Forget the Average—It’s the Top Students Who Drive National Innovation Progress

When it comes to a nation’s innovation and global competitiveness, the Pareto Principle holds true: A small group drives most of the impact. In this context, it doesn’t make sense to judge a country’s future potential by the average performance of all its students. After all, not every student will go on to become a scientist, engineer, or researcher, the roles that fuel breakthrough products and processes. What truly matters is how well the top students perform, since they will play a disproportionate role in shaping the nation’s technological and economic edge. That is why the current trend in U.S. education to eliminate programs for top students, under the rubric of greater equity, is so damaging to the nation and must be immediately reversed.

An ITIF analysis of data from the Organization of Economic Cooperation and Development’s Programme for International Student Assessment (PISA) and the World Intellectual Property Organization’s Global Innovation Index shows the most innovative nations tend to rank higher based on students’ 90^th percentile scores rather than their average scores. This indicates that top-percentile performance is a better predictor of a nation’s innovativeness and competitiveness. The policy implications are clear: Expand, don’t contract, K-12 gifted and talented programs, and provide more advanced and honors classes for high school students. Unfortunately, under the guise of diversity, equity, and inclusion, some states and cities are moving in the opposite direction—cutting these kinds of programs. That is a path to national decline.

The top 15 most innovative nations, based on the Global Innovation Index, tended to rank higher in their students’ 90^th percentile cumulative math, science, and reading scores compared to their mean scores. Of these 15 nations, 11 had higher or equal rankings based on their 90^th percentile scores compared to their mean scores. These nations include Sweden, the United States, and Singapore. For example, the United States, one of the world’s most innovative economies, ranked 18^th based on mean scores but 12^th based on 90^th percentile scores. (See figure 1.) This suggests that 90^th percentile performance is likely a stronger indicator of a nation’s innovativeness and competitiveness than average scores. To note, the asterisk in the chart indicates: “Caution is required when interpreting estimates because one or more PISA sampling standards were not met.”

Figure 1: Ranking of the top 15 most innovative nations' 90th percentile and mean cumulative math, science, and reading scores (parentheses indicate nation’s rank in Global Innovation Index)

An analysis of 76 nations with available data from PISA and the Global Innovation Index shows a correlation of -0.87 between 90^th percentile cumulative scores in math, science, and reading scores and a nation’s ranking in innovativeness, meaning that higher scores are associated with higher rankings. In comparison, the correlation between the mean score and a nation’s innovativeness ranking was -0.84, indicating that 90^th percentile scores are slightly better predictors of a nation’s innovativeness. When limiting the data to the 50 most innovative nations, 90^th percentile scores remained the stronger predictor, with a correlation coefficient of 0.72 compared to 0.66 for mean scores.

Figure 2: Correlation between 90th percentile cumulative math, reading, and science scores and Global Innovation Index ranking

The 90^th percentile reading scores are also a slightly better predictor of a nation’s competitiveness and innovativeness than mean scores. The correlation between 90^th percentile reading scores and a nation’s innovativeness ranking was -0.83, compared to -0.81 for mean scores. (See figure 2.) When limiting the analysis to the 50 most innovative nations, the 90^th percentile reading scores remained the stronger predictor, with a correlation coefficient of -0.70 compared to -0.63 for mean scores.

Figure 3: Correlation between 90th percentile reading scores and Global Innovation Index ranking

For science scores, the 90^th percentile scores are again a slightly better predictor of a nation’s competitiveness. The correlation between the 90^th percentile scores and a nation’s innovativeness ranking was -0.87, compared to -0.84 for mean scores. (See figure 4.) When the data is limited to the top 50 most innovative nations, the correlation coefficient for 90^th percentile score was -0.72, compared to -0.66 for mean scores.

Figure 4: Correlation between 90th percentile science scores and Global Innovation Index ranking

The 90^th percentile math scores are not necessarily a better indicator of a nation’s innovativeness than mean scores. The correlation between 90^th percentile math scores and a nation’s ranking in innovativeness was -0.85, the same as the correlation for mean scores. (See figure 5.) However, when limiting the analysis to the 50 most innovative nations, the correlation coefficient for 90^th percentile scores was -0.68, slightly stronger than -0.66 for mean scores.

Figure 5: Correlation between 90th percentile math scores and Global Innovation Index ranking

Policymakers should support, rather than undermine, programs for gifted and talented students, particularly those focused on STEM. This includes strengthening specialty math and science high schools and ensuring that admission is based on merit, not lottery. It also means expanding, rather than cutting, Advanced Placement and other honors classes in high schools, and increasing resources for gifted and talented students. To the extent there is still a U.S. Department of Education, it should withhold funding from any school district that eliminates these kinds of programs focused on high scholarly achievement.

The current assault on academic excellence, especially in STEM, is a path to U.S. decline, particularly vis-à-vis China. As such, it is time to prioritize programs for high-performing students alongside efforts to promote equity, rather than subordinate them.