Robert Zajonc: Family Dynamics and Intellectual Development

By Christine VanDeVelde Luskin, writer and former Bing parent

Is there a connection between birth order and intelligence? This was the question examined by Professor Robert Zajonc, one of the world’s foremost social psychologists, in the 2002 Bing Distinguished Lecture held on May 29th at Bing Nursery School at Stanford University.
Dr. Zajonc began with a look at the 1980 national SAT scores, which had been the highest in many years. The verbal scores reached 506, the quantitative scores 514
—a 20-year high. In the United Kingdom that year, a record number of students passed a comparable examination called the A-levels. Each country reacted in its distinctive way, Professor Zajonc notes. “The United Kingdom was shocked that standards were eroding, that teaching was diluted, that there was grade inflation, that there was cheating,” says Zajonc. “In contrast, the Americans were self-congratulatory. Experts were declaring that more rigorous courses had started to pay off, that the infusion of money into the school system was a good investment, and President Reagan took full credit for the rising SATs at that time in his State Of The Union address.”
But rising SAT scores were not always the case. There was a period when the SAT national averages were dropping, as Zajonc says, “very, very badly and for a long time.” From 1967 to 1980, in verbal SAT mean scores, there was a precipitous drop over 13 years. This data, too, had been greeted with strong reactions and various explanations. Families were blamed for not spending enough quality time with their children. TV was rotting students’ brains. Communities were not giving enough support to schools. There was crime, drugs, smoking, drinking,
rap music, the erosion of standards of excellence. All these things, it was said, contributed to dropping SAT scores.
However, in 1976, Dr. Zajonc published a paper meant to reassure. It said “Don’t worry. This drop is temporary and this trend will not continue. Wait until 1980 and things will change.” And that’s
exactly what happened.
“When I made this prediction in 1976,” says Zajonc, “when scores were falling, a friend, an economist, told me that I was very foolish. I was not doing the kind of thing that economists do. They make
predictions, but they make predictions about the time of change or the direction of change. It is foolhardy to make a
prediction about both, he said. And I made a prediction that in 1980, there
will be a reversal of SAT scores.”
How did Zajonc get it so right? First of all, he notes, there was no evidence that any of the causes people were attributing to declining SAT scores—parental neglect, TV, lack of financial support—had any relationship to the declining scores.
What Zajonc focused on instead was a strong parallel between the falling test scores and family trends. “In particular,” he says, “it appeared that the falling test scores tracked quite accurately the
average orders of birth of children born into this population of children then
taking the SATs.”
What exactly is average birth order? When you are born into a family, you may be a firstborn, a second born, a third born, and so on—that’s your birth order. To get the average birth order for any group of people, you take the number of first births multiplied by one, the number of second births multiplied by two, the number of third births multiplied by
three and so on. Sum these products and divide by the total number of births and you’ll get the average order of birth for that group.
Zajonc recognized a simultaneous decline in average birth orders and SAT scores. Between 1949 and 1962, families were having more and more children, so the average birth order declined. Eighteen years later, these children were taking their SATS
and scores were showing a simultaneous decline. Beginning in 1962, families began to decrease in size, so the average birth order rose. Eighteen years later, Zajonc forecasted, SAT scores would start to rise.
Because SATs are taken by only a
fraction of the total population, but the Census data report all births, there is some discrepancy, what social scientists call “noise” in the data. However, another source of excellent data supports the same conclusions.
In Iowa, every single child is tested every single year. Since the data in Iowa are based on the total population, there should be closer correspondence between average birth order and scores. And, when examined, the correspondence was very close. In Iowa, as in the nation, the lowest average birth order level was for children born in 1962-63, and it was also the year where the scores on the Iowa basic skills test were the lowest. From then on, things changed.
Now, remember that in the United Kingdom, there had been a rise in scores of the A-levels in 1980, corresponding
to the US rise in SAT scores. So, the question was, “Do A-levels in the United Kingdom correspond to the average birth order in the United Kingdom?” Again, the correspondence was very high. And interestingly, the point of change was 1962—the same as in the United States.
“Now,” noted Zajonc, “everything I have told you up until now begs the question: Why should SATs be in any way related to such factors as family size and birth order? Of course, there are some theories about birth order. There are theories that postulate depleting resources within the family; theories about the physiological exhaustion of the mother; theories of ‘uterine fatigue’ as a factor; theories about the association between socioeconomic status and birth order and intelligence. But not all these theories are on solid ground.”
Zajonc’s prediction that test scores would begin to rise, made in 1976, was based on a theory about the emergence of
individual differences in intellectual
performance within the social context
of the family. In 1973, a study was
published that examined the relationship between birth order, family size, and intelligence. The data from that study was drawn on nearly 400,000 recruits
in the Dutch armed forces. And it very clearly showed that intellectual test scores systematically declined with
family size. Within each family, each successive child in the birth order scored a little lower in intelligence.
But this data also contained some
anomalies. The last-born child in a family showed a precipitous drop in intelligence. And there was a discontinuity in a family with only one child. If family size, after all, is a systematic factor in intellectual scores, you would expect the only child to score the highest, but he does not. Those two anomalous factors are related.
Two factors enter into creating and changing the dynamics of the intellectual environment. One of these factors is birth order. However, there is another, very important mitigating factor—the “teaching function.” And this applies to all
children except the last-born or the
only child.
In each family, the older children serve as mentors, teachers, and caretakers of the younger children. That function is important and, in fact, benefits the teacher more than the learner. But the opportunity to teach is reserved only for children who have younger siblings. The only child is a last-born. Any last-born never has anybody to teach. As a consequence, lastborns and only children score lower in intelligence.
“The differences that I have shown you are small differences,” cautioned Zajonc. Such aggregate data of mass proportions cannot be applied to any one individual or any individual family. Birth order and family size effects are risk factors, not determining tendencies that have practical solutions. But, we should know
about them. Because, for a nation or for a community, the effects can be significant.
“Let me give you an illustration of what
I mean,” continued Zajonc. “Let’s take a community of 1,000 individuals. If this community has an average IQ of 100, and a standard deviation of 15, this
community will have 22 individuals who score 130—in the gifted range—and they will also have 22 individuals who score below 70—and these individuals will have to be taken care of by the commu-nity. Now let’s take a community whose mean IQ is 85, one standard deviation lower, and this community will have
only one individual who is gifted and 159 individuals who may be a burden on the community. That is an enormous difference, which accumulates simply by
virtue of very small differences, and has significant and important effects on the community.”
Zajonc admits that IQ is not everything—other effects accrue from different dynamics of the environment in the
family. Other studies have found later-born children are more creative than
earlier born children. And, although
earlier born children are more intelligent, there are questions of leadership, personality differences, effectiveness, and
well-being, which also are effects of
family size. Socioeconomic status has some effect, as well. For each $10,000 of income, Zajonc said, SAT scores rise 16 points, a significant change. Interestingly, no study on birth order has found any gender differences.
While Zajonc hesitated to offer any
policy suggestions based on his data, he believes the effect that accrues from
acting as a teacher should be seriously considered. “We’ve not recognized
sufficiently how much a child can benefit from being a teacher of someone else,” he says. “It would be possible, for
example, to run classes in which half of the students learn about decimals and the other half learn about common denominators. Then, they would teach each other about what they know and perhaps they would both gain a little bit more.”
Dr. Robert Zajonc joined the Stanford faculty six years ago from the University of Michigan, where he received his Ph.D. and was the Charles Horton Cooley Distinguished Professor of the Social Sciences and the Director of the Institute of Social Research. He is a world expert on the psychology of emotions and affect and on the development of preferences and has studied topics as diverse as social behavior in animals and the causes of collective violence in people. He is also a member of the American Academy of Arts and Sciences.

