CCBC-Net Archives
Re: Multicultural Literature - Causal Directionality
- Contemporary messages sorted: [ by date ] [ by subject ] [ by author ]
From: Charles Bayless <charles.bayless_at_ttmd.com>
Date: Fri, 07 Feb 2014 14:41:10 -0500
Both business decisions and government policy decisions are plagued by two fallacies, post hoc ergo propter hoc and cum hoc ergo propter hoc, and I suspect we are falling prey to one or both. Post hoc is concluding that because one thing happened after another, then it must have been caused by it. Cum hoc is even more common and that is mistaking correlation for causation. Only 20% of corporate projects are delivered on time, on budget and achieve their stated goals. Only 10% of government policy implementations do so. These two fallacies are behind both those results. It is often hard to determine 1) causation, and 2) the directionality of causation. Advocates for more POC authors and content argue that low POC publication rates cause the low reading rates among POC communities and therefore low demand. Publishers, librarians, and retailers claim that there is low demand for POC content/authors and that is why there are fewer POC authors/content books. Chicken and egg. Supply Side Hypothesis (need more POC authors/content) or Demand Side Hypothesis (need more demand for such content). Of course, it is also possible that both are equally true. It is even more probable that both have some, but perhaps unequal, degrees of truth. But which and how much? What is the evidence for either hypothesis?
A couple of neutral examples of the prevalence of this causal directionality problem. I am head of a couple urban green space/park conservation groups. We have a standard list of arguments we use in our advocacy. Ten years ago, a common argument was that better and increased parks and green space in urban environments would help address health and obesity issues that are prevalent in such environments. It was a logical argument - more green space, more opportunity to exercise, better health, lower obesity rates. We do not use that argument these days because the direction of causation was wrong. In the past ten years there have been a couple of reasonably robust longitudinal studies that indicate that presence or absence of urban green space has no impact on community health and obesity. It's not that people couldn't exercise because there was no green space. There was little green space because people weren't interested in exercising. In our efforts to do good (and there are other more important reasons for urban green spaces, health was just one of the arguments used), we actually exacerbated things a bit. It was not the sedentary and obese (who happened to also be poor) who used the new green spaces but the already fit (who were usually well off) - creating an impression around classism. It is not enough to want to do good, you actually have to achieve it and that can be complicated. Because of the longitudinal studies, it can no longer be argued that health and obesity are likely to be improved by increased green space.
I just saw yesterday a similar example of mistaken causal directionality. Over the past five years there has been a five hundred million dollar effort to address food deserts (urban environments with reduced access to healthy food, particularly fresh fruit and vegetables). The argument has been that people have poor health and high obesity rates in these areas because there is poor access to fresh F&V. The government subsidized a number of pilot projects to bring fresh F&V to these communities and test the hypothesis. The results study is out for the first five years and it appears that there is no change in consumption and certainly no change in obesity rates and health outcomes. (New Neighborhood Grocery Store Increased Awareness Of Food Access But Did Not Alter Dietary Habits Or Obesity by Steven Cummins, et al. http://content.healthaffairs.org/content/33/2/283.abstract) If these results persist, as is expected that they will, then it will be another example of mistaken causation. Poor health and obesity were assumed to be caused by low availability of fresh F&V. However, it now appears that low availability of fresh F&V happened because there was low demand.
In terms of children's books, we have one observable fact and that is a low and very constant level of reading books by and about minorities, approximately 15% over the past twenty years per the CCBC numbers, taking those numbers at face value (3,200 mass market trade books as part of the 25-35,000 new children's titles released each year, i.e. are the figures for mass market trade books reflective of the total population?; and performing the ratio calculations on the correct base, i.e. omitting animal based books, etc.).
