Every city in our report has dozens or even hundreds of programs, projects, facilities, and initiatives aimed at improving the lives of children. We have chosen one project - not necessarily the biggest or the most important - to help illustrate the breadth and diversity of this work. These programs represent tens of thousands of hours (by professional staff and volunteers) and millions of dollars (in public and private funds) which are going to improve the quality of lives of kids in our cities. There may be no more important work being done to ensure a quality future for all.

Data on total population was obtained from the 2000 U.S. Census survey.

Data on the percent of the population under 18 years of age was obtained from the 2000 U.S. Census survey.

Data on the number of children's museums was obtained from the Association of Youth Museums. These are museums geared specifically to children and/or youth.

The Health Improvement Grade was based on changes in three indicators of health between 1990 and 1998 - percent births to teens, percent low birthweight births, and infant mortality rate. Where city-level data was unavailable in 1990, but was available in 1998, we only used county or state data to compute the change in health over those eight years.

Although we did not include this category in the cities' final grades and ranks, it is an important indication of both positive and negative changes taking place. Consider this an "effort" grade, much like the "effort" grades given to our students in their report cards.

Percent population change is the difference between the city's 1998 population and its 2000 population. Total population for 1998 was obtained from the Census Bureau's Population Estimates Program, which produces estimates for the years following the last decennial estimate. The decennial census counts are updated using existing data series such as births, deaths, federal tax returns, Medicare enrollment, and immigration. Total population for 2000 was obtained from Census 2000.

Too often, a city's rapid population growth is not matched by equal growth in hospitals, schools, recreational areas, and other vital service sectors. Both rapid population growth and rapid population decline were considered to be negative characteristics, because the stability of a community is essential for meeting the needs of its residents, especially children.

Percent births to teens reflects the number of live births to women under 20 years of age divided by the total number of live births in a city. Data was obtained from the Division of Vital Statistics at the National Center for Health Statistics. The data is from 1998 and is reported by mother's residence, not place of birth.

Where city-level data was unavailable for some small cities, we used county-level data. Only state-level data was available for Fargo, Laredo, Huntington, Casper, and Cheyenne.

Babies born to teen mothers are at a higher risk of low birthweight and infant mortality compared with babies born to older mothers. They are more likely to grow up in homes that have lower levels of emotional support, and they are less likely to earn high school diplomas. Teen motherhood is often accompanied by poverty, diminishing opportunities for both mother and child.

Percent low birthweight births reflects the percent of live births weighing less than 2,500 grams, or about 5.5 pounds. Data was obtained from the Division of Vital Statistics at the National Center for Health Statistics. The data is from 1998 and is reported by mother's residence, not place of birth.

Where city-level data was unavailable for some small cities, we used county-level data. Only state-level data was available for Fargo, Laredo, Huntington, Casper, and Cheyenne.

Low birthweight infants (infants born weighing less than 2,500 grams, or about 5.5 pounds) are at higher risk of death or long-term illness and disability than are infants of normal birthweight. Low birthweight babies are more than 20 times more likely to die during the first year of life than normal birthweight infants.

Infant mortality is defined as the death of an infant before his or her first birthday. Data was obtained from the Division of Vital Statistics at the National Center for Health Statistics. The data is from 1998 and is reported by mother's residence, not place of birth.

Where city-level data was unavailable for some small cities, we used county-level data. Only state-level data was available for Fargo, Laredo, Huntington, Casper, and Cheyenne.

The infant mortality rate is an important measure of the well-being of infants, children, and pregnant women because it reflects maternal health, quality of access to medical care, and socioeconomic conditions. It captures the intrinsic link between the health of a mother and that of her child.

Number of Title X-funded clinics was obtained from the 2001/2002 Directory of the Office of Population Affairs. In the case of Florida and Hawaii, the data was obtained directly from the states' family planning divisions.

Title X clinics play a crucial role in preventing unintended pregnancies, including adolescent pregnancies. These services are the principal source of health care for many, particularly the uninsured. Nearly 85 percent of the population served by Title X clinics is from low-income households, 30 percent is less than 20 years of age, and 60 percent is less than 25 years of age.

Average SAT score data was obtained from the National Center for Education Statistics. SAT scores are from 1999-2000. The highest possible score is a 1600.

Both the SAT and the ACT are used primarily to assess how students will perform during the first year of college. A large number of colleges require students to report either the ACT or the SAT on admission applications. We used both ACT and SAT scores because of the varying proportions of students in each state taking these two tests.

Please Note: Because the U.S. Census Bureau hasn't yet released city-level education data, we used state-level information rather than the more outdated 1990 city-level Census data.

Average ACT scores were obtained for 2000 by ACT, Inc. The highest possible score is 36.

Both the SAT and the ACT are used primarily to assess how students will perform during the first year of college. A large number of colleges require students to report either the ACT or the SAT on admission applications. We used both ACT and SAT scores because of the varying proportions of students in each state taking these two tests.

Please Note: Because the U.S. Census Bureau hasn't yet released city-level education data, we used state-level information rather than the more outdated 1990 city-level Census data.

Average elementary class size data was obtained from the National Center for Education Statistics. SAT scores are from 1999-2000. The data is from 1993-94, but was published last year by the NCES.

The National Center for Education Statistics in a May 2000 report concluded that smaller class sizes--more than any other factor-determine higher student achievement. The study analyzed 18 variables, including school size, class size, cohesion of faculty, teacher qualifications, and teacher perceptions of school climate.

