From Foresight, No. 45
published 2006
Something about the human condition compels us to compare and rank nearly every facet of our lives. Wines, consumer electronics, and recruiting classes, which are evaluated regularly and ranked according to their quality and value, quickly come to mind. Likewise, states are ranked in numerous ways according to demographic, economic, educational, and social characteristics. Such comparisons enable us to track progress over time and determine where we stand relative to other states. States, however, are multidimensional, making it difficult to derive a composite or overall ranking. For example, relative to other states, Kentucky has a low crime rate, a high home ownership rate, low per capita income, and high toxic air emissions. We know where Kentucky ranks on these individual factors, but how does it look overall?
We have developed an approach that combines 32 different factors into a single quality-of-life index. The State of the Commonwealth Index is a single number that summarizes Kentucky’s overall quality of life relative to other states over time. Based on 1990 to 2003 data from national surveys and studies of various indicators of well-being in the states, the State of the Commonwealth Index includes factors ranging from teen pregnancy, poverty, and voter participation rates to toxic releases to air, water, and land. Together, they form a data-driven index that offers a richer understanding of how we are faring now and how our status has changed relative to other states.
Based on our index, Kentucky made slow but steady progress between 1990 and 2003. Our national ranking improved from the mid 40s to the low 40s while our regional ranking remained about the same. In the sections that follow, we describe how the index was created, which indicators are used, and how Kentucky’s rank has changed relative to the nation and its peer states.
The State of the Commonwealth Index combines 32 long-term quality-of-life indicators covering 1990 to 2003,(1) including measures of community attributes, education, the economy, the environment, and government (see Table 1).(2) The index uses summary statistical information about each indicator to construct a number ranging from 0 to 1000 that expresses how each state’s measure compares to other states. The higher the score, the better a state ranks among the states.(3) The final index score is the average of five subindex scores based on the quality of life areas measured by the indicators. They include the subindexes of communities, education, the economy, the environment, and government, ranging from 0 to 200 based on the equal weights given to each area. In addition to comparing Kentucky to all the states, a second index was created comparing Kentucky to its peer states.(4) This group of states includes those demographically, geographically, and economically similar in makeup to Kentucky; 17 states are included in the estimation of this second index.(5)
Table 1: The 32 Long-Term Quality-of-Life Indicators Used in the State of the Commonwealth Index
Although it is a comprehensive, data-driven index, caveats and complicating factors that could potentially affect the outcome of its values include choice of the weighting scheme, the quantity and types of indicators included, and the inherent quantitative bias of the method. The framework used to construct the index reflects findings from 15 public forums held throughout Kentucky and a statewide conference in 1994 in which over 500 people participated. From these meetings, a vision statement of Kentucky’s future emerged which highlighted the five main themes used here to construct the subindexes. The weighting scheme reflects the values and priorities highlighted during these meetings.(6) In addition, while there are countless quality-of-life indicators available, the indicators chosen reflect the same values Kentuckians expressed in their vision of the state’s future. It is also important to note that, arguably, many facets define quality of life which do not easily lend themselves to quantification, inherently biasing any index of this kind towards those that can be quantified. In light of these caveats, however, the final form and methodology used here is reasonable, given our approach, which aims to retain the values and ideals generally held by many Kentuckians from around the state.
As a result of feedback we received since the initial release of the State of the Commonwealth Index in 2004, we have removed two indicators and added eight. Our rationale for these changes is grounded in recent trends. Obesity, for example, has been identified as a major risk factor for a number of ailments, increasing its importance in the health status of a state’s population. A measure of educational achievement by minorities was added in recognition of its prominence in the federal No Child Left Behind Act, which is becoming an integral measure of advancement. Previously, the index had no indicator for the arts, which has been positively linked to other areas of educational achievement as well as economic development; thus, an arts indicator was added to lend greater balance to the Index. Because research has shown a positive relationship between high-speed Internet use and economic growth and development, we added an indicator on broadband access. A more comprehensive indicator, entrepreneurial capacity, has replaced our previous measure of business formation. A measure of road conditions has been added in recognition of the integral role that a state’s transportation system plays in the development of its economy. Also, in light of recent developments concerning the availability of oil and natural gas, a measure of motor fuel use was included to account for the stewardship of this valuable natural resource. Finally, we now use a more refined measure of voting rates that excludes ineligible noncitizens and felons from the final calculation.
Kentucky made slow but steady progress from 1990 to 2003. Figure 1 shows Kentucky’s rank in the State of the Commonwealth Index from 1990 to 2003, both nationally and relative to its peer states.(7) In the national ranking, Kentucky progressed from ranking in the mid 40s (1990-1994), to the low to mid 40s (1995-1997, to the low 40s (1998-2002). Kentucky’s rank did decline in 2003, but at this point we cannot determine whether this is a temporary aberration or the beginning of a trend. Kentucky’s ranking relative to its peer states has remained fairly constant throughout this period.
