Methodology

Employment and Earnings

Poverty and Opportunity

Health & Well-Being

Reproductive Rights

Violence & Safety

Work & Family

Political Participation

 

Employment and Earnings

To analyze the status of women in the states, IWPR selected indicators that prior research and experience have shown illuminate issues that are integral to women’s lives and that allow for comparisons between each state and the United States as a whole. The data in IWPR’s Status of Women in the States reports come from federal government agencies and other sources; many of the figures rely on analysis of the U.S. Census Bureau’s American Community Survey (ACS) from the Minnesota Population Center’s Integrated Public Use Microdata Series (IPUMS). Much of the analysis for IWPR’s 1996–2004 Status of Women in the States reports relied on the Current Population Survey (CPS).

The tables and figures present data for individuals, often disaggregated by race and ethnicity. In general, race and ethnicity are self-identified; the person providing the information on the survey form determines the group to which he or she (and other household members) belongs. People who identify as Hispanic or Latino may be of any race; to prevent double counting, IWPR’s analysis of American Community Survey microdata separates Hispanics from racial categories—including white, black (which includes those who identified as black or African American), Asian/Pacific Islander (which includes those who identified as Chinese, Japanese, and Other Asian or Pacific Islander, including Native Hawaiians), or Native American (which includes those who identified as American Indian or Alaska Native). The ACS also allows respondents to identify with more specific racial groups and/or Hispanic origins. Detailed racial/ethnic information is available for American Indians and Alaska Natives, Asian/Pacific Islanders, and Hispanics, but not for blacks or whites. IWPR conducted analysis of selected indicators for the groups for which detailed information is available (when sample sizes were not large enough, detailed races/ethnicities were combined into “other” categories based on their corresponding major racial or ethnic group). Published data from the Bureau of Labor Statistics that are cited in the text do not include Pacific Islanders and classify Hispanics in the racial groups with which they identify as well as separately, and in the data that come from these datasets Hispanics are double counted.

When analyzing state- and national-level ACS microdata, IWPR used 2013 data, the most recent available, for most indicators. When disaggregating data at the state level by race and ethnicity, analyzing median annual earnings for young women by state, and analyzing the employment and earnings of women by detailed racial and ethnic group nationally, IWPR combined three years of data (2011, 2012, and 2013) to ensure sufficient sample sizes. IWPR constructed a multi-year file by selecting the 2011, 2012, and 2013 datasets, adjusting dollar values to their 2013 equivalents using the Consumer Price Index for All Urban Consumers, and averaging the sample weights to represent the average population during the three year period. Data on median earnings are not presented if the unweighted sample size is less than 100 for any cell; data on other indicators are not presented if the average cell size for the category total is less than 35.

Earnings lost over time due to the gender wage gap were estimated by comparing the median annual earnings of women and men who worked full-time, year-round using the 1980–2014 CPS Annual Social and Economic Supplements (ASEC). Birth year was estimated by subtracting age from the year of the survey data collection. Earnings were adjusted to 2014 dollars using the CPI-U. The differences in earnings between women and men by single year of age were calculated within five-year birth cohorts and summed to calculate the cumulative losses for all women and for women with a bachelor’s degree or higher.

IWPR used personal weights to obtain nationally representative statistics for person-level analyses. Weights included with the IPUMS ACS for person-level data adjust for the mixed geographic sampling rates, nonresponses, and individual sampling probabilities. Estimates from IPUMS ACS samples may not be consistent with summary table ACS estimates available from the U.S. Census Bureau due to the additional sampling error and the fact that over time, the Census Bureau changes the definitions and classifications for some variables. The IPUMS project provides harmonized data to maximize comparability over time; updates and corrections to the microdata released by the Census Bureau and IPUMS may result in minor variation in future analyses.

Differences Between the ACS and the CPS

The differences between the ACS and CPS and their impact on measures of employment and earnings are described in detail in Kromer and Howard (2011). These differences have some bearing on this report’s comparisons with data from IWPR’s 2004 report, as well as on the reported differences in data for 2013 that come from the two surveys. While both the ACS and the CPS survey households, their sample frames also include noninstitutionalized group quarters, such as college dorms and group homes for adults. The ACS also includes institutionalized group quarters, such as correctional facilities and nursing homes. College students away at school and living in a dormitory are treated differently in the two surveys. In the ACS they would be residents of the dorm in the group quarters population while in the CPS they remain a member of their family household. While all CPS interviews are collected using computer-assisted interviews, about half of the ACS households respond using the paper mail-back form and half by computer-assisted interview (Census Bureau 2014). The ACS collects data on work and earnings in the previous 12 months throughout the year while the CPS-ASEC collects work and earnings information for the previous calendar year during interviews collected February-April each year. Finally, the two surveys have differences in wording of some questions that aim to collect similar social and demographic information.