There is an argument to be made that even that low level is explicable based on reading demographics rather than population demographics. Different groups in the US have different reading proclivities (amount of time spent reading overall, amount of time spent reading books, amount of money spent on books). According to Pew, ( http://libraries.pewinternet.org/2012/12/27/e-book-reading-jumps-print-book- reading-declines/), only 75% of the population reads any book at all in a year. If you look at only those who are electively reading (i.e. not those who read a book for work or study), the number drops to 50%. So readers are a self-selected group whose demographics are likely to be different than those of the population at large. And they are different. In that Pew study, a higher percentage of the white population read a book in a year
(78%) than do African Americans (74%) or Hispanics (60%). Even more critically, white readers read 33% more books than either of the other two groups. The implication is that while whites might be 70% of the population
(depending on definitions of ethnicity and race), they are 80% of the readers. This is consistent with two other sources of data: For book reading, the ethnic distribution based on Bureau of Labor Statistics (BLS) time-use data for average time spent reading per year by ethnic group and Kaiser Foundation data for reading form preference (books vs. newspapers and magazines) by ethnic group is 82% white, 4% African American, 6% Asian American, 8% Hispanic, and 0.4% Native American. The distribution in terms of money spent on reading (BLS Consumer Expenditures 2011) is 85% white, 6% African American, 4% Asian, and 5% Hispanic (Native American numbers not broken out in the BLS expenditure data).
The net of this is that the available data indicates that the race distribution of the CCBC numbers (85% white and 15% POC), while inconsistent with at-large population demographics, are quite close to the numbers for those who are reading books (80% based on Pew numbers, 82% based on BLS time-use, and 85% based on BLS consumer expenditures). This makes logical sense. But in terms of rectifying the disparity, it points away from publishers and towards the reading habits of the at-large population. If we want to bring the reading demographics into alignment with the population demographics, then the different populations need to be consuming books at comparable rates.
But setting all that aside, we still have a chicken and egg conundrum. Several posts have made the logical inference that minority reading figures are low because they can't find minority authors/content to read. If POC authors were published more, then POC would read more. In other words, that there is a supply side issue. (BTW, I am uncomfortable with the blanket POC term because the reading numbers vary significantly between census groups, but use the term for simplicity). But what if the causal direction is the opposite and there is a demand side problem instead. This is important to resolve because the solutions to supply side problems are quite different from solutions to demand side problems.
The supply side hypothesis has been the dominant assumption since 1965 with Nancy Larrick's article, The All-What World of Children's Books. In the nearly five decades since then, there has been persistent advocacy to increase minority presence in books (both authorial and protagonist). Most/all the major publishing houses have had internal initiatives towards that end. Most/all of them have some form of multicultural imprint. SCBWI and ALA have had (and still have) various initiatives. Minority oriented prizes such as CSK and Pura Belpre have emerged to increase attention and market interest and to recognize new talent. Substantial independents such as Lee & Low have been founded and grown with a target focus on minority publication. Black History month provides a means of focusing on minority oriented reading in bookstores and libraries. All of these activities are intended to bring greater attention and supply to POC oriented reading. Despite all these efforts to increase supply, as the CCBC numbers show, and as illustrated by the Lee & Low graphic (once adjusted to the correct base), and as Cheryl Klein highlights
(http://chavelaque.blogspot.com/2014/02/in-defense-of-cbc-diversity-and-on.h tml), the percentage of minority presence remains ~15%.
Maybe the problem is not a supply side issue but a demand side issue. What is the data for that hypothesis?
The BLS , Pew and Kaiser Foundation data all document that there is indeed a material component that is demand side. If different groups spend less time reading, read fewer books and spend less money on reading, then that is explicitly a demand side problem. Providing more authors is unlikely to make a material difference unless their volume of reading increases.
The low commercial viability of CSK awards (limited editions and high out of print rates) also supports that demand is the issue. This is a pretty important data point because it negates the causal claim of identity, i.e. they are books by AA authors, about AA experiences and therefore, based on the causal claim for identity (need to see themselves to want to read them) ought to be well received by AA audiences.