Is there a connection between birth order and intelligence? This was the question examined by Professor Robert Zajonc, one of the world’s foremost social psychologists, in the 2002 Bing Distinguished Lecture held on May 29th at Bing Nursery School at Stanford University.

Dr. Zajonc began with a look at the 1980 national SAT scores, which had been the highest in many years. The verbal scores reached 506, the quantitative scores 514—a 20-year high. In the United Kingdom that year, a record number of students passed a comparable examination called the A-levels. Each country reacted in its distinctive way, Professor Zajonc notes. “The United Kingdom was shocked that standards were eroding, that teaching was diluted, that there was grade inflation, that there was cheating,” says Zajonc. “In contrast, the Americans were self-congratulatory. Experts were declaring that more rigorous courses had started to pay off, that the infusion of money into the school system was a good investment, and President Reagan took full credit for the rising SATs at that time in his State Of The Union address.”

But rising SAT scores were not always the case. There was a period when the SAT national averages were dropping, as Zajonc says, “very, very badly and for a long time.” From 1967 to 1980, in verbal SAT mean scores, there was a precipitous drop over 13 years. This data, too, had been greeted with strong reactions and various explanations. Families were blamed for not spending enough quality time with their children. TV was rotting students’ brains. Communities were not giving enough support to schools. There was crime, drugs, smoking, drinking, rap music, the erosion of standards of excellence. All these things, it was said, contributed to dropping SAT scores.

However, in 1976, Dr. Zajonc published a paper meant to reassure. It said “Don’t worry. This drop is temporary and this trend will not continue. Wait until 1980 and things will change.” And that’s exactly what happened.

“When I made this prediction in 1976,” says Zajonc, “when scores were falling, a friend, an economist, told me that I was very foolish. I was not doing the kind of thing that economists do. They make predictions, but they make predictions about the time of change or the direction of change. It is foolhardy to make a prediction about both, he said. And I made a prediction that in 1980, there will be a reversal of SAT scores.”

How did Zajonc get it so right? First of all, he notes, there was no evidence that any of the causes people were attributing to declining SAT scores—parental neglect, TV, lack of financial support—had any relationship to the declining scores.

What Zajonc focused on instead was a strong parallel between the falling test scores and family trends. “In particular,” he says, “it appeared that the falling test scores tracked quite accurately the average orders of birth of children born into this population of children then taking the SATs.”

