It looks like a bill to increase Kentucky’s minimum high school dropout age to 18 will enter the legislative process again this year, so I decided to update a little study I did last year.

For starters, there is already research that questions the effectiveness of the Age 18 idea.

When Massachusetts looked at similar legislation in 2009, the Rennie Center published a study titled, “Raise the Age, Lower the Dropout Rate?”

This 18-page report does a very nice job of outlining the pros and cons along with a survey of the available research. The Rennie Center says:

“Our review of research revealed little evidence to support the idea that raising the compulsory age to 18 decreases dropout rates and increases graduation rates.”

Later, the report says:

“We urge policymakers to first consider other policies to address the Commonwealth’s dropout crisis.”

One research example cited in the Rennie report compares the year states moved to Age 18 dropout minimums and the states’ performance with the Averaged Freshman Graduation Rate statistic that is now reported by the National Center for Education Statistics. The data only ran to the 2004-05 school year, so last year I took a look at the then available AFGR data to 2007. I found out that only five states of 15 jurisdictions (included: 14 states and Washington, DC) had graduation rate trends that notably exceeded the national average growth.

It is now a year later, so I have been able to update the original study with the 2007-08 graduation rate data. This new graph shows what I found.

As you can see, now only four of the Age 18 states with decent trend lines – Hawaii, Oregon, Wisconsin, and Connecticut – produced graduation rates notably above the national average. That is a further decay from the performance last year when five states performed notably above the national average.

Seven jurisdictions on the graph actually saw a decline in their graduation rates over time while the rest of the nation was posting improvement (*Note: Nebraska changed to Age 18 in 2004, but had five years of data to determine a useful trend line and therefore is also included).

To reiterate, the latest available data shows only four of the 15 Age 18 Dropout Minimum jurisdictions in the graph had graduation rate improvements notably above the overall national average improvement from 2001 to 2008.

This graph and the more extensive comments in the Rennie Center study indicate that simply raising the dropout age to 18 isn’t a sure cure for low high school graduation rates.

Technical Information

Averaged Freshman Graduation Rates (AFGR) for each state were obtained from several years of the Digest of Education Statistics and other documents from the National Center for Education Statistics (NCES). The AFGR was extensively researched by the NCES in the early part of the decade and the findings were made available in a two-volume report, “User’s Guide to Computing High School Graduation Rates,” in 2006. Per NCES, the AFGR represents the best comparison data currently available on high school graduation rates for the 50 states.

I picked 2001 as the starting year for my analysis because the US average AFGR was in decline until the turn of the century and then began to increase again. Thus, any state with a small increase or even worse, a decline, in its graduation rate would be performing below the national norm.

Most of the information concerning the first year the Age 18 minimum was adopted in each state was obtained from an e-mail from the Education Commission of the States – My thanks to Jennifer Dounay Zinth of the ECS for compiling that information. I also relied on information in the Rennie Center report and in a 2008 Scripps Howard news release for a few states where ECS was unable to locate data.

I did not examine states that had only two or one year of experience with the Age 18 minimum as of 2007, as these states offered insufficient data.

Once those states with at least four years of Age 18 graduation rate data were determined, I ran a regression analysis to determine the slope of the best fit line for the data points for each state. Those slopes are shown by the graph.