The unchanged level of POC books over twenty years despite the emergence of successful multicultural focused independent publishers such as Lee & Low is further evidence of a demand side problem. They obviously are effective publishers since they are successfully competing with the big six. However, if the market has not grown, that means they are taking market share from the big six, not growing the market by meeting pent up demand. Again, that undermines the associative identity argument. If these independent publishers are offering what is said to be needed but the market is not growing larger, then there is no unmet need. In other words, it's not a supply issue but a demand issue.
The data is neither extensive enough or robust enough to support firm conclusions, but right now, with what we know, it appears that the low POC presence is substantially a Demand Side issue. In Cheryl K.'s post, there is also a clear point that there remain supply side issues in terms of both raw number of submissions and the quality of submissions. Even if the primary issue is Demand Side, that doesn't mean that the Supply Side initiatives don't remain pertinent. However, absent some change in the environment or material change in the nature of the initiatives, there is no reason to believe future Supply Side results will be any different than past results, i.e. 15% POC content/presence.
If the low POC of 15% is substantially a Demand Side issue, what can be done and by whom? All the data points to this being rooted in cultural values and personal decisions which are notoriously difficult to budge. It can be done. Smoking is down from 45% in the early 1950s to 20% of the population today, but that was a material sustained national effort over many decades. Change in culture and individual decision making are hard. But there are some things that can be done.
We have already mentioned some. Be tactical, concrete and material.
. Be a bundler - Going to a book conference with publisher's present? Get around to as many libraries beforehand as possible, find out their acquisitions budget and the portion to be dedicated to diversity. Sales leads are catnip to publishers. They will respond. They will publish more of what they are selling.
. Start book clubs in the affected communities - Parent's clubs or children's clubs or parents with children's clubs. Our census based categories are useful for superficial monitoring but actually can stand in the way of local real knowledge. A community of third generation Cuban American middle class professionals with their children in private pre-school are technically Hispanic but their book club issues and readings are materially different from those of a ten year undocumented resident from a rural community in Mexico with a five year old starting kindergarten. Trying to provide an "Hispanic" book list to either misses the mark. Hard work to do this, but I suspect this might be by far the most powerful way to address POC disparities in publishing in the long run by creating very targeted grass roots demand. All children start out with the potential to be enthusiastic readers and that spark is suffocated in a thousand different ways, most of which are hard to identify and even harder to fix because they are so variable. Community based book clubs might serve as a lifeline for those wanting to cultivate an environment of enthusiastic reading for their children.
. Be concrete and link to institutional goals - Susan P. provided a great example. People/institutions respond to specific issues a lot better than to nebulous ones. You can make an effective rhetorical argument. You can make an effective empirical argument. Most effective all is to make a passionate rhetorical argument grounded in gimlet-eyed empirical evidence. Everyone is short of time, money and attention span. To Christine T-B's point, What Gets Measured Gets Done (As an aside, it is pretty critical that you understand what you are measuring, it is surprising how often that is done poorly or incompletely). I would expand on Susan's example even further to provide context (made up numbers follow): "Our student population is 35% Hispanic, 10% of our monthly circulations are Hispanic content but only 3% of our collection is Hispanic content. The average age of our Hispanic content books is 15 years versus 6 years for the collection overall. Students who check out books have test scores that are 25% higher than those who are not reading library books. Among circulating books, those that are less than five years old are circulated at three times the rate of those older than 10 years." The administration or board may or may not have the budget to address the issue but the claim can no longer be dismissed as someone's pet peeve or hobby horse. Goals need budgets, budgets need priorities, priorities need numbers.
I believe the book club suggestion would dovetail with Cheryl K.'s comments about the need to cultivate a larger and better pipeline of POC talent.
