Correlations

[Damian Chroboczek]

Hello, I'm writing a thesis about mismatch on the Labour Market. I have in my dataset dummy variables for mismatch, 0 for match, 1 for mismatch (overeducation) and the same variables for undereducation. Also I have gender 1 = man, 2 = woman, age, monthly wage. I want the mistmach correlate with gender, age and wage, but I don't know which command and which correlations is the best to use. The main is to get from correlations is a p-value. Thank you very much for help!

[Maarten Buis]

These are a combination of categorical and continuous variables, so I will assume you mean association and not correlation. I will also assume that you want those p-values to refer to bivariate association. In that case you could do something like the example below:

. // open example data
. sysuse nlsw88, clear
(NLSW, 1988 extract)

.
. // prepare some data
. gen byte urban = c_city + smsa

. label define urban 2 "central city" ///
>                    1 "suburban"     ///
>                    0 "rural"

. label value urban urban

. label variable urban "urbanicity"

.
. // the command:
. dtable i.urban i.south wage, by(race, test)
note: using test pearson across levels of race for urban and south.
note: using test regress across levels of race for wage.

----------------------------------------------------------------------------------
                                                 Race                             
                       White         Black         Other          Total      Test
----------------------------------------------------------------------------------
N                  1,637 (72.9%)   583 (26.0%)     26 (1.2%) 2,246 (100.0%)       
urbanicity                                                                        
  rural              513 (31.3%)   145 (24.9%)     7 (26.9%)    665 (29.6%) <0.001
  suburban           762 (46.5%)   152 (26.1%)    12 (46.2%)    926 (41.2%)       
  central city       362 (22.1%)   286 (49.1%)     7 (26.9%)    655 (29.2%)       
Lives in the south                                                                
  Not south        1,071 (65.4%)   210 (36.0%)    23 (88.5%)  1,304 (58.1%) <0.001
  South              566 (34.6%)   373 (64.0%)     3 (11.5%)    942 (41.9%)       
Hourly wage        8.083 (5.955) 6.845 (5.076) 8.551 (5.209)  7.767 (5.756) <0.001
----------------------------------------------------------------------------------