Calculating the Composite Index

To construct the employment and earnings composite index, each of the four component indicators was first standardized. For each of the indicators, the observed value for the state was divided by the comparable value for the entire United States. The resulting values were summed for each state to create a composite score. Each of the four component indicators has equal weight. The states were ranked from the highest to the lowest scores.

To grade the states on this composite index, values for each of the components were set at desired levels to provide an “ideal score.” Women’s earnings were set at the median annual earnings for men in the United States overall; the wage ratio was set at 100 percent, as if women earned as much as men; women’s labor force participation was set at the national number for men; and women in managerial or professional occupations was set at the highest score for all states. Each state’s score was compared with the ideal score to determine the state’s grade.

WOMEN’S MEDIAN ANNUAL EARNINGS: Median annual earnings of women aged 16 and older who worked full-time, year-round (50 or more weeks per year and 35 or more hours per week) in 2013. The sample size for women ranged from 713 in Alaska to 44,866 in California. These earnings data have not been adjusted for cost-of-living differences between the states because the federal government does not produce an index of such differences. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center.

RATIO OF WOMEN’S TO MEN’S EARNINGS: Median annual earnings of women aged 16 and older who worked full-time, year-round (50 or more weeks per year and 35 or more hours per week) in 2013 divided by the median annual earnings of men aged 16 and older who worked full-time, year-round in 2013. Sample sizes ranged from 713 in Alaska to 44,866 in California for women’s earnings, and from 1,074 in Alaska to 62,903 in California for men’s earnings. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center.

WOMEN’S LABOR FORCE PARTICIPATION (proportion of the adult female population in the labor force): Percent of women aged 16 and older who were employed or looking for work in 2013. This includes those employed full-time, part-time voluntarily, or part-time involuntarily, and those who are unemployed but looking for work. The percent of women in the labor force in IWPR’s 1996–2004 Status of Women in the States reports included the civilian, noninstititutionalized population. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center.

WOMEN IN MANAGERIAL AND PROFESSIONAL OCCUPATIONS: Percent of women aged 16 and older who were employed in executive, administrative, managerial, or professional specialty occupations in 2013. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center.

 

Poverty and Opportunity

To analyze the status of women in the states, IWPR selected indicators that prior research and experience have shown illuminate issues that are integral to women’s lives and that allow for comparisons between each state and the United States as a whole. The data in this report come from federal government and other sources; many of the figures rely on analysis of the U.S. Census Bureau’s American Community Survey (ACS) from the Minnesota Population Center’s Integrated Public Use Microdata Series (IPUMS; Ruggles et al.). Much of the analysis for IWPR’s 1996–2004 Status of Women in the States reports relied on the Current Population Survey (CPS).

The tables and figures present data for individuals, often disaggregated by race and ethnicity. In general, race and ethnicity are self-identified; the person providing the information on the survey form determines the group to which he or she (and other household members) belongs. People who identify as Hispanic or Latino may be of any race; to prevent double counting, IWPR’s analysis separates Hispanics from racial categories—including white, black (which includes those who identified as black or African American), Asian/Pacific Islander (which includes those who identified as Chinese, Japanese, and Other Asian or Pacific Islander), or Native American (which includes those who identified as American Indian or Alaska Native). The ACS also allows respondents to identify with more specific racial/ethnic groups and/or Hispanic origins. Detailed racial/ethnic information is available for American Indians and Alaska Natives, Asian/Pacific Islanders, and Hispanics, but not for blacks or whites. IWPR conducted analysis of selected indicators for the groups for which detailed information is available. When sample sizes were not large enough, detailed races/ethnicities were combined into “other” categories based on their corresponding major racial or ethnic group.