Charles
Date: Fri, 07 Feb 2014 14:41:10 -0500
Both business decisions and government policy decisions are plagued by two fallacies, post hoc ergo propter hoc and cum hoc ergo propter hoc, and I suspect we are falling prey to one or both. Post hoc is concluding that because one thing happened after another, then it must have been caused by it. Cum hoc is even more common and that is mistaking correlation for causation. Only 20% of corporate projects are delivered on time, on budget and achieve their stated goals. Only 10% of government policy implementations do so. These two fallacies are behind both those results. It is often hard to determine 1) causation, and 2) the directionality of causation. Advocates for more POC authors and content argue that low POC publication rates cause the low reading rates among POC communities and therefore low demand. Publishers, librarians, and retailers claim that there is low demand for POC content/authors and that is why there are fewer POC authors/content books. Chicken and egg. Supply Side Hypothesis (need more POC authors/content) or Demand Side Hypothesis (need more demand for such content). Of course, it is also possible that both are equally true. It is even more probable that both have some, but perhaps unequal, degrees of truth. But which and how much? What is the evidence for either hypothesis?
A couple of neutral examples of the prevalence of this causal directionality problem. I am head of a couple urban green space/park conservation groups. We have a standard list of arguments we use in our advocacy. Ten years ago, a common argument was that better and increased parks and green space in urban environments would help address health and obesity issues that are prevalent in such environments. It was a logical argument - more green space, more opportunity to exercise, better health, lower obesity rates. We do not use that argument these days because the direction of causation was wrong. In the past ten years there have been a couple of reasonably robust longitudinal studies that indicate that presence or absence of urban green space has no impact on community health and obesity. It's not that people couldn't exercise because there was no green space. There was little green space because people weren't interested in exercising. In our efforts to do good (and there are other more important reasons for urban green spaces, health was just one of the arguments used), we actually exacerbated things a bit. It was not the sedentary and obese (who happened to also be poor) who used the new green spaces but the already fit (who were usually well off) - creating an impression around classism. It is not enough to want to do good, you actually have to achieve it and that can be complicated. Because of the longitudinal studies, it can no longer be argued that health and obesity are likely to be improved by increased green space.
I just saw yesterday a similar example of mistaken causal directionality. Over the past five years there has been a five hundred million dollar effort to address food deserts (urban environments with reduced access to healthy food, particularly fresh fruit and vegetables). The argument has been that people have poor health and high obesity rates in these areas because there is poor access to fresh F&V. The government subsidized a number of pilot projects to bring fresh F&V to these communities and test the hypothesis. The results study is out for the first five years and it appears that there is no change in consumption and certainly no change in obesity rates and health outcomes. (New Neighborhood Grocery Store Increased Awareness Of Food Access But Did Not Alter Dietary Habits Or Obesity by Steven Cummins, et al. http://content.healthaffairs.org/content/33/2/283.abstract) If these results persist, as is expected that they will, then it will be another example of mistaken causation. Poor health and obesity were assumed to be caused by low availability of fresh F&V. However, it now appears that low availability of fresh F&V happened because there was low demand.
In terms of children's books, we have one observable fact and that is a low and very constant level of reading books by and about minorities, approximately 15% over the past twenty years per the CCBC numbers, taking those numbers at face value (3,200 mass market trade books as part of the 25-35,000 new children's titles released each year, i.e. are the figures for mass market trade books reflective of the total population?; and performing the ratio calculations on the correct base, i.e. omitting animal based books, etc.).
There is an argument to be made that even that low level is explicable based on reading demographics rather than population demographics. Different groups in the US have different reading proclivities (amount of time spent reading overall, amount of time spent reading books, amount of money spent on books). According to Pew, ( http://libraries.pewinternet.org/2012/12/27/e-book-reading-jumps-print-book- reading-declines/), only 75% of the population reads any book at all in a year. If you look at only those who are electively reading (i.e. not those who read a book for work or study), the number drops to 50%. So readers are a self-selected group whose demographics are likely to be different than those of the population at large. And they are different. In that Pew study, a higher percentage of the white population read a book in a year
(78%) than do African Americans (74%) or Hispanics (60%). Even more critically, white readers read 33% more books than either of the other two groups. The implication is that while whites might be 70% of the population
(depending on definitions of ethnicity and race), they are 80% of the readers. This is consistent with two other sources of data: For book reading, the ethnic distribution based on Bureau of Labor Statistics (BLS) time-use data for average time spent reading per year by ethnic group and Kaiser Foundation data for reading form preference (books vs. newspapers and magazines) by ethnic group is 82% white, 4% African American, 6% Asian American, 8% Hispanic, and 0.4% Native American. The distribution in terms of money spent on reading (BLS Consumer Expenditures 2011) is 85% white, 6% African American, 4% Asian, and 5% Hispanic (Native American numbers not broken out in the BLS expenditure data).
