01


As a follow-up: what is the interpretation if only some of the categorical variables are significant (and others are not)? Further, do I keep them all if this is the case?

Here is an example of the format of my data:

    year season   temp.avg   ppt.avg       GDD   pdo
1   1922      1  0.4935484 0.3535484  14.40737 -0.45
2   1923      1  4.3892857 0.4542857  56.03017 -1.51
3   1924      2  7.3032258 0.5435484 106.49244 -1.76
4   1925      3 12.8533333 0.2583333 239.07739 -1.71
5   1926      4 19.7903226 0.4667742 458.50000 -1.61
6   1927      1 24.2766667 0.3146667 578.30000 -1.11

>summary(gls(ppt.avg ~ I(year - 1950) + factor(season), data = df, method = 'ML'))
Generalized least squares fit by maximum likelihood
Model: ppt.avg ~ I(year - 1950) + factor(season)
Data: df
AIC       BIC   logLik
-672.3348 -602.2411 350.1674