What Textbooks Teach Us — And What They Don’t

Texas and California often appear to be worlds apart when it comes to politics and culture, but the education students are getting — as far as their textbooks go, at least — may not be so different.

University of Chicago Assistant Professor Anjali Adukia, Ed.M.’03, Ed.M.’12, Ed.D.’14, investigated more than 260 textbooks used in both public and religiously affiliated schools in the two states, analyzing their portrayal of race, gender, religion, and historical events. “I think the part that was the most surprising to me is despite this narrative of political polarization, we actually don't necessarily see that in the books themselves that are given to kids on average,” Adukia says.

While there are differences, especially regarding religious content, textbooks used in both states tend to emphasize similar themes such as family, nature, and history, she says. Additionally, the textbooks feature similar portrayals of females in passive and stereotypical roles, while males are more often linked to power, politics, and military.

Textbooks play a crucial role in shaping students' identities and worldviews, transmitting cultural values and societal norms, she argues. Despite changing public attitudes, these textbooks remain largely unchanged, posing important questions about how educational content influences future generations and the values that schools are endorsing.

“The process of education and its associated books and curriculum materials necessarily, and by design, transmit the knowledge that we care about, the values that we care about. They transmit messages about who belongs in what spaces in society,” Adukia says. “Also, the presence and the absence of different identities can send messages to kids which can contribute to how they view their own potential and the potential of others.”

In this episode, we explore the similarities and differences across textbooks in public and religious schools, and the role textbooks play in shaping students’ identities and worldviews.

Transcript

JILL ANDERSON: I’m Jill Anderson. This is the Harvard EdCast.

What are students really learning about America and does it change depending on where or how they're educated? University of Chicago assistant professor Anjali Adukia has been exploring these questions by diving deep into the textbooks used across public schools, religious private schools, and home schools.

Her latest research analyzed more than 260 textbooks from Texas and California, spanning from the 1980s to today. What she found challenges our assumptions about ideological divides in education and reveals surprising similarities across schools. I wanted to know more about how textbooks shape students' understanding of history and identity and why it matters. First, I asked what led her to study textbooks between public and religious private schools.

ANJALI ADUKIA: I'm really interested in understanding what shapes our collective memory and our understanding of the world. How do we get our beliefs and how do those then contribute to what later life decisions we make or what opportunities we afford ourselves or others? And we learn from popular culture like songs and movies, and we learn from stories, narratives that are told. There's monuments and tributes in the history that they represent. The process of education and its associated books and curriculum materials necessarily, and by design, transmit the knowledge that we care about, the values that we care about. They transmit messages about who belongs in what spaces in society.

Also, the presence and the absence of different identities can send messages to kids which can contribute how they view their own potential and the potential of others, which can then shape subconscious defaults. And so just even at a first stage or as first step, what is it that kids are learning? What are the knowledge and values? Who is being highlighted? And then also importantly, how are they being portrayed? Because you can think about representation both as whether or not there are topics or identities. And so you can think of that as a numerical accounting essentially, but you could have a book full of underrepresented identities. But if they're being portrayed in stereotypical or reductive ways, then that might be just counter to what your goals are. We also care about the depiction and how different identities and topics are being talked about.

JILL ANDERSON: So you looked at over 200 textbooks from 1980 to today, and my understanding is this was third and fifth grade textbooks.

ANJALI ADUKIA: Correct.

JILL ANDERSON: Were there any clear trends you saw over time in terms of gender, race, and religious representation?

ANJALI ADUKIA: Are looking at public school textbooks from Texas and California that are written for third and fifth graders for science, social studies and reading or language arts. We're also looking at a set of textbooks that are published by conservative Christian publishers that are commonly used in homeschooling settings. And they're also often used in private religious school settings where increasingly, there are more tax dollars that are being directed towards schools that use those textbooks.

And actually, what we were really surprised about, when you think about just even Texas and California, often Texas is considered to be on more on the conservative end of the spectrum, and California is often considered to be more on the progressive end of the spectrum. And so when you watch the news media, when you listen to politicians, you kind of think of them at different ends of what it is that they might be teaching kids, right? In California, there has been the Fair Act, which is supposed to be all about greater inclusion. In other places, there have been more educational gag orders or bans on certain topics.