When analyzing state- and national-level ACS microdata, IWPR used 2013 data, the most recent available, for most indicators. When analyzing poverty rates by household type at the state level, poverty and opportunity indicators by detailed racial and ethnic group nationally, and educational attainment among young women by state, IWPR combined three years of data (2011, 2012, and 2013) to ensure sufficient sample sizes. IWPR constructed a multi-year file by selecting the 2011, 2012, and 2013 datasets, adjusting dollar values to their 2013 equivalents using the Consumer Price Index for All Urban Consumers, and averaging the sample weights to represent the average population during the three year period. When examining employment-based health insurance coverage among women by union status, IWPR used three-year (2012–2014) data from the Current Population Annual Social and Economic Supplement. Data are not presented if the average cell size for the category total is less than 35.

Same-sex households were identified by matching the gender of the household head with the gender of their spouse or partner living in the same household using 2013 ACS microdata. If the genders matched, the couple was defined as a same-sex male/same-sex female couple. Due to sample size limitations, no distinction was made between same-sex cohabiting and same-sex married couples. For analyses of individuals living in same-sex households, the sample was limited to only heads of households and their spouses or partners to eliminate the possibility of children, other relatives, or roommates of same-sex couples being classified as a woman or man in a same-sex relationship. To estimate poverty rates among same-sex, different-sex, and single men/single women households, IWPR calculated the total income from all individuals in the household, then compared the total household income to the 2013 poverty thresholds from the Census Bureau using the “weighted average thresholds” by number of people in the household (without regard to the age of the householder). Households with total income that is less than the threshold value for the household’s size are considered to be “poor.”

IWPR used personal weights to obtain nationally representative statistics for person-level analyses, and household weights for household-level analyses. Weights included with the IPUMS ACS for person-level data adjust for the mixed geographic sampling rates, nonresponses, and individual sampling probabilities. Estimates from IPUMS ACS samples may not be consistent with summary table ACS estimates available from the U.S. Census Bureau due to the additional sampling error and the fact that over time, the Census Bureau changes the definitions and classifications for some variables. The IPUMS project provides harmonized data to maximize comparability over time; updates and corrections to the microdata released by the Census Bureau and IPUMS may result in minor variation in future analyses.

Differences Between the ACS and the CPS

The differences between the ACS and CPS and their impact on measures related to poverty have some bearing on this report’s comparisons with data from IWPR’s 2004 report and on the reported differences in data for 2013 that come from the two surveys. While both the ACS and the CPS survey households, their sample frames also include noninstitutionalized group quarters, such as college dorms and group homes for adults. The ACS also includes institutionalized group quarters, such as correctional facilities and nursing homes (U.S. Department of Commerce 2014b). College students away at school and living in a dormitory are treated differently in the two surveys. In the ACS they would be residents of the dorm in the group quarters population while in the CPS they remain a member of their family household (Kromer and Howard 2011). While all CPS interviews are collected using computer-assisted interviews, about half of the ACS households respond using the paper mail-back form and half by computer-assisted interview (U.S. Department of Commerce 2014c). The ACS collects income and health insurance information in the previous 12 months throughout the year while the CPS-ASEC collects income and health insurance information for the previous calendar year during interviews collected February-April each year. While the ACS asks eight questions about income from different sources, the CPS-ASEC interview includes questions on more than 50 income sources (U.S. Department of Commerce 2014b). Finally, the two surveys have differences in wording of some questions that aim to collect similar social and demographic information.

Calculating the Composite Index

To construct the poverty and opportunity index, each of the four component indicators was first standardized. For each of the indicators, the observed value for the state was divided by the comparable value for the entire United States. The resulting values were summed for each state to create a composite score. Women’s health insurance coverage, educational attainment, and business ownership were given a weight of 1.0 each, while poverty was given a weight of 4.0 (in IWPR’s first three series of Status of Women in the States reports published in 1996, 1998, and 2000, this indicator was given a weight of 1.0, but in 2002 IWPR began weighting it at 4.0).  The states were ranked from the highest to the lowest scores.

To grade the states on this composite index, values for each of the components were set at desired levels to provide an “ideal score.” The percentage of women with health insurance and with a bachelor’s degree or higher were set at the highest values for all states; the percentage of businesses owned by women was set as if 50 percent of businesses were owned by women; and the percentage of women in poverty was set at the national value for men. Each state’s score was then compared with the ideal score to determine its grade. In previous IWPR Status of Women in the States report, the desired level of educational attainment was set at the national value for men. In 2013, however, the percentage of women aged 25 and older in the United States overall with a bachelor’s degree or higher slightly surpassed the percentage of men with this level of education; in 21 states and the District of Columbia in this year, the percentage of women with at least a bachelor’s degree was higher than the national average for men (29.5 percent).

