If we expand the range of outcomes that we care about beyond economics, I’ll bet we find strong impacts from parenting that are not highly genetic. How many children will they have? Will they believe in God? How do they think about relationships, art, where do they find community and meaning in life? Maybe these studies have been done already?
This doesn't seem like too much of mystery if you remember what the twin studies actually say.
They inherently condition any results on adoptive parents. I'm no expert in the criteria used as a gate to adoption, but they appear to include factors like:
- Two parent household (or extremely capable single parent)
- Some minimal amount of stable social connections
- History of conscientiousness in general
- Enough material resources to provide a basic quality of life
- Sincere desire to have children
Once you have these basic macro factors in place, micro factors don't seem to matter. Reading Bryan's book, one of my takeaways was the difference between macro-parenting and micro-parenting. Focus on the macro; don't sweat the micro. Spend engaged time with your children. Don't worry about what you do during that time.
Now, this bar doesn't seem like much to people like Bryan. But it in no way conflicts with the "Success Sequence" hypothesis.
My daughter was the sweetest little kid – empathetic and mature well beyond her years. People who met her would compliment me and my wife on our parenting, but we kind of knew better (“actually we’re not doing anything, she’s just a sweet kid”).
Fast forward a few years to my son who is completely different from my daughter. My wife and I really feel like we’re taking exactly the same approach to parenting both kids. But needless to say, we get far fewer compliments with him.
I know a sample size of 1 isn’t robust, but I really don’t know how to explain the difference between our kids’ behavior by anything other than massive genetic effects on a whole host of behaviors.
I suspect some aspects of parenting matter much more than others and we are often wrong about which are which. Llikewise, I suspect parenting has a much greater impact on some kids than others. That said, I have three thoughts.
2 All kids need good models, whether it be peers, parents, or others. Maybe it works best if they are peers, maybe not.
3 Not all kids need the same thing and those needs may change as they mature. Some need lots of rules, some less. Some need rigid enforcement, others not. Learning styles vary. Some need challenged, others need more opportunity to explore on their own. Etc.
I work in edtech, serving K12 (charter and public in the US) and all the research we refer to in the ed space shows one of the strongest correlations between parent engagement and student success of any factor.
Surely education research establishing parent engagement with schools as key to student success supports the idea that parenting is a factor on children's overall outcomes...
Might pay to look at research strands like this? I'm not an academic, so some of these references may not be from university research!
-"Parent involvement is consistently associated with higher academic achievement measures, including higher test scores, grades, and graduation rates." - Harvard Family Research Project
-"Parental involvement is a key strategy for closing the achievement gap, particularly for low-income students. When parents are engaged in their children's education, they can help address the systemic challenges that contribute to educational disparities." - Education Trust
-According to a report by the National Center for Family and Community Connections with Schools:
Students from low-income families with involved parents are more than twice as likely to succeed academically compared to their peers with less parental involvement. Parental involvement in education can account for up to a 10-point increase in standardized test scores.
I agree with the general gist that past a certain point parental involvement has a smallish effect, but it might be worth pointing out the areas where it does seem to have an outsized impact.
For example, I believe there are very lopsided statistics about white collar professional choice among people who already have parents in that profession. There's almost certainly a genetic effect in play, but if offspring of professors are 15x more likely to go into academia compared to non-offspring then surely there's a type of socialization that's happening while growing up.
I haven't looked into it, but I imagine access to other sorts of zero-sum societal goods might show similar effects as well.
A longstanding, occasionally fashionable hypothesis in research has been that, in families with more than one child, *birth order* shapes personality. However, recent research, with sounder methods, finds that birth order has some effect, on average, on intelligence, but no effect , on average, on extraversion, emotional stability, agreeableness, conscientiousness, or imagination.