JILL ANDERSON: Right.

ANJALI ADUKIA: And yet what we see is that Texas and California on average are actually pretty similar across the board. However it is that we look at this, there are some important differences with these textbooks that are published by these conservative Christian publishers, but there's some similarities as well. And so I think the part that was the most surprising to me is despite this narrative of political polarization, we actually don't necessarily see that in the books themselves that are given to kids on average.

JILL ANDERSON: That's so, so interesting. We'll get more into the specifics of what some of those differences were. But I want to start first looking at what differences you saw between the religious private school books and the public school books.

ANJALI ADUKIA: Actually, in terms of differences, you could start with the topics themselves. We do see across all the subjects, the religious private school textbooks are more likely to include mentions of religious topics and religious values across their pages, even in science. But when you look at the top topics, cross-reading, science, social studies, generally, they're actually pretty similar. So it's about education or family or nature and reading and science, it's around observation and whether it's astronomy or biology and in social studies, it's around geography and exploration and history.

Now, in the religious books, the religious collection, as we call them, again, they're not necessarily religious, they're science books or social studies or reading. There is actually among the top topics in reading actually topics such as fear and survival and perseverance, which you don't actually see in the California and Texas collections. You also do see that religion tends to be a top topic. And in terms of values, faith is one of the top values that is espoused. And Christianity is the most common religion actually across the board, but it's especially in the religious textbooks. Especially when you think about contentious topics. Topics such as slavery and the Civil War on average are actually talked about pretty similarly.

Now, when you look at religion, this is where there might be a difference. When religion or religious topics come up in California and Texas, they tend to be connected to the founding of our country. So you think about Baptists and Quakers, whereas in the religious collection or this religiously published textbooks, it's much more connected to God and Jesus and sin and prayer. And so it is more specific in that front. Some people think that especially in places like California, they're much more likely to talk about issues or topics related to LGBTQI+ individuals or their rights. And actually across these 261 textbooks, we found a total of two pages that actually were identified by our models as actually discussing or at least mentioning these topics.

It was once in a California textbook, which was all around California heroes and they talked about Harvey Milk and Chaz Bono, about how they worked hard to protect rights for people or fighting against discrimination. And in the religiously published textbooks, there was one page which talked about how God designed marriage to be between a man and a woman, and that gay marriage would harm society and could threaten religious liberties. Actually, I think another place where people sometimes think there are big differences is how the founding of the United States might be talked about.

Sometimes there are folks who think that, especially among progressives, that people are really wanting to teach kids, quote unquote, to "hate this country". And so we looked at, for example, the founding fathers and how are the founding fathers spoken about? How often are they talked about, how are they spoken about, or at least in which contexts are they spoken about? And actually across the board, they're all spoken about, whether it's in California, in Texas, or in these religiously published books, they're more likely to be spoken about in a positive light. So we look at the valence, which is kind of about how pleasant or how positive the context in which something is or whatever group you're looking at is talked about.

And they're also more likely to be spoken about in what they call a dominant context, so a more powerful context compared to a weaker context. And so it's the same across these collections. This notion that there are these real differences, it's not something that's actually born out in our analysis. And actually, across each of these collections, a vast majority, anywhere from 80 to 90% of the famous individuals who are mentioned are white. And then in California and Texas, you go down to about 10 or 11% of the famous individuals who are mentioned are Black. 3% of the famous individuals in the religious collection are Black. And then from there, it's much, much smaller.

And so it's not that you see these really big differences in terms of even by race, who is being included. You could also look at the skin color or the pictured characters, and especially at younger ages, now third and fifth graders can read, but even then, just the images themselves send these initial messages. And so when we think about what are kids seeing, skin color is one of those first things, even from infants, from the ages of two months, three months, that's one of the first ways they distinguish people. And actually you do see that California and Texas are more likely to picture characters with slightly darker skin. And actually California and Texas, there's no significant difference between the skin color of the pictured characters on average. Whereas in the religious collection, the characters are more likely to have lighter skin.

JILL ANDERSON: What was interesting to me is despite these differences that we all hear about politically, that between these states, there was a lot of similarities in these books regarding identity and representation, some of which you just talk about. One of the most interesting ones was that public and religious textbooks, regardless of where it was, tend to fall short in female representation.

