The Ladder

A Blog from New America's Asset Building Program

Publication Image
Image provided by Ross Belmont via Flickr

The link between household wealth and youth’s educational outcomes has been confirmed by numerous research studies, finding that youth tend to do better educationally when their households have more wealth (click here and here for more information). A new paper from the Assets and Education Initiative at the University of Kansas School of Social Welfare confirms the relationship between household wealth and math achievement for youth from the U.S. and Ghana. The paper adds a unique contribution to existing research.

A unique contribution from this study is the inclusion of households’ wealth accumulation. The difference here is that accumulating wealth may have effects on math achievement beyond owning wealth. For the most part, previous research studies have looked at households’ wealth ownership and math achievement. That is, whether youth have better math achievement when they are from high wealth households. In this case, wealth is static—a stock of resources owned by households. Along these lines, researchers ask whether accumulating household wealth relates to math achievement among youth from U.S. and Ghanaian households.

Study Design                         

Researchers used longitudinal data from two sources: youth from U.S. households that participated in the Panel Study of Income Dynamics and youth from Ghanaian households that participated in the Ghana YouthSave experiment. Among U.S. households, there was a full sample of 1,097 youth with an average age of 15 from the Panel Study of Income Dynamics. In addition to the full sample, researchers looked at math achievement separately for black youth (n = 520) and low-to-moderate income youth (n = 459). Among Ghanaian households, there was a full sample of 2,558 youth with an average age of 15.

Net worth was the measure of wealth for U.S. households, which summed assets and then subtracted liabilities. Asset index was the measure of wealth for Ghanaian households, which included properties, household possessions, and livestock owned by households.Additional controls included youth’s and heads of households’ demographic characteristics, home and family environment characteristics, and household income.

In terms of the study design, this paper makes methodological contributions because it uses two analytic techniques of potential interest to researchers. The first is the Inverse Hyperbolic Sine (IHS) transformation, which may be useful for retaining zero and negative values typically found in household wealth data and retaining the continuous form of the variable. Common transformations like the natural log set zero and negative values to 1, meaning a household with net worth of −$10,000 would be equivalent to having net worth of $1. The second are splines, which can be used to test non-linear relationships and accumulation thresholds between wealth and math achievement. For example, accumulating negative net worth might negatively relate to math achievement whereas accumulating positive net worth might positively relate to math achievement. These analytic techniques are discussed in greater detail in the paper, but are noteworthy for a research audience.

What Researchers Found

  • In the full sample of youth from U.S. households, net worth significantly improves math achievement.
    • Accumulating zero and negative net worth is significantly related to decreases in youth’s math achievement whereas accumulating moderate values of positive net worth is significantly related to increases in youth’s math achievement.

    • Among black households in the U.S., IHS transformed net worth significantly predicts youth’s math achievement at trend level.

    • For black households, owning net worth sufficient to remain above the poverty line for three months significantly improves their math achievement at trend level.

    • Among low-to-moderate income households, IHS transformed net worth significantly predicts youth’s math achievement.

    • For low-to-moderate income households, owning net worth sufficient to weather a 25% drop in income and to remain above the poverty line for three months significantly improved math achievement.

  • In Ghanaian households, the IHS transformation of the asset index only becomes significant when splines are introduced.
    • Accumulating assets between the 25th and 50th percentiles is related to a significant increase in youth’s math achievement.

    • There is a negative relationship between the asset index and math achievement when households start accumulating assets from zero assets and accumulating small amounts.
  • IHS transformed wealth performs similarly compared to commonly used log and categorically transformed wealth within U.S. and Ghanaian households.

  • Non-linear relationships and accumulation thresholds emerge that significantly predict youth’s math achievement when IHS transformed wealth is combined with splines.

What Does this Mean?

A main point from these findings is that household wealth is related to math achievement for youth in the U.S. and Ghana. Moreover, these youth have better outcomes when their households are accumulating wealth. It may not be sufficient for households to retain wealth they already own; households may also need to accumulate wealth in order to improve youth’s math achievement. These findings may be informative for policies that aim to build households’ wealth.

This also means that researchers can explore non-linear relationships and accumulating thresholds for household wealth when the IHS transformation is combined with splines. The authors recommend that researchers consider the advantages of the IHS transformation for wealth data in their own analyses, especially when zero and negative values are present.

For more details, please read the complete paper available on the Assets and Education Initiative’s website.

Join the Conversation

Please log in below through Disqus, Twitter or Facebook to participate in the conversation. Your email address, which is required for a Disqus account, will not be publicly displayed. If you sign in with Twitter or Facebook, you have the option of publishing your comments in those streams as well.