PERCENT WITH HEALTH INSURANCE: Percent of women from ages 18 through 64 who are insured. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center (Ruggles et al. 2010).

EDUCATIONAL ATTAINMENT: In 2013, the percent of women aged 25 and older with a bachelor’s degree or higher. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center (Ruggles et al. 2010).

WOMEN’S BUSINESS OWNERSHIP: In 2007, the percent of all firms (legal entities engaged in economic activity during any part of 2007 that filed an IRS Form 1040, Schedule C; 1065; any 1120; 941; or 944) owned by women. The Bureau of the Census 2007 Survey of Business Owners asked the sex of the owner(s); a business is classified as woman-owned based on the sex of those with a majority of the stock or equity in the business.  Source: Calculations of data from the U.S. Department of Commerce, Bureau of the Census (IWPR 2015b).

PERCENT OF WOMEN ABOVE POVERTY: In 2013, the percent of women living above the federal poverty threshold, which varies by family size and composition. In 2013, the poverty level of a family of four (with two children) was $23,624 (U.S. Department of Commerce 2015). This report uses the official Federal definition of poverty that compares the cash income received by family members to an estimate of the minimum amount the family would need to meet their basic needs. Source: Calculations of 2013 American Community Survey microdata as provided by the Integrated Public Use Microdata Series (IPUMS) at the Minnesota Population Center (Ruggles et al. 2010).

 

Health & Well-Being

To analyze the status of women in the states, IWPR selected indicators that prior research and experience have shown illuminate issues that are integral to women’s lives and that allow for comparisons between each state and the United States as a whole. The data in IWPR’s Status of Women in the States reports come from federal government agencies and other sources; much of the analysis of women’s health relies on data from the Centers for Disease Control and Prevention (CDC), including the CDC’s Wide-ranging OnLine Data for Epidemiologic Research (WONDER), Web-based Injury Statistics Query and Reporting System (WISQARS), and National Center for HIV, STD, and TB Prevention Atlas databases. In addition, IWPR analyzed microdata from the Behavioral Risk Factor Surveillance System (BRFSS) survey, which is conducted by the CDC annually in conjunction with the states, the District of Columbia, and five U.S. territories. BRFSS measures behavioral risk factors for the noninstitutionalized adult population (aged 18 and older) living in the United States. Interviews are collected using telephone interviews; in 2011, the data collection methods were refined to include both land line and mobile telephone numbers in the sample to ensure all segments of the population were covered. Some of the changes noted in poor mental health, diabetes, and activities limitations could be due to methodological differences (Centers for Disease Control and Prevention 2012), but these data represent the best estimates of population health behaviors at the state level. In 2013, 491,733 interviews were completed (Centers for Disease Control and Prevention 2014e).

When analyzing state- and national-level BRFSS microdata, IWPR used 2013 data, the most recent available. When disaggregating data at the state level by race/ethnicity and by age, IWPR combined three years of data (2011, 2012, and 2013) to ensure sufficient sample sizes, with several exceptions. Data on the percent of women who exercise at least 150 minutes per week were available only for 2013; data on the percent of women who eat at least five servings of fruits or vegetables per day and have been screened for cholesterol in the past five years were available only for 2011 and 2013; and data on the percent of women who have had a pap test in the past three years and a mammogram in the past two years were available only for 2012 for all states and 2013 for five states. When analyzing the percent of women who have had a mammogram or pap test nationally, IWPR used 2012 data only. State-level estimates on these indicators combine 2012 and 2013 data. IWPR used sample weights provided by the CDC to obtain nationally representative statistics that adjust for sampling both landline and mobile telephone numbers. Data are not presented if the average cell size for the category total is less than 35.
The tables and figures present data for individuals, often disaggregated by race and ethnicity. In general, race and ethnicity are self-identified; the person providing the information for the survey determines the group to which he or she (and other household members) belongs. People who identify as Hispanic or Latino may be of any race; to prevent double counting, IWPR’s analysis separates Hispanics from racial categories—including white, black (which includes those who identified as black or African American), Asian/Pacific Islander (which includes those who identified as Chinese, Japanese, or other Asian or Pacific Islander), or Native American (which includes those who identified as American Indian or Alaska Native). Hispanics may be of any race or two or more races.

