Why don’t the descriptive statistics from Mplus using type=basic match those from another package? | Mplus FAQ

Below we see descriptive statistics for all of the variables in our dataset. Note that the total number of cases in the dataset is 200, but several variables have missing values. Note that several of these variables are categorical, so the means don’t have much meaning substantively, but they can still help us assess whether the data was read in correctly.

    Variable |       Obs        Mean    Std. Dev.       Min        Max
-------------+--------------------------------------------------------
          id |       200       100.5    57.87918          1        200
      female |       182    .5549451    .4983428          0          1
        race |       200        3.43    1.039472          1          4
         ses |       200       2.055    .7242914          1          3
      schtyp |       200        1.16     .367526          1          2
-------------+--------------------------------------------------------
        prog |       182    2.027473    .6927511          1          3
        read |       191    52.28796    10.21072         28         76
       write |       183    52.95082    9.257773         31         67
        math |       185     52.8973    9.360837         33         75
     science |       184    51.30978    9.817833         26         74
-------------+--------------------------------------------------------
       socst |       200      52.405    10.73579         26         71

Data:
  File is descriptives.dat ;
Variable:
  Names are 
     id female race ses schtyp prog read write math science socst; 
  Missing are all (-9999) ; 
Analysis: 
  Type = basic;

Part of the output from the above command file is shown below. The first thing to notice is that the number of observations is listed as 200. Looking at the variable means, notice that for each of the variables with missing values (i.e. female, prog, read, write, math, and science) the mean produced by Mplus does not match the mean shown above. The difference has to do with how Mplus treats missing values, starting in version 5.0, by default Mplus uses full information maximum likelihood to estimate models in which there are any missing values, including the model estimated by type=basic. In other words, the means, and other values produced by Mplus are estimates of what the mean would be if there were no missing values.

SUMMARY OF ANALYSIS

Number of groups                                                 1
Number of observations                                         200

Number of dependent variables                                   11

<output omitted>

     ESTIMATED SAMPLE STATISTICS


           Means
              ID            FEMALE        RACE          SES           SCHTYP
              ________      ________      ________      ________      ________
      1       100.500         0.551         3.430         2.055         1.160


           Means
              PROG          READ          WRITE         MATH          SCIENCE
              ________      ________      ________      ________      ________
      1         2.024        52.358        52.929        52.873        51.612


           Means
              SOCST
              ________
      1        52.405