Please Note: Because the U.S. Census Bureau hasn't yet released city-level education data, we used state-level information rather than the more outdated 1990 city-level Census data.

Average secondary class size data was obtained from the National Center for Education Statistics. SAT scores are from 1999-2000. The data is from 1993-94, but was published last year by the NCES.

The National Center for Education Statistics in a May 2000 report concluded that smaller class sizes--more than any other factor-determine higher student achievement. The study analyzed 18 variables, including school size, class size, cohesion of faculty, teacher qualifications, and teacher perceptions of school climate.

Please Note: Because the U.S. Census Bureau hasn't yet released city-level education data, we used state-level information rather than the more outdated 1990 city-level Census data.

Data on violent crime per 1000 people was obtained from the Federal Bureau of Investigation log of crimes reported in 1999. Violent crime is defined as murder and non-negligent manslaughter, forcible rape, robbery, and aggravated assault. The data has not been adjusted for under-reporting, which may affect comparability between geographic areas or over time.

According to the Department of Justice's Office of Juvenile Justice and Delinquency Prevention (OJJDP), children are at much greater risk of being victims of crime than they are of being perpetrators. Juveniles make up a total of 12 percent of all crime victims, as reported to the police. They are 71 percent of the victims of all sex crimes and 38 percent of all kidnapping. One out of every 18 victims of violent crime, and one of every 3 victims of sexual assault, is under age 12.

Data on property crime per 1000 people was obtained from the Federal Bureau of Investigation log of crimes reported in 1999. Property crime is defined as burglary, larceny-theft, and motor vehicle theft. Arson was excluded from property crime data due to the unavailability of data for many cities. The data has not been adjusted for under-reporting, which may affect comparability between geographic areas or over time.

According to the Department of Justice's Office of Juvenile Justice and Delinquency Prevention (OJJDP), children are at much greater risk of being victims of crime than they are of being perpetrators. Juveniles make up a total of 12 percent of all crime victims, as reported to the police. They are 71 percent of the victims of all sex crimes and 38 percent of all kidnapping. One out of every 18 victims of violent crime, and one of every 3 victims of sexual assault, is under age 12.

Unemployment rate is the percent of the civilian labor force that had no employment during the reference week, despite being available for work (except for temporary illness) and making specific efforts to find employment some time during the four-week period ending with the reference week. Data is from 1998 and is reported by the Bureau of Labor Statistics.

Lack of family income is a serious threat to children's health and well-being because it exposes children to poverty. Parents who suffer from the emotional stress of thwarted attempts to provide for their family's basic needs are less effective in supporting, guiding, and nurturing their children. Since most parents obtain health insurance for themselves and their children through their employers, lack of employment can mean no health care. According to the Children's Defense Fund, children's risks multiply when families cannot afford learning materials, good quality child-care, decent, stable housing, or the hope of living outside a crime-ridden neighborhood.

For the Major cities , we were able to obtain data on the percent of homes that are affordable for a median income family from the National Association of Homebuilders (NAHB). Data is from the fourth quarter of 2000.

Without decent, affordable housing, working-class families face difficulties in realizing their dreams. The stability that comes with homeownership is important for child development and achievement. According to NAHB, "Homeownership is the cornerstone of family security, stability and prosperity. It strengthens the nation's communities, encourages civic responsibility and provides a solid foundation from which Americans can work to support their families, enhance their communities and achieve their personal goals."

The number of bad air days was obtained from the 1998 National Air Quality and Emissions Trends Report, published by the Environmental Protection Agency (EPA). EPA reports the number of days with Air Quality Index (AQI) values greater than 100 for the nation's 94 largest metropolitan areas. An AQI value over 100 is considered unhealthy for sensitive groups.

The AQI is a recent revision of the older Pollution Standards Index (PSI). Because the new index values are not readily available for all metropolitan areas, data for the Independent cities came from AIRData, an EPA on-line database. Since data for the Independent cities is based on the older PSI, it cannot be directly compared to data for the Major and Component cities.

While five criteria pollutants can contribute to AQI, the index primarily reflects ozone levels. Our report looks specifically at "bad air days" resulting from ozone. According to the EPA, children and people with asthma are most at risk from ozone "bad air days." Exposure to ozone for several hours at relatively low concentrations has been found to significantly reduce lung function. Children are especially at risk because they spend a lot of time outdoors.

Park acreage per 1000 persons for the 25 Major cities came from Peter Harnik's book Inside City Parks (Urban Land Institute, 2000). Data for all other cities was obtained by contacting each city's mayor and local department of parks and recreation. Data was imputed in select cases where we were unable to gather data from city officials. Parks data is "self-reported" and is an estimate of all park space located within the city's boundary, regardless of the managing agency. In many cases, park space includes undeveloped land, sports fields, natural reserves, lakes and reservoirs, and trails.

Recreational facilities provide children with infinite social, health, and cultural benefits. Public parks, recreation facilities and creative programs offer alternatives proven to help prevent high-risk youth from entering the juvenile justice system.

Library circulation per child data was obtained from the American Library Association's 2000 Statistical Report. Where data was not available, we contacted the city and/or city library directly. Data was imputed in select cases where we were unable to gather data from city officials.

Children's library program attendance counts were obtained from the American Library Association's 2000 Statistical Report. Where data was not available, we contacted the city and/or city library directly. Data was imputed in select cases where we were unable to gather data from city officials.