See Rohrer et al., "Examining the effects of birth order on personality" (Proceedings of the National Academy of Sciences: Psychological and Cognitive Sciences, 2015):
Oddly one of the things going on now with AI systems is that vendors are trying to figure out in essence how to "parent" them in terms of teaching them how to behave. Should they respond in a "woke" fashion? Should they display pictures of Mohammed, contrary to the views of many Muslims? What political or religious viewpoints should they express by default, or have subtly built into implicit bias underlying their responses?
OpenAI has put out a call for proposals for "democratic input" from the global populace to make these decisions, and Anthropic (another major LLM vendor) has also discussed wanting democratic input: essentially the "it takes a village" approach to what might be viewed as raising AI children. Yet that risks 1 size fits all tyranny of the majority, or Taleb's dictatorship of the most intolerant minority. There are culture wars over what Americans teach their kids, imagine a global choice for the AIs that will be nudging everyone writing content in a year or so, or teaching kids, etc.
What are global norms likely to be on free markets and free speech if they decide them democratically?
Another approach is to push for many diverse AIs, or personas of an AI, that correspond to each culture rather than one size fits all. In a transparent appeal to leftists, this site argues for a rainbow of AI with different viewpoints:
I think a lot of the research and twin studies assumes a certain level of first world middle class stability within an overall normie culture. That's probably true when you are deciding whether to make your kid eat their broccoli tonight, but not necessarily true if you're contemplating out of wedlock birth or divorce or doing drugs.
A lot of them are also limited to socioeconomic outcomes. Yeah, you will probably end up earning what people with your IQ level end up earning. But how do you get there? Was the journey joyful? Was there any way it could have gone better?
For what it's worth something of a natural experiment was run between me and my half brother. I had the normie two parent up brining. He had a mother that partied and did drugs and married five other dudes along the way. I turned out good and my half brother has had a basket case life. Though he probably has the same IQ as I do.
I suspect that while good parenting may not be crucial, bad parenting is.
If we expand the range of outcomes that we care about beyond economics, I’ll bet we find strong impacts from parenting that are not highly genetic. How many children will they have? Will they believe in God? How do they think about relationships, art, where do they find community and meaning in life? Maybe these studies have been done already?
This doesn't seem like too much of mystery if you remember what the twin studies actually say.
They inherently condition any results on adoptive parents. I'm no expert in the criteria used as a gate to adoption, but they appear to include factors like:
- Two parent household (or extremely capable single parent)
- Some minimal amount of stable social connections
- History of conscientiousness in general
- Enough material resources to provide a basic quality of life
- Sincere desire to have children
Once you have these basic macro factors in place, micro factors don't seem to matter. Reading Bryan's book, one of my takeaways was the difference between macro-parenting and micro-parenting. Focus on the macro; don't sweat the micro. Spend engaged time with your children. Don't worry about what you do during that time.
Now, this bar doesn't seem like much to people like Bryan. But it in no way conflicts with the "Success Sequence" hypothesis.
My daughter was the sweetest little kid – empathetic and mature well beyond her years. People who met her would compliment me and my wife on our parenting, but we kind of knew better (“actually we’re not doing anything, she’s just a sweet kid”).
Fast forward a few years to my son who is completely different from my daughter. My wife and I really feel like we’re taking exactly the same approach to parenting both kids. But needless to say, we get far fewer compliments with him.
I know a sample size of 1 isn’t robust, but I really don’t know how to explain the difference between our kids’ behavior by anything other than massive genetic effects on a whole host of behaviors.
I suspect some aspects of parenting matter much more than others and we are often wrong about which are which. Llikewise, I suspect parenting has a much greater impact on some kids than others. That said, I have three thoughts.
1 I'm sure there are some other activities parents could do with kids to improve outcome but this one is probably more certain in benefit than most. https://www.greatschools.org/gk/articles/word-gap-speak-more-words-to-your-preschooler-daily/
2 All kids need good models, whether it be peers, parents, or others. Maybe it works best if they are peers, maybe not.
3 Not all kids need the same thing and those needs may change as they mature. Some need lots of rules, some less. Some need rigid enforcement, others not. Learning styles vary. Some need challenged, others need more opportunity to explore on their own. Etc.