What exactly is average birth order? When you are born into a family, you may be a firstborn, a second born, a third born, and so on—that’s your birth order. To get the average birth order for any group of people, you take the number of first births multiplied by one, the number of second births multiplied by two, the number of third births multiplied by three and so on. Sum these products and divide by the total number of births and you’ll get the average order of birth for that group.

Zajonc recognized a simultaneous decline in average birth orders and SAT scores. Between 1949 and 1962, families were having more and more children, so the average birth order declined. Eighteen years later, these children were taking their SATS and scores were showing a simultaneous decline. Beginning in 1962, families began to decrease in size, so the average birth order rose. Eighteen years later, Zajonc forecasted, SAT scores would start to rise.

Because SATs are taken by only a fraction of the total population, but the Census data report all births, there is some discrepancy, what social scientists call “noise” in the data. However, another source of excellent data supports the same conclusions.

In Iowa, every single child is tested every single year. Since the data in Iowa are based on the total population, there should be closer correspondence between average birth order and scores. And, when examined, the correspondence was very close. In Iowa, as in the nation, the lowest average birth order level was for children born in 1962-63, and it was also the year where the scores on the Iowa basic skills test were the lowest. From then on, things changed.

Now, remember that in the United Kingdom, there had been a rise in scores of the A-levels in 1980, corresponding to the US rise in SAT scores. So, the question was, “Do A-levels in the United Kingdom correspond to the average birth order in the United Kingdom?” Again, the correspondence was very high. And interestingly, the point of change was 1962—the same as in the United States.

“Now,” noted Zajonc, “everything I have told you up until now begs the question: Why should SATs be in any way related to such factors as family size and birth order? Of course, there are some theories about birth order. There are theories that postulate depleting resources within the family; theories about the physiological exhaustion of the mother; theories of ‘uterine fatigue’ as a factor; theories about the association between socioeconomic status and birth order and intelligence. But not all these theories are on solid ground.”

Zajonc’s prediction that test scores would begin to rise, made in 1976, was based on a theory about the emergence of individual differences in intellectual performance within the social context of the family. In 1973, a study was published that examined the relationship between birth order, family size, and intelligence. The data from that study was drawn on nearly 400,000 recruits in the Dutch armed forces. And it very clearly showed that intellectual test scores systematically declined with family size. Within each family, each successive child in the birth order scored a little lower in intelligence.

But this data also contained some anomalies. The last-born child in a family showed a precipitous drop in intelligence. And there was a discontinuity in a family with only one child. If family size, after all, is a systematic factor in intellectual scores, you would expect the only child to score the highest, but he does not. Those two anomalous factors are related.

Two factors enter into creating and changing the dynamics of the intellectual environment. One of these factors is birth order. However, there is another, very important mitigating factor—the “teaching function.” And this applies to all children except the last-born or the only child.

In each family, the older children serve as mentors, teachers, and caretakers of the younger children. That function is important and, in fact, benefits the teacher more than the learner. But the opportunity to teach is reserved only for children who have younger siblings. The only child is a last-born. Any last-born never has anybody to teach. As a consequence, lastborns and only children score lower in intelligence.

“The differences that I have shown you are small differences,” cautioned Zajonc. Such aggregate data of mass proportions cannot be applied to any one individual or any individual family. Birth order and family size effects are risk factors, not determining tendencies that have practical solutions. But, we should know about them. Because, for a nation or for a community, the effects can be significant.

“Let me give you an illustration of what

I mean,” continued Zajonc. “Let’s take a community of 1,000 individuals. If this community has an average IQ of 100, and a standard deviation of 15, this community will have 22 individuals who score 130—in the gifted range—and they will also have 22 individuals who score below 70—and these individuals will have to be taken care of by the commu-nity. Now let’s take a community whose mean IQ is 85, one standard deviation lower, and this community will have only one individual who is gifted and 159 individuals who may be a burden on the community. That is an enormous difference, which accumulates simply by virtue of very small differences, and has significant and important effects on the community.”

Zajonc admits that IQ is not everything—other effects accrue from different dynamics of the environment in the family. Other studies have found later-born children are more creative than earlier born children. And, although earlier born children are more intelligent, there are questions of leadership, personality differences, effectiveness, and well-being, which also are effects of family size. Socioeconomic status has some effect, as well. For each $10,000 of income, Zajonc said, SAT scores rise 16 points, a significant change. Interestingly, no study on birth order has found any gender differences.

While Zajonc hesitated to offer any policy suggestions based on his data, he believes the effect that accrues from acting as a teacher should be seriously considered. “We’ve not recognized sufficiently how much a child can benefit from being a teacher of someone else,” he says. “It would be possible, for example, to run classes in which half of the students learn about decimals and the other half learn about common denominators. Then, they would teach each other about what they know and perhaps they would both gain a little bit more.”

Dr. Robert Zajonc joined the Stanford faculty six years ago from the University of Michigan, where he received his Ph.D. and was the Charles Horton Cooley Distinguished Professor of the Social Sciences and the Director of the Institute of Social Research. He is a world expert on the psychology of emotions and affect and on the development of preferences and has studied topics as diverse as social behavior in animals and the causes of collective violence in people. He is also a member of the American Academy of Arts and Sciences.