The net of this is that the available data indicates that the race distribution of the CCBC numbers (85% white and 15% POC), while inconsistent with at-large population demographics, are quite close to the numbers for those who are reading books (80% based on Pew numbers, 82% based on BLS time-use, and 85% based on BLS consumer expenditures). This makes logical sense. But in terms of rectifying the disparity, it points away from publishers and towards the reading habits of the at-large population. If we want to bring the reading demographics into alignment with the population demographics, then the different populations need to be consuming books at comparable rates.
But setting all that aside, we still have a chicken and egg conundrum. Several posts have made the logical inference that minority reading figures are low because they can't find minority authors/content to read. If POC authors were published more, then POC would read more. In other words, that there is a supply side issue. (BTW, I am uncomfortable with the blanket POC term because the reading numbers vary significantly between census groups, but use the term for simplicity). But what if the causal direction is the opposite and there is a demand side problem instead. This is important to resolve because the solutions to supply side problems are quite different from solutions to demand side problems.
The supply side hypothesis has been the dominant assumption since 1965 with Nancy Larrick's article, The All-What World of Children's Books. In the nearly five decades since then, there has been persistent advocacy to increase minority presence in books (both authorial and protagonist). Most/all the major publishing houses have had internal initiatives towards that end. Most/all of them have some form of multicultural imprint. SCBWI and ALA have had (and still have) various initiatives. Minority oriented prizes such as CSK and Pura Belpre have emerged to increase attention and market interest and to recognize new talent. Substantial independents such as Lee & Low have been founded and grown with a target focus on minority publication. Black History month provides a means of focusing on minority oriented reading in bookstores and libraries. All of these activities are intended to bring greater attention and supply to POC oriented reading. Despite all these efforts to increase supply, as the CCBC numbers show, and as illustrated by the Lee & Low graphic (once adjusted to the correct base), and as Cheryl Klein highlights
(http://chavelaque.blogspot.com/2014/02/in-defense-of-cbc-diversity-and-on.h tml), the percentage of minority presence remains ~15%.
Maybe the problem is not a supply side issue but a demand side issue. What is the data for that hypothesis?
The BLS , Pew and Kaiser Foundation data all document that there is indeed a material component that is demand side. If different groups spend less time reading, read fewer books and spend less money on reading, then that is explicitly a demand side problem. Providing more authors is unlikely to make a material difference unless their volume of reading increases.
The low commercial viability of CSK awards (limited editions and high out of print rates) also supports that demand is the issue. This is a pretty important data point because it negates the causal claim of identity, i.e. they are books by AA authors, about AA experiences and therefore, based on the causal claim for identity (need to see themselves to want to read them) ought to be well received by AA audiences.
The unchanged level of POC books over twenty years despite the emergence of successful multicultural focused independent publishers such as Lee & Low is further evidence of a demand side problem. They obviously are effective publishers since they are successfully competing with the big six. However, if the market has not grown, that means they are taking market share from the big six, not growing the market by meeting pent up demand. Again, that undermines the associative identity argument. If these independent publishers are offering what is said to be needed but the market is not growing larger, then there is no unmet need. In other words, it's not a supply issue but a demand issue.
