As a follow-up of Chuck’s previous post, here are the IQ scores for whites and non-whites, by nativity status, using the NLSY97.

IQ, Race, Nativity (NLSY97)The tables generated by SPSS or PSPP program presents the ASVAB scores for race, broken down by immigrant status. In the table at the top, “White” is a variable for designating non-hispanic whites, created by collapsing the variables R0538700 (race, where 1 = whites) and R0538600 (ethnicity, where 0 = non-hispanic) into a single variable. Here is the syntax we have entered (using SPSS or PSPP ) :

Compute White =0.
If R0538700 =1 and R0538600 =0 White =1.

And here is the sample size for non-Hispanic Whites, with the numbers corresponding to the numbers of US-born grandparents :

,00 = 360
1,00 = 89
2,00 = 386
3,00 = 331
4,00 = 2174

The table at the bottom displays ASVAB scores for minorities, using the variable KEY!RACE_ETHNICITY (R1482600), and where 1 = Black, 2 = Hispanic, 3 = “Mixed Race (Non-Hispanic)”, 4 = “Non-Black / Non-Hispanic”. The numbers (1,00 … 4,00) in the GPUSBORN column correspond to the number of grandparents reported as being born in the US but ,00 means however that all 4 grandparents were either not US born, unknown, or missing data. Here are the samples sizes :

Blacks :
,00 = 104
1,00 = 38
2,00 = 180
3,00 = 121
4,00 = 1127

Hispanics :
,00 = 122
1,00 = 56
2,00 = 176
3,00 = 97
4,00 = 256

Mixed Race :
,00 = 6
1,00 = 2
2,00 = 7
3,00 = 5
4,00 = 25

Non-Black / Non-Hispanic :
,00 = 374
1,00 = 95
2,00 = 409
3,00 = 341
4,00 = 2205

To note, the reported scores are weighted. In the data editor, go to “Data”, “Weigh Cases”, and put the variable R1236201 (SAMPLE WEIGHT PANEL). But if you want to know what the sample size is, just put the sampling weights off by simply checking the box “Do not weight cases”. Then, we have used the Filter Command. Here is the syntax we have entered :

COMPUTE filter_$=(RUSBORN=1).
VALUE LABELS filter_$ 0 ‘Not Selected’ 1 ‘Selected’.
FORMAT filter_$ (f1.0).
FILTER BY filter_$.

Then, highlight it, and click on “Execute”, “Selection”. But to be honest, it’s easier to just go into the data editor, “Data”, “Select Cases”, and check the box “If condition is satisfied”, and then type RUSBORN=1 into the field. Click on “Continue”, and check the box “Filter out unselected cases”. In this way, you can filter off the data more easily.

To note, RUSBORN is a variable created after collapsing, for example the variables, R1201300, S7642200, T0135800, T2110700 and T372170, (questionnaires asking whether the respondent was born in the US), into a single, new variable. Check the NLS investigator for further detail (use the Search option, with the term search “born”).

Tha same process has been repeated for creating the variable GPUSBORN, after collapsing some variables asking whether or not the biological maternal/paternal grand mother/father was born in the US.

Anyway, after filtering, the respondents in the NLSY data should be 2nd+ generation (all born in the U.S.). Individuals with 4 US born grandparents are 4th+ generation.

Next step. On the SPSS data editor, go to “Data”, “Comparison of means”. In the list of dependent variables, put the variable R9829600 (ASVAB). In the list of independent variables (layer 1), put White and R1482600 (KEY!RACE_ETHNICITY). Hit the box “Next”, and in the list of independent variables (layer 2), just put the variable GPUSBORN. Click “OK”.

And you should find the above result.