The Bell Curve:

1
The Bell Curve
Intelligence and Class Structure in American Life
Richard J. Herrnstein Charles Murray
2
Emergence of the Cognitive Elite
The twentieth century dawned on a
world segregated into social classes defined
in terms of money, power, and status. ...
Our thesis is that the twentieth century
has continued the transformation of the
origin of social class, so that the
twenty-first will open on a world in which
cognitive ability is the decisive driving force.
... Social class remains the vehicle of
social life, but intelligence now pulls the
train.
Herrnstein and Murray, p. 25
3
Arguments
1) Greater access to higher education.
3) Occupational status has became more dependent
upon cognitive ability.
4
Assumptions Evidence
1. Intelligence, as measured by IQ tests, is
meaningful and stable.
2. Intelligence runs in families assortative
mating for intelligence.
3. Intelligence has a direct causal role in
eventual social status.
4. Society has become increasingly efficient
in selecting for intelligence in education
and occupations.
5
Predictions

• An increasingly isolated cognitive elite.
• A merging of the cognitive elite with the
  affluent.
• A deteriorating quality of life for people at the
  bottom end of the cognitive ability distribution.

Herrnstein and Murray, , p. 509
6
Unchecked, these trends will lead the U.S.
towards something resembling a caste system, with
the underclass mired ever more deeply at the
bottom and the cognitive elite ever more firmly
anchored at the top, restructuring the rules of
society so that it becomes harder and harder for
them to lose.
Herrnstein and Murray, p.509
7
Topics
1. What is intelligence?
2. How stabile is intelligence?
3. Does it run in families?
4. Does it influence eventual social status?
5. Is a meritocracy evolving?
8
1. What is Intelligence?
9
Analogies
past is to present as present is to ______
dog is to god as rat is to ______
10
Vocabulary
Respirate means
A. sweat
B. give back
C. breathe
D. continue
Garrulous means
A. angry
B. talkative
C. unruly
D. stupid
11
Sentences
Rearrange the following
him not ask do.
four square every has sides.
12
Syntactic Inference
(fill in the blank)
A pervish wib kelted edfully.
I bellishly smusted the ____.
a) pervish
b) edfully
c) kelted
d) wib
13
Numerical Reasoning
Jane is 6. She is twice as old as Billy.
How old will Billy be when Jane is 8?
14
Number Series
2, 4, 8, 16, ___ .
2, 3, 5, 8, ___ .
15
Figure Completions
____
a
b
c
d
e
16
Correlates of IQ Tests
1) Other IQ tests (.90)
2) Standardized Tests
2.a) verbal math ability (.80)
2.b) content areas (.70)
3) School grades (.60)
17
2. How stabile is intelligence?
18
Stability of IQ
Up to about 4 or 5 ..., measures of IQ are not
of much use in predicting later IQ. Between
ages 5 and 10, the tests rapidly become more
predictive of adult IQ. After about the age of
10, the IQ score is essentially stable within the
constraints of measurement error.
Herrnstein Murray, p. 130
19
Stability of IQ
Age
Age
7
10
9
5
6
8
5
1.00
6
.88
1.00
7
.78
.87
1.00
8
.81
.88
.91
1.00
9
.77
.87
.91
.92
1.00
10
.78
.83
.90
.71
.94
1.00
Fels Longitudinal Study (N80)
20
Stability of IQ
Age
Age
12
17
15
10
11
14
10
1.00
11
.94
1.00
12
.92
.93
1.00
14
.86
.86
.90
1.00
15
.86
.86
.90
.88
1.00
17
.69
.67
.74
.71
.89
1.00
Fels Longitudinal Study (N80)
21
3. Does intelligence run in families?
22
Correlations Raised Together
Identical Twins
.86
.60
Fraternal Twins
.47
Siblings
.42
Parent-offspring
Half-siblings
.31
Cousins
.15
Mates
.40
Bouchard McGue (1981), Science, 212, 1055-1059.
23
Correlations Raised Apart
Identical Twins
.72
.24
Siblings
.22
Parent-offspring
Bouchard McGue (1981), Science, 212, 1055-1059.
24
Correlations Nongenetic Relatives
.31
Adoptive Siblings
.19
Adoptive Parent-offspring
Bouchard McGue (1981), Science, 212, 1055-1059.
25
Correlations
Raised Together
Raised Apart
Identical Twins
.72
.86
.24
.47
Siblings
.22
.42
Parent-offspring
Bouchard McGue (1981), Science, 212, 1055-1059.
26
4. Does intelligence influence eventual social
status?
27
Herrnstein and Murray, p. 149
28
Herrnstein and Murray, p. 134
29
Social Outcomes and IQ
Percent
Cognitive Class
Unemployed gt 1 month (1989)
Poverty (1989)
No HS Diploma
Chronic Welfare
Divorced
2
0
2
0
9
I. Very Bright
3
0
7
2
15
II. Bright
6
6
7
8
23
III. Normal
16
35
10
17
22
IV. Dull
30
55
12
31
21
V. Very Dull
Herrnstein Murray, NLSY data (whites)
30
Social Outcomes and IQ
Percent
Cognitive Class
Low Birth- weight
Illegitimate Child
Low IQ Child
Low HOME Index
5
2
0
0
I. Very Bright
2
4
2
7
II. Bright
3
8
6
6
III. Normal
7
17
11
17
IV. Dull
6
32
24
39
V. Very Dull
Herrnstein Murray, NLSY data (whites)
31
IQ and Social Status
Fathers
Own
IQ
Education
Occupation
.38
.37
.47
Education
.26
.29
.36
Occupation
.18
.18
.35
Income
PSID Data (N1774)
32
IQ and Social Status
Fathers
Own
IQ
Education
Occupation
.30
.31
.55
Education
.26
.30
.43
Occupation
.14
.27
.35
Income
Veterans (N803)
33
IQ and Social Status
Fathers
Own
IQ
Education
Occupation
.40
.38
.58
Education
First
.35
.39
.45
Occupation
Later
.22
.22
.45
Occupation
.17
.20
.36
Income
Kalamazoo Brothers (N300)
34
Standardized Regression Coefficients
Independent Variables
Family Back- ground
Educa- tion
Occupa- tion
Dependent Variable
Test
Test
.43
Education
.29
.44
Occupation
.06
.15
.53
Earnings
.06
.06
.08
.15
Project Talent (N839)
35
Standardized Regression Coefficients
Independent Variables
Family Back- ground
Occupa- tion
Educa- tion
Dependent Variable
Test
Test
.25
.28
.40
Education
Occupation
.07
.07
.55
-
Earnings
.01
.17
.24
.20
PSID (N1,774)
36
Standardized Regression Coefficients
Independent Variables
Family Back- ground
Occupa- tion
Educa- tion
Dependent Variable
Test
Test
.34
.30
.45
Education
Occupation
.16
.12
.45
-
Earnings
.14
.20
.01
.23
Veterans (N803)
37
Standardized Regression Coefficients
Independent Variables
Family Back- ground
Dependent Variable
Current Occ.
First Occ.
Educa- tion
Test
Test
.38
Education
.41
.43
First Occup.
.19
.03
.59
-
Current Occ.
.09
.29
.15
.34
.03
Current Earn.
.12
.14
.10
.21
Kalamazoo Brothers (N300)
38
FB
IQ