The data is neither extensive enough or robust enough to support firm conclusions, but right now, with what we know, it appears that the low POC presence is substantially a Demand Side issue. In Cheryl K.'s post, there is also a clear point that there remain supply side issues in terms of both raw number of submissions and the quality of submissions. Even if the primary issue is Demand Side, that doesn't mean that the Supply Side initiatives don't remain pertinent. However, absent some change in the environment or material change in the nature of the initiatives, there is no reason to believe future Supply Side results will be any different than past results, i.e. 15% POC content/presence.
If the low POC of 15% is substantially a Demand Side issue, what can be done and by whom? All the data points to this being rooted in cultural values and personal decisions which are notoriously difficult to budge. It can be done. Smoking is down from 45% in the early 1950s to 20% of the population today, but that was a material sustained national effort over many decades. Change in culture and individual decision making are hard. But there are some things that can be done.
We have already mentioned some. Be tactical, concrete and material.
. Be a bundler - Going to a book conference with publisher's present? Get around to as many libraries beforehand as possible, find out their acquisitions budget and the portion to be dedicated to diversity. Sales leads are catnip to publishers. They will respond. They will publish more of what they are selling.
. Start book clubs in the affected communities - Parent's clubs or children's clubs or parents with children's clubs. Our census based categories are useful for superficial monitoring but actually can stand in the way of local real knowledge. A community of third generation Cuban American middle class professionals with their children in private pre-school are technically Hispanic but their book club issues and readings are materially different from those of a ten year undocumented resident from a rural community in Mexico with a five year old starting kindergarten. Trying to provide an "Hispanic" book list to either misses the mark. Hard work to do this, but I suspect this might be by far the most powerful way to address POC disparities in publishing in the long run by creating very targeted grass roots demand. All children start out with the potential to be enthusiastic readers and that spark is suffocated in a thousand different ways, most of which are hard to identify and even harder to fix because they are so variable. Community based book clubs might serve as a lifeline for those wanting to cultivate an environment of enthusiastic reading for their children.
. Be concrete and link to institutional goals - Susan P. provided a great example. People/institutions respond to specific issues a lot better than to nebulous ones. You can make an effective rhetorical argument. You can make an effective empirical argument. Most effective all is to make a passionate rhetorical argument grounded in gimlet-eyed empirical evidence. Everyone is short of time, money and attention span. To Christine T-B's point, What Gets Measured Gets Done (As an aside, it is pretty critical that you understand what you are measuring, it is surprising how often that is done poorly or incompletely). I would expand on Susan's example even further to provide context (made up numbers follow): "Our student population is 35% Hispanic, 10% of our monthly circulations are Hispanic content but only 3% of our collection is Hispanic content. The average age of our Hispanic content books is 15 years versus 6 years for the collection overall. Students who check out books have test scores that are 25% higher than those who are not reading library books. Among circulating books, those that are less than five years old are circulated at three times the rate of those older than 10 years." The administration or board may or may not have the budget to address the issue but the claim can no longer be dismissed as someone's pet peeve or hobby horse. Goals need budgets, budgets need priorities, priorities need numbers.
I believe the book club suggestion would dovetail with Cheryl K.'s comments about the need to cultivate a larger and better pipeline of POC talent.
Charles
---
You are currently subscribed to ccbc-net as: ccbc-archive_at_post.education.wisc.edu.
To post to the list, send message to: ccbc-net_at_lists.wisc.edu
To receive messages in digest format, send a message to...
ccbc-net-request_at_lists.wisc.edu
...and include only this command in the body of the message:
set ccbc-net digest
CCBC-Net Archives
The CCBC-Net archives are available to all CCBC-Net listserv members. The archives are organized by month and year. A list of discussion topics (including month/year) is available at http://www.education.wisc.edu/ccbc/ccbcnet/archives.asp
To access the archives, go to:
http://ccbc.education.wisc.edu/ccbc-net
and enter the following:
username: ccbc-net
password: Look4Posts
Received on Fri 07 Feb 2014 01:42:21 PM CST