ANJALI ADUKIA: So if we think about gender, and you want to think about the presence of gender representation versus the portrayal of gender representation. So Texas and California, at least overall in the text, are closer to 50%. They're not quite at 50%, but they're closer, whereas the religious collection is not, it's much less in terms of presence. 

Now you could say, well, when I was growing up, the gender-neutral term was he. And so maybe that's actually what you're picking up is you're just actually getting a lot of neutrality. And frankly, if you wanted to even look at subjects, even though yes, were there women who did substantial things in history and who were scientists? Absolutely, but they're not typically represented in science or social studies. But in reading, you can make more of these choices. And so actually when you look by subject, you actually see that California and Texas, in terms of presence, just the proportion of gendered words that are female versus male, it's [inaudible 00:11:45] 50% actually for reading. And then it's much less for science and social studies and then the religious collection for all of them, it's actually less.

Again, what was surprising to me is when you look at how women or females are talked about in the text relative to males. So if you just look at the top words that are associated with females versus males, for females it's words like husband or words related to appearance or clothing or sewing. Whereas the top words associated with males are really around military. So king or admiral or master sword or horses. My goodness, there is something about when people talk about boys or males, they love to talk about horses. And this is across California, Texas, the religious collection.

And then when you think about the measures of, as I mentioned, valence, which is how positive or negative a particular group is talked about or what context they're talked about, there's something called arousal, which is how active or passive a group might be talked about. And then dominance, how powerful or weak, the context in which they're talked about. And while there's slightly different magnitudes, the direction is actually very similar. So females across each of the collections tend to be spoken about in a more positive context, but they tend to be spoken about in a more passive and a weaker context. But again, that's across the board for all three collections.

And just one last thing actually, in terms of where we see gender skews, we do see that females are just much more likely to be connected to domains related to family or the home. And males are more likely to be connected to domains related to politics, for example.

JILL ANDERSON: You already talked about the LGBTQIA representation and how it was fairly minimal, almost non-existent at this point, except for small amount in California's textbooks. 

ANJALI ADUKIA: Which you would not know by watching the news, given how much attention is put on this. And it's like, "What are we teaching our young kids?" And it's actually not much in terms of the curriculum itself.

JILL ANDERSON: Yeah, it's surprising to talk to you in a way because there are things that one might expect. You would expect religious textbooks to talk about creationism and faith and things like that. So those don't seem all that surprising.

ANJALI ADUKIA: That's right. And frankly, there's actually very little mention of evolution. There is some, but not as much as you might think in the California and Texas books. Evolution is more likely to be mentioned or the topic is more likely to come up in these religiously published books. But when it comes up, it is almost always talked about in the context of creationism as well. So the two topics actually come up more simultaneously in that religiously published collection compared to Texas and California.

JILL ANDERSON: What do you think are some of the implications of what you've discovered, which is that there's some question how females are represented and even how males are discussed and some absences there that are fairly strong across the board?

ANJALI ADUKIA: So I think of curriculum in three different parts. There's curriculum is written. That's these textbooks or whatever, the materials that are given to kids. There's curriculum is implemented, and that's really how the teacher decides to talk about a topic. And then there's curriculum is discussed. So how is it that the students themselves are talking about something? And these patterns are relatively flat over time. Just take female representation. But if you actually map on top of that survey results of, do you think a woman can be president? Do you think a woman should work out of the home and whatnot, you do see that public belief in these different settings is actually changing, but we don't actually see these changes in curriculum.

So I think what this really does is it begs the question of, well, what is contributing to differences in beliefs and what is contributing to this narrative of polarization and how much of it is a narrative versus reality? And so I think this just opens up this set of questions. And there are all these fights around content and standards and public schools. And I think it's important for us to understand where are taxpayer dollars going? What are the values that we're leaving the next generation with? Because then that will also affect the generations after that.

And it is important to understand what is in these textbooks because in some ways, whatever we put in these textbooks, we are then putting in our schools across an entire state, across every single grade, a whole cohorts of kids are seeing those messages. And so whether these are implicit or explicit messages, they are a signal of what we value as a society. And these are sanctioned by the schools to say, "Yes, this is knowledge that we think you should know." But I think it is this empirical question as to, well, how much does it matter what's in these books? I think it must matter and I think that it just opens up a whole set of questions that we really just need to pursue in future research.