Calculating the Composite Index

This composite index includes nine measures of women’s physical and mental health: mortality from heart disease, mortality from lung cancer, mortality from breast cancer, incidence of diabetes, incidence of chlamydia, incidence of AIDS, mean days of poor mental health, mortality from suicide, and mean days of activity limitations. To construct the composite index, each of the component indicators was converted to scores ranging from 0 to 1 by dividing the observed value for each state by the highest value for all states. Each score was then subtracted from 1 so that high scores represent lower levels of mortality, poor health, or disease. Scores were then given different weights. Mortality from heart disease was given a weight of 1.0. Lung and breast cancer were each given a weight of 0.5. Incidence of diabetes, chlamydia, and AIDS were each given a weight of 0.5. Mean days of poor mental health and women’s mortality from suicide were given a weight of 0.5. Activity limitations were given a weight of 1.0. The resulting values for each of the component indicators were summed for each state to create a composite score. The states were then ranked from the highest to the lowest score.

To grade the states on this composite index, values for each of the components were set at desired levels to produce an “ideal score.” For each indicator, the desired level was set at the lowest rate or lowest level among all states. Each state’s score was then compared with the ideal score to determine the state’s grade. In previous Status of Women in the States reports, mortality rates from heart disease, lung cancer, and breast cancer were set according to national objectives for the year 2010, as determined by the U.S. Department of Health and Human Services under the Healthy People 2010 program, and all other indicators were set at the lowest rate or lowest level among all states.

MORTALITY FROM HEART DISEASE: Average annual mortality from heart disease among women of all ages per 100,000 population (in 2011–2013). Data are age-adjusted to the 2000 U.S. standard population. Source: Centers for Disease Control and Prevention, National Center for Health Statistics (CDC 2015a).

MORTALITY FROM LUNG CANCER: Average mortality among women of all ages from lung cancer per 100,000 population (in 2011–2013). Data are age-adjusted to the 2000 U.S. standard population. Source: Centers for Disease Control and Prevention, National Center for Health Statistics (CDC 2015a).

MORTALITY FROM BREAST CANCER: Average mortality among women of all ages from breast cancer per 100,000 population (in 2011–2013). Data are age-adjusted to the 2000 U.S. standard population. Source: Centers for Disease Control and Prevention, National Center for Health Statistics (CDC 2015a).

PERCENT OF WOMEN WHO HAVE EVER BEEN TOLD THEY HAVE DIABETES: As self-reported by female respondents in the Behavioral Risk Factor Surveillance System (BRFSS) survey in 2013. The Centers for Disease Control and Prevention conduct BRFSS in conjunction with the states among men and women at least 18 years of age. Source: IWPR analysis of BRFSS 2013 microdata (IWPR 2015b).

INCIDENCE OF CHLAMYDIA: Reported rate of chlamydia among women of all ages per 100,000 population in 2013. Source: Centers for Disease Control, National Center for HIV, STD, and TB Prevention, Division of STD Prevention 2013 (CDC 2015c).

INCIDENCE OF AIDS: Average incidence of HIV Stage 3 (AIDS)-indicating diseases among females aged 13 years and older per 100,000 population in 2012. Source: Centers for Disease Control and Prevention, National Center for HIV, STD, and TB Prevention Atlas (CDC 2015b).

POOR MENTAL HEALTH: Mean number of days in the past 30 days on which mental health was not good, as self-reported by female respondents in the BRFSS survey in 2013. The Centers for Disease Control and Prevention conduct BRFSS in conjunction with the states among men and women at least 18 years of age. Source: IWPR analysis of BRFSS 2013 microdata (IWPR 2015b).

MORTALITY FROM SUICIDE: Average annual mortality from suicide among women of all ages per 100,000 population in 2011–2013. Data are age-adjusted to the 2000 U.S. standard population. Source: Centers for Disease Control and Prevention Web-based Injury Statistics Query and Reporting System (CDC 2015e).