I work in edtech, serving K12 (charter and public in the US) and all the research we refer to in the ed space shows one of the strongest correlations between parent engagement and student success of any factor.
Surely education research establishing parent engagement with schools as key to student success supports the idea that parenting is a factor on children's overall outcomes...
Might pay to look at research strands like this? I'm not an academic, so some of these references may not be from university research!
-https://www.waterford.org/education/how-parent-involvment-leads-to-student-success/ (this one references five different studies)
-"Parent involvement is consistently associated with higher academic achievement measures, including higher test scores, grades, and graduation rates." - Harvard Family Research Project
-"Parental involvement is a key strategy for closing the achievement gap, particularly for low-income students. When parents are engaged in their children's education, they can help address the systemic challenges that contribute to educational disparities." - Education Trust
-According to a report by the National Center for Family and Community Connections with Schools:
Students from low-income families with involved parents are more than twice as likely to succeed academically compared to their peers with less parental involvement. Parental involvement in education can account for up to a 10-point increase in standardized test scores.
The evidence against parenting styles mattering is sticker charts
I agree with the general gist that past a certain point parental involvement has a smallish effect, but it might be worth pointing out the areas where it does seem to have an outsized impact.
For example, I believe there are very lopsided statistics about white collar professional choice among people who already have parents in that profession. There's almost certainly a genetic effect in play, but if offspring of professors are 15x more likely to go into academia compared to non-offspring then surely there's a type of socialization that's happening while growing up.
I haven't looked into it, but I imagine access to other sorts of zero-sum societal goods might show similar effects as well.
A longstanding, occasionally fashionable hypothesis in research has been that, in families with more than one child, *birth order* shapes personality. However, recent research, with sounder methods, finds that birth order has some effect, on average, on intelligence, but no effect , on average, on extraversion, emotional stability, agreeableness, conscientiousness, or imagination.
See Rohrer et al., "Examining the effects of birth order on personality" (Proceedings of the National Academy of Sciences: Psychological and Cognitive Sciences, 2015):
https://www.pnas.org/doi/10.1073/pnas.1506451112
Perhaps birth order matters in some circumstances or in particular families?
Oddly one of the things going on now with AI systems is that vendors are trying to figure out in essence how to "parent" them in terms of teaching them how to behave. Should they respond in a "woke" fashion? Should they display pictures of Mohammed, contrary to the views of many Muslims? What political or religious viewpoints should they express by default, or have subtly built into implicit bias underlying their responses?
OpenAI has put out a call for proposals for "democratic input" from the global populace to make these decisions, and Anthropic (another major LLM vendor) has also discussed wanting democratic input: essentially the "it takes a village" approach to what might be viewed as raising AI children. Yet that risks 1 size fits all tyranny of the majority, or Taleb's dictatorship of the most intolerant minority. There are culture wars over what Americans teach their kids, imagine a global choice for the AIs that will be nudging everyone writing content in a year or so, or teaching kids, etc.
What are global norms likely to be on free markets and free speech if they decide them democratically?
Another approach is to push for many diverse AIs, or personas of an AI, that correspond to each culture rather than one size fits all. In a transparent appeal to leftists, this site argues for a rainbow of AI with different viewpoints:
https://RainbowOfAI.com
I think a lot of the research and twin studies assumes a certain level of first world middle class stability within an overall normie culture. That's probably true when you are deciding whether to make your kid eat their broccoli tonight, but not necessarily true if you're contemplating out of wedlock birth or divorce or doing drugs.
A lot of them are also limited to socioeconomic outcomes. Yeah, you will probably end up earning what people with your IQ level end up earning. But how do you get there? Was the journey joyful? Was there any way it could have gone better?
For what it's worth something of a natural experiment was run between me and my half brother. I had the normie two parent up brining. He had a mother that partied and did drugs and married five other dudes along the way. I turned out good and my half brother has had a basket case life. Though he probably has the same IQ as I do.