Figure 1: Kentucky Ranked by Index Score Relative to the US and Peer States, 1990 to 2003
On the State of the Commonwealth Index, Kentucky gained ground in the areas of education, the economy, and the environment, held steady in the area of communities, and lost ground in the area of government between 1990 and 2003 both nationally and among its peer states (see Table 2). Kentucky also improved its standing nationally and among its peer states in many of the 32 long-term quality-of-life indicators. The state made progress on 16 of the indicators nationally and relative to the comparison states. On the remaining indicators, Kentucky held steady on 4 indicators and lost ground on 12 nationally, and on 3 and 13, respectively, compared to its peer states.
Improvements in the areas of education, economy, and the environment account for all of the growth that Kentucky experienced from 1990 to 2003, with 13 indicators increasing in rank nationally in these three areas and 14 relative to Kentucky’s peers. In 1990, Kentucky placed 44th out of the 50 states and 12th out of the peer states based on a ranking of the education subindex scores (see Table 2). By 2003, Kentucky’s education subindex had advanced three places in rank to 41st in the nation and three places to 9th among the peer states in 2003. A ranking of the states based on the indicator index scores shows that Kentucky’s ranking remained unchanged or improved on each one, with the exception of college attainment rates relative to Kentucky’s peer states and the educational achievement gap relative to both the nation and Kentucky’s peers. Kentucky’s high school attainment rates improved the most from 1990 to 2003, increasing from 47th in the nation in 1990 to 35th by 2003 and from 14th to 9th among its peers. The percent of Kentucky adults 25 to 64 years old with at least a high school diploma increased from 77 percent in 1990 to 86 percent in 2003. Kentucky improved its National Assessment of Educational Progress (NAEP) Eighth-Grade Math Exam scores, moving from 43rd nationally in 1990 to 35th in 2003 on this indicator index. The percentage of Kentucky eighth-grade students scoring at or above the basic level on the NAEP Math Exam increased from 43 percent in 1990 to 65 percent in 2003.
While the overall trend for Kentucky between 1990 and 2003 was one of slow but steady progress, the state remains well below average nationally. Improvements in education helped boost Kentucky’s ranking and kept it on its chosen path of slow but steady progress. Sustaining this commitment to high-quality, accessible education at all levels is key to achieving the goals that the architects of educational reform envisioned. To accelerate the state’s rate of progress, targeted investment and creative uses of the resources we have will be needed to leapfrog ahead of other states and create and sustain an enviable quality of life across the Commonwealth.
* Dr. Watts is a Policy Analyst with the Kentucky Long-Term Policy Research Center. Return to text.
1 2003 is the last year for which we have data for all the indicators and all the states. Return to text.
2 For further information on the indicators and their sources please see http://www.kltprc.net/stateofthecommonwealthappendix.htm. Return to text.
3 The indicators were standardized to facilitate their comparison with each other and their combination into one summary statistic. By transforming all outcomes to Z-scores, with the same mean (0) and standard deviation (1), all of the indicators could be compared and combined using a common yardstick. Although the use of standardized outcome measures provides a common yardstick with which to compare and combine the different indicator measures, it still is not completely satisfying for the purpose of presentation. This drawback is attributable to the fact that standardized outcomes can take on an infinite range of values that indicate only the direction and number of standard deviations of the difference between the given score and the mean score for the particular outcome. In contrast, the probability values associated with the standardized outcome scores represent a measure with more intuitive appeal. They range from 0 to 1, or, in this case, from 0 to 1000 with an average of 500. These values were derived directly from the Z-scores, using a cumulative standard normal distribution. For example a Z-score of 0 equals a probability of 50 percent or, here, an index score of 500. Conceptually, the result represents the percentile ranking of the Z-scores, and indicates the extent to which the state performed well or poorly relative to the other states included in the calculation of the index.
For example, using per capita income, the first step in this method is to calculate the mean and standard deviation across all the states for a particular year. In 2003, Kentucky’s per capita income was $26,352. The mean and standard deviation across all 50 states for that year were $30,375 and $4,236, respectively. The Z-score was calculated as ($26,352-$30,375)/$4,236, which equals a value of –0.9. The probability value for this Z-score value is 0.171. This value was then multiplied by 1000 to obtain 171—Kentucky’s per capita income index score for 2003 relative to the nation. The economy subindex score was then obtained by calculating the average of this score and the eight other indicators included in this quality-of-life area. Upon calculation of this score, the final index score was the average of each of the five subindex scores. Return to text.
4 For further information on how these states were selected please see http://www.kltprc.net/stateofthecommonwealthappendix.htm. Return to text.
5 The peer states include Alabama, Arkansas, Florida, Georgia, Illinois, Indiana, Louisiana, Michigan, Mississippi, Missouri, North Carolina, Ohio, South Carolina, Tennessee, Virginia, and West Virginia. Including Kentucky, the total number of states used in this part of the analysis is 17. Return to text.
6 Choosing a weighting scheme can be problematic in that any one chosen inherently makes assumptions about the relative importance of the indicators and imposes those assumptions on the final calculation. For example, weighting the indicators equally imposes the assumption that those areas with more indicators are more important than those with fewer. In this case, the results are comparable to those shown here. The current weighting system, while it does not avoid this inherent bias, draws upon the input of a multitude of people throughout the state, rather than a few. Return to text.
7 For further information on how these states were selected please see http://www.kltprc.net/stateofthecommonwealthappendix.htm. Return to text.