MEAN DAYS OF ACTIVITY LIMITATIONS: Mean number of days in the past 30 days on which activities were limited due to health status, as self-reported by female respondents in the BRFSS survey in 2013. The Centers for Disease Control and Prevention conduct BRFSS in conjunction with the states among men and women at least 18 years of age. Source: IWPR analysis of BRFSS 2013 microdata (IWPR 2015b).

 

Reproductive Rights

To analyze the status of women in the states, IWPR selected indicators that prior research and experience have shown illuminate issues that are integral to women’s lives and that allow for comparisons between each state and the United States as a whole. The data in IWPR’s Status of Women in the States reports come from federal government agencies and other sources; much of the data in this report rely on analysis from the Centers for Disease Control and Prevention and organizations such as the Guttmacher Institute, NARAL Pro-Choice America, and the National Center for Lesbian Rights. The tables present data for individuals, in some cases disaggregated by race and ethnicity. Data disaggregated by race and ethnicity are compiled from reports published by the Centers for Disease Control and Prevention (CDC) that define racial categories (white, black, Asian/Pacific Islander, and Native American) in different ways. In the data tables on prenatal care and low birthweight presented in the report, only whites and blacks are non-Hispanic; other racial categories include Hispanics. In the data on infant mortality, all racial categories are defined as non-Hispanic. Hispanics may be of any race or two or more races.

The reproductive rights composite index reflects a variety of indicators of women’s reproductive rights. These include access to abortion services without mandatory parental consent or notification laws for minors, access to abortion services without a waiting period, public funding for abortions if a woman is income eligible, the percent of women living in counties with at least one abortion provider, whether the governor and state legislature are pro-choice, whether states have adopted the Medicaid expansion under the ACA and/or expanded eligibility for Medicaid family planning services, policies that mandate insurance coverage of infertility treatments, whether same-sex marriage is recognized or second-parent adoption is allowed for same-sex couples, and mandatory sex education for children in the public school system. These indicators reflect two changes from IWPR’s previous Status of Women in the States reports that take into account recent policy developments: the indicator on Medicaid expansion replaces a previous indicator on state contraceptive coverage laws, and the indicator on same-sex marriage or second-parent adoption modifies a previous indicator on second-parent adoption.

Calculating the Composite Index

To construct this composite index, each component indicator was rated on a scale of 0 to 1 and assigned a weight. The notification/consent and waiting-period indicators were each given a weight of 0.5. The indicators of public funding for abortions, pro-choice government, women living in counties with an abortion provider, and Medicaid expansion and/or Medicaid family planning eligibility expansions were each given a weight of 1.0. The infertility coverage law and same-sex marriage and/or second-parent adoption laws were each given a weight of 0.5. Finally, states were given 1.0 point if they mandate sex education for students. The weighted scores for each component indicator were summed to arrive at the value of the composite index score for each state. The states were ranked from the highest to the lowest score.

To grade the states on this composite index, values for each of the components were set at desired levels to produce an “ideal score.” An ideal state was assumed to have no notification/consent or waiting period policies, public funding for abortion, a pro-choice government, 100 percent of women living in counties with an abortion provider, a Medicaid expansion or state Medicaid family planning eligibility expansion, infertility coverage, legal guarantees of same-sex marriage or second-parent adoption, and mandatory sex education for students. Each state’s score was then compared with the resulting ideal score to determine its grade.

MANDATORY CONSENT: States received a score of 1.0 if they allow minors access to abortion without parental consent or notification. Mandatory consent laws require that minors gain the consent of one or both parents before a physician can perform the procedure, while notification laws require they notify one or both parents of the decision to have an abortion. Source: Guttmacher Institute 2015b.

WAITING PERIOD: States received a score of 1.0 if they allow a woman to have an abortion without a waiting period. Waiting-period legislation mandates that a physician cannot perform an abortion until a certain number of hours after notifying the woman of her options in dealing with a pregnancy. Source: Guttmacher Institute 2015b.

RESTRICTIONS ON PUBLIC FUNDING: If a state provides public funding for all or most medically necessary abortions, exceeding federal requirements, for women who meet income eligibility standards, it received a score of 1.0. Source: Guttmacher Institute 2015b.

PERCENT OF WOMEN LIVING IN COUNTIES WITH AT LEAST ONE ABORTION PROVIDER: States were given a scaled score ranging from 0 to 1, with states with 100 percent of women living in counties with abortion providers receiving a 1. Source: Guttmacher Institute 2014.

