Widgets Magazine

Income gap creates achievement gap, study says

Correction: The first sentence of the online version of this article originally did not match the more up-to-date version printed in The Daily Feb. 22, 2012. The article has now been updated to reflect this change. 


In the past 50 years, socioeconomic status has become an increasingly important predictor of academic achievement on standardized tests, according to a recently published study by Stanford associate professor of education Sean Reardon.


University officials, however, said the data that would be necessary to determine whether a similar achievement gap exists between students from low- and high-income families at Stanford is dispersed between different administrative offices, making it difficult to track.

Koren Bakkegard, associate dean of Undergraduate Advising and Research (UAR), told The Daily she does not know of a centralized organization in the University that analyzes both academic achievement and socioeconomic status.


“It may very well be that no office at the University has or combines the data that would be necessary to generate those kinds of records,” Bakkegard said.


“In UAR, for example, we do not have information about students’ socioeconomic status,” Bakkegard continued. “To my knowledge, only the Financial Aid Office would have this information, and I would be surprised if they tracked students’ academic performance since that is not their function.”


Tommy Lee Woon, director of diversity and first-generation programs, also could not speak to whether or not an achievement gap exists between students of varying income levels due to the way data is kept by the University.


“I unfortunately do not have the answers,” Woon said.


“Data is very decentralized here,” he added. “Some of the data may be embedded in reports, but tracking it down can be a challenge.”


Reardon said that although he has no statistical evidence to back up his claim, he is doubtful that such an achievement gap exists at Stanford, as judged by his research methodology, because of Stanford’s highly selective admissions process.


The data that Reardon used for his study came from several different nationwide standardized tests, including data collected by the National Center for Educational Statistics and the National Assessment of Educational Progress.


“One of the criteria I used to decide which studies to use was that I didn’t want them to be local,” Reardon said.


The results of his study showed that in the late ‘60s and early ‘70s, family income, rather than race, determined how students performed on standardized tests. The gap between children in the 90th percentile and children in the 10th percentile of income was roughly twice as large as the difference in achievement between white and black children.


In addition, the study showed that the achievement gap between children of high- and low-income families was about 30 or 40 percent larger for children born in 2001 as compared to children born 25 years earlier.


“Because a place like Stanford is so selective, most students would score very high on these tests,” Reardon said.


Stanford students who come from low-income families would have scored high on standardized tests, and therefore their performance would not be remarkably different from that of students from middle class or upper class backgrounds.


He added, however, that “the way [the achievement gap] is reflected is that there are fewer students with low-income backgrounds at a place like Stanford than kids from a higher-income background.”


The achievement gap that Reardon saw in his research manifests itself in the fact that lower rates of low-income students attend college at all, let alone elite universities such as Stanford.


With its need-blind admissions process and extensive financial aid program, Stanford aims to attract as many qualified low-income students as it can. However, Reardon said that most students from low-income backgrounds never get the chance to succeed at a high enough level to be considered by universities.


“Kids with lower income backgrounds do not get the opportunities to succeed academically and don’t often get to go to places like Stanford,” Reardon said.

  • This article has some errors. For example, the first sentence says socioeconomic status is the primary factor but even the abstract of Reardon’s paper (book chapter) reveals that parental education is a stronger factor.

    Interested readers can find the report here: 
    Personally, I’m confused by the focus on comparing income to race. Race still has a strong effect, and income and race are still correlated. Instead of creating this bizarre “income is more important than race” narrative, the article could have just pointed out that the effect of income on education has been growing over time. That is an important and alarming fact which stands by itself.

    Also, I’ve been here for less than a year and I’ve already developed a gag reflex for Stanford’s consuming self-fascination. More than half the article was *wasted* on a completely inconclusive (and pointless imo) effort to determine if the conclusions of the report also hold at Stanford… Meanwhile, orders of magnitude more students in this same geographic area are *definitely* affected by this issue and are facing insane tuition hikes, cuts to their school budgets, etc.

    (As usual with the Daily missing the real story: poor journalism, but good preparation for future careers in the mainstream media)

  • YuhGurl

    So I kind of feel like holding the entire Daily or Estanfor in general in contempt is a bit much. 

  • MH

     If race alone had a strong effect, the data would show it. For a long time people considered race to be the most important factor in the achievement gap, but once you account for income, race becomes mostly insignificant. Many, many studies have shown this. Income is, has been, and always will be more important than factors like race. After all, wealth issues and classism have been around even longer than racism.

    And this article focuses more on Stanford’s work in the area because that’s what university newspapers do: focus on the work and impact of the university itself. Stanford doesn’t have any more self-fascination than any other institution, whether it’s a newspaper organization or a university.

    Articles like this one are useful partly because they’re raising awareness about this issue on Stanford’s campus. One of Stanford’s biggest cultural problems is a lack of awareness about socioeconomic issues, hence why the first director of low-income/first-generation programs was appointed this year. When I was a freshman a few years ago, this issue was never discussed, so even if we don’t know about the achievement gap specifically at Stanford, we can at least begin discussion about it and figure out a good way of assessing that gap here. In other words, you’re holding a university newspaper to an unreasonable standard.

  • Being a factor (significant variable in a statistical model) can have different meanings or interpretations and we can draw more than one conclusion about it. For another standardized testing example, the gender gaps do NOT necessarily mean that boys are “naturally” better at math or girls somehow predisposed to literacy. There can be some other variable, correlated with gender or race, that offers a better causative explanation. You probably already know this, since the point here is that income is such a causative variable.

    But it’s important to not ignore one factor just because another appears to account for more of the variability. Even if we could attribute 100% of the testing gap to income, that wouldn’t change the outcome: people of some races would still have lower test scores on average than others. And the fact would remain that income is, tragically, still strongly correlated with race.

    I don’t think that race (or gender) are good scientific/causative explanatory variables for test scores (especially because race is not even well-defined- see http://www.aaanet.org/stmts/racepp.htm for example). But this isn’t just some objective scientific finding isolated from social context. Some well-meaning people want to interpret this as proof that race does not affect academic ability (to me, that should not even have been in question). But others want to interpret this as proof that we are now a post-racial society, which is wrong on many levels. That is why I am taking issue over the wording.

    Also, I agree that asking whether or not the income effect exists at Stanford is a perfectly legitimate question. But there are so many other questions which are, in my opinion, both more interesting and more important. For example, the article doesn’t even mention how large the gap is. It only compares it to other gaps, or the same one at a previous time. The reader is left with no actual quantified sense of how important income is for test scores.

    Raising awareness is (by definition) an admirable goal, and an especially relevant one inside this bubble. That’s exactly why I’m confused by the tendency to reflexively look inside the bubble.