ED
OC
39
IQ and Crime
Among the most firmly established facts about
criminal offenders is that their distribution of
IQ scores differs from that of the population at
large. Taking the scientific literature as a
whole, criminal offenders have average IQs of
about 92, eight points below the mean.
Herrnstein and Murray, p. 235
40
IQ and Crime Conclusions
In trying to understand how to deal with the
crime problem, much of the attention now given to
problems of poverty and unemployment should
be shifted to another question altogether coping
with cognitive disadvantage.
Herrnstein and Murray, p. 251
41
Crime and IQ
Percent
Cognitive Class
Ever Convicted
Ever Incarcerated
Stopped by Police
Ever Booked
3
18
0
I. Very Bright
5
7
27
1
II. Bright
12
15
37
3
III. Normal
20
21
46
7
IV. Dull
27
14
7
V. Very Dull
17
33
Herrnstein Murray, NLSY data (white males), p.
247
42
IQ and Delinquency Boys
adapted from White, Moffitt, Silva (1989)
43
IQ and Delinquency Girls
adapted from White, Moffitt, Silva (1989)
44
Substance Abusing and Control Male Teenage
Probands
ASA-I (unpublished)
45
Relatives of Substance Abusing and Control Male
Teenage Probands
ASA-I (unpublished)
46
.46
P
iq
-.13
-.05
-.07
.01
-.06
-.03
alco- hol
CD/ ASP
drugs
ASA-I (unpublished)
47
5. Is a cognitive meritocracy evolving?
48
Heritocracy Meritocracy
49
Rethinking Ethnic Differences
If the reader is now convinced that either the
genetic or environmental explanation has won out
to the exclusion of the other, we have not done
a sufficiently good job of presenting one side or
the other. It seems highly likely to us that both
genes and the environment have something to do
with racial differences. What might the mix
be? We are resolutely agnostic on that issue.
Herrnstein and Murray, p. 311
50
And, on the next page ...
... we have found that the genetic aspect of
ethnic differences has assumed an overwhelming
importance. One symptom of this is that while
this book was in preparation and regardless of
how we described it to anyone who asked, it was
assumed that the books real subject had to be
not only ethnic differences in cognitive ability
but the genetic source of those differences.
Herrnstein and Murray, p. 312