JILL ANDERSON: Some of your research also investigates representation in children's books. Do you see similarities across identity representation, how people are portrayed in children's books to textbooks?

ANJALI ADUKIA: That's right. So in this other work that you're mentioning, we look at award-winning children's books over the last century. And so there are these awards called the Newbery and Caldecott Awards, and they're kind of like the Oscars of children's book awards or children's books. And they're very commonly present in homes and schools and libraries. They are much more likely to get checked out. They're much more likely to get purchased. And so when we actually look what is the representation in these books, we actually see less female representation than in California and Texas. 

In California and Texas, especially in the reading books, so if you just take the reading books, that is around 50%. On average, if you actually take reading science and social studies, it's a little less than 50% overall. But there is more female representation in California and Texas textbooks than there are in these books that we're commonly giving to kids for pleasure or for whatever other value that they're getting out of just whatever books they are assigned in school or in the libraries, what people are more likely to purchase for either their own kids or other kids. And we see that there is actually probably a little bit more representation in terms of race, but even then, it doesn't seem like it's that much more.

JILL ANDERSON: As you mentioned, there's a lot of questions that need to be answered going forward about all of this work and the books that children have access to. And you also noted about how there are several components that go into teaching in a classroom, and only one of those is the books that get used. But knowing what you've uncovered in your research, even across the board with children books, what do you think that means? Does this mean there needs to be changes made in the books, or we need to approach representation differently in the classroom?

ANJALI ADUKIA: When we first started doing this project, one of the things that we really wanted to do was how is it that the people who are actually choosing the textbooks, if it's a 700-page textbook or if there are books that you need to get across all subjects for all students, how do you really get a good picture of what are the topics and the values and the identities and all the different things that we care about? How do you get a good sense of what is in all of these books?

Now, you can have humans go through and read all of them, but of course what's going to happen there is that first of all, humans get tired. Who you are on page seven is different than who you are on page 70 and different who you are on page 777. And so then you need to have a number of humans go through and do this to try to figure out even just what are all of this content that we're giving kids. And then because you want to be a social architect of like, "Oh, I want to weigh a bunch of different perspectives so that kids are getting that."

And so we thought, well, how is it we can leverage a lot of these new computational social science or artificial intelligence tools to help raise the floor of our understanding? And maybe that'll actually raise the ceiling as well. So we thought, oh, if we can make these tools available, a third grade teacher can run their curriculum through and make those decisions. But the fact is that a third grade teacher doesn't have digitized books. They don't have the time to be able to run everything through each time. And why actually have it be at the individual level? What if it's something that publishers themselves could use tools like this?

And again, it's not a scorecard, it's just awareness, right? It's just some sort of awareness rubric for every single book or every single set of materials. Maybe you really want to have a bunch of books that like, oh, well, they're talking about soccer, they're talking about football, and these kids, this set of books has kids who are depicted with glasses or wheelchairs or what. And oh, I want to make sure that there's some combination. But if that awareness or that information can be given, then people can make those kinds of decisions. And I think part of it right now is that people just don't know.

And sometimes people think, "Oh, representation, that means you want to erase whoever was in the dominant group or erase history as we know it." No, it simply means that we just need to have an understanding of what we're teaching kids. And I think there are these philosophical questions of, do you teach things in a quote unquote "all positive light", or do we want to teach kids nuance? Or do we want to teach kids about themselves and understand that even adults can make mistakes? But actually there's a lot of amazing work that's happened, whether it's in our history or contemporaneously across groups, within groups. And so I think part of this is just even figuring out, how does content get chosen and giving educators tools to be able to make intentional decisions.

JILL ANDERSON: Well, Anjali, this is amazing. Thank you so much. Lots and lots to think about.

ANJALI ADUKIA: Well, thank you so much for having me. It's been a pleasure talking with you.

JILL ANDERSON: Anjali Adukia is an assistant professor at the University of Chicago and the director of the MiiE Lab. I'm Jill Anderson. This is the Harvard EdCast, produced by the Harvard Graduate School of Education. Thanks for listening.