Independently of income, intelligent people tend to be more libertarian

The NLSY97 contains questionnaires that allows us to create a libertarianism index. The questionnaires, listed below with the SPSS syntax used, having 4 values where, 1= Definitely should be, 2= Probably should be, 3= Probably should not be, 4= Definitely should not be, asked whether or not the government should keep the prices under control, reduce income inequality, provide jobs for everyone, and so on.

I summed the different questionnaires, which are moderately correlated, so as to create a more reliable index (Rushton, 1997, chap. 2). By the same logic, I averaged the respondent’s income (2005-2010). Finally, I restrict the analysis for the white sample.

The result from multiple regression is shown as below :

NLSY97 Regression Libertarianism ASVAB holding income and grade constant - Descriptives

NLSY97 Regression Libertarianism ASVAB holding income and grade constant

Generally speaking, multiple regression uses a continuous variable as the dependent variable. And clearly, a 4-point scale does not appear to be a continuous one, but summing the scores of the NLSY97 respondents creates in fact a continuous variable.

Sex/gender is negatively correlated with my libertarianism index. This means that, all else constant, women are more likely to embrace leftist views, with more government interventions. I suspect that one reason could be that they are more anxious, less risk-prone, or something like that.

Education is not correlated with libertarianism, when ASVAB is taken into account (Model 3). I expected a small negative correlation though. I believed that education, holding all else constant, was left biased. Especially today.

Personal income and intelligence are independently associated with libertarian views. No one would be surprised if people tend to be more libertarian as they become richer. But ASVAB is the most important predictor and its independent effect is little affected by the inclusion of the other variables. The zero-order is +0.250 and Beta weight +0.197.

Regarding the histogram of the standardized residual as well as the P-P plot, they are normally distributed. So this is not an issue.

The NLSY97 variables needed are :

S86469.00 YTEL-11A GOVERNMENT RESPONSIBILITY – PROVIDE JOBS FOR EVERYONE 2006
S86470.00 YTEL-11B GOVERNMENT RESPONSIBILITY – KEEP PRICES UNDER CONTROL 2006
S86471.00 YTEL-11C GOVERNMENT RESPONSIBILITY – PROVIDE HEALTH CARE FOR SICK 2006
S86472.00 YTEL-11D GOVERNMENT RESPONSIBLITY – PROVIDE STANDARD OF LIVING FOR OLD 2006
S86473.00 YTEL-11E GOVERNMENT RESPONSIBLITY – PROVIDE INDUSTRY WITH HELP 2006
S86474.00 YTEL-11F GOVERNMENT RESPONSIBLITY – PROVIDE LIVING STANDARD FOR UNEMPLOYED 2006
S86475.00 YTEL-11G GOVERNMENT RESPONSIBLITY – REDUCE INCOME DIFFERENCES 2006
S86476.00 YTEL-11H GOVERNMENT RESPONSIBLITY – PROVIDE COLLEGE FINAN AID 2006
S86477.00 YTEL-11I GOVERNMENT RESPONSIBLITY – PROVIDE DECENT HOUSING 2006
S86478.00 YTEL-11J GOVERNMENT RESPONSIBLITY – PROTECT ENVIRONMENT 2006
R05364.02 KEY!BDATE_Y KEY!BDATE, RS BIRTHDATE MONTH/YEAR (SYMBOL) 1997
R05363.00 KEY!SEX KEY!SEX, RS GENDER (SYMBOL) 1997
R05387.00 KEY!RACE KEY!RACE, RACE OF R (SYMBOL) 1997
R14826.00 KEY!RACE_ETHNICITY KEY!RACE_ETHNICITY, COMBINED RACE AND ETHNICITY (SYMBOL) 1997
R98296.00 ASVAB_MATH_VERBAL_SCORE_PCT ASVAB MATH_VERBAL SCORE PERCENT 1999
Z90838.00 CVC_HGC_EVER RS HIGHEST GRADE COMPLETED XRND
S65010.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2005
S84965.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2006
T08898.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2007
T30030.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2008
T44060.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2009
T60555.00 YINC-1700 TOTAL INCOME FROM WAGES AND SALARY IN PAST YEAR 2010

As for the SPSS syntax I used :

IF R0538700=2 BHW=1.
IF R1482600=2 BHW=2.
IF R0538700=1 and R1482600=4 BHW=3.

RECODE Z9083800 (1 thru 20=COPY) (ELSE=SYSMIS) INTO RGRADE.
EXECUTE.

COMPUTE R_INCOME = MEAN(S6501000, S8496500, T0889800, T3003000, T4406000, T6055500).
EXECUTE.

COMPUTE sqrtRINCOME = SQRT(R_INCOME+1).
VARIABLE LABELS sqrtRINCOME ‘square root of respondent income’.
EXECUTE.

COMPUTE LIBERTARIANISM_SUM=SUM(S8646900, S8647000, S8647100, S8647200, S8647300, S8647400, S8647500, S8647600, S8647700, S8647800).
EXECUTE.

COMPUTE ScaledWeights1=(R1236201*741/217904240).
EXECUTE.
WEIGHT BY ScaledWeights1.

USE ALL.
COMPUTE filter_$=(BHW=3).
VARIABLE LABELS filter_$ ‘BHW=3 (FILTER)’.
VALUE LABELS filter_$ 0 ‘Not Selected’ 1 ‘Selected’.
FORMATS filter_$ (f1.0).
FILTER BY filter_$.
EXECUTE.

REGRESSION
/DESCRIPTIVES MEAN STDDEV CORR SIG N
/MISSING LISTWISE
/STATISTICS COEFF OUTS R ANOVA CHANGE ZPP
/CRITERIA=PIN(.05) POUT(.10)
/NOORIGIN
/DEPENDENT LIBERTARIANISM_SUM
/METHOD=ENTER R0536402 R0536300 sqrtRINCOME
/METHOD=ENTER RGRADE
/METHOD=ENTER R9829600
/PARTIALPLOT ALL
/SCATTERPLOT=(*ZRESID ,*ZPRED)
/RESIDUALS DURBIN HISTOGRAM(ZRESID) NORMPROB(ZRESID)
/CASEWISE PLOT(ZRESID) OUTLIERS(3).

CORRELATIONS
/VARIABLES=S8646900 S8647000 S8647100 S8647200 S8647300 S8647400 S8647500 S8647600 S8647700 S8647800
/PRINT=TWOTAIL NOSIG
/MISSING=PAIRWISE.

NONPAR CORR
/VARIABLES=S8646900 S8647000 S8647100 S8647200 S8647300 S8647400 S8647500 S8647600 S8647700 S8647800
/PRINT=SPEARMAN TWOTAIL NOSIG
/MISSING=PAIRWISE.

FILTER OFF.
USE ALL.
EXECUTE.

One comment. I would not recommend to over-generalize this result to other ethnicities, to other countries, just because it holds true for US whites. There is also the possibility that over time, such relationship might change.

2 comments on “Independently of income, intelligent people tend to be more libertarian

  1. Luke Lea says:

    Maybe because intelligent people know they can make it on their own in the world? Sort of like setting the stakes in poker after you’ve had a peek at your cards.

  2. 猛虎 says:

    I once believed that one reason could be that people wit low intelligence are more religious, and they see the government as an omnipotent identity, like a god, something that can protect them. But over time religiosity has decreased along with economic freedom. So, I think the most plausible explanation now is simply that low intelligent people feel more confident when they believe that an omnipotent identity is protecting them. This is exactly as you said : intelligent people know they can make it on their own in the world. In other words, high IQ people don’t believe they should be nurtured by the state in order to make a living.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s