PRO-CHOICE GOVERNOR OR LEGISLATURE: This indicator is based on NARAL’s assessment of whether governors and legislatures would support a ban or restrictions on abortion. Governors and legislatures who would support restrictions on abortion rights are considered anti-choice, and those who would oppose them are considered pro-choice. Legislatures with a majority that are neither anti- or pro-choice are considered mixed. Each state received 0.33 points per pro-choice governmental body—governor, upper house, and lower house—up to a maximum of 1.0 point. Those governors and legislatures with mixed assessments received half credit. Source: NARAL Pro-Choice America and NARAL Pro-Choice America Foundation 2015 and NARAL Pro-Choice America 2015.

MEDICAID EXPANSION: Whether a state had expanded Medicaid under the ACA or enacted a state Medicaid family planning eligibility expansion through either a waiver of federal policy from the Centers for Medicare and Medicaid Services or a state plan amendment: family planning eligibility expansions extend Medicaid coverage of family planning services to women who would be otherwise ineligible, and in some cases to women who are exiting the Medicaid program. States received a score of 1.0 if they have adopted the Medicaid expansion under the ACA or enacted a state Medicaid family planning eligibility expansion. Sources: Guttmacher Institute 2015e and Kaiser Family Foundation 2015.

COVERAGE OF INFERTILITY TREATMENTS: As of June 2014, states mandating that insurance companies provide coverage of infertility treatments received a score of 1.0, while states mandating that insurance companies offer policyholders coverage of infertility treatments received a score of 0.5. Louisiana, which enacted a statute that prohibits the exclusion of coverage for a medical condition that would otherwise be covered solely because it results in infertility, received a score of 0.0. Source: National Conference of State Legislatures. 2014.

SAME-SEX MARRIAGE or SECOND-PARENT ADOPTION: Whether a state recognizes same-sex marriage or allows same-sex couples the option of second-parent adoption (which occurs when a nonbiological parent in a couple adopts the child of his or her partner), or stepparent adoption through marriage: states were given 1.0 point if the state recognizes same-sex marriage or second-parent adoption statewide, 0.5 if same-sex marriage or second-parent adoption is allowed in some localities within the state, and no points if the state does not recognize same-sex marriage or allow second-parent adoption for same-sex couples in any county or the state overall. Alabama was given a score of 1.0; in February 2015, a federal district court order overturning the state’s ban on same-sex marriage took effect. The 11th Circuit Court of Appeals and the U.S. Supreme Court declined to stay that order. In response, the Alabama Supreme Court ordered probate judges not to issue marriage licenses to same-sex couples; as of April 2015, same sex couples were pursuing a class-action in federal court (Freedom to Marry 2015). Arkansas and Michigan were given scores of zero; court decisions affirming the right of same-sex couples to marry were issued in both states in 2014, but the decisions were subsequently stayed. As of April 2015, both the Arkansas and Michigan marriage cases were pending further appeal. Missouri was given a 0.5; the state has banned same-sex marriage, but that order is stayed pending appeal. In addition, a state court in Missouri also overturned the ban on same-sex marriage; as of April 2015, this ban was on appeal but had not been stayed, and same-sex couples were able to marry in some Missouri counties. In addition, Missouri has a state court order (which the Attorney General declined to appeal) that marriages same-sex couples entered into elsewhere must be recognized in the state. Sources: National Center for Lesbian Rights 2014 and 2015.

MANDATORY SEX EDUCATION: States received a score of 1.0 if they require public schools (including K-12) to provide sex education classes. Two amendments to IWPR’s 2004 Status of Women in the States report affect the discussion of trend data on this indicator: both the Guttmacher Institute (2004c) and the earlier Status of Women in the States report listed Alaska and Wyoming as having mandatory sex education in 2004; neither state, however, had mandatory sex education at that time (Elizabeth Nash, Guttmacher Institute, personal communication, April 14, 2015). Source: Guttmacher Institute 2015f.

 

Violence & Safety

This report draws on data from multiple sources that are referenced in the text, including published reports from the Centers for Disease Control and Prevention (CDC), the Bureau of Justice Statistics, and other sources to examine issues related to violence and safety among women in the United States. Much of the data are drawn from published reports from the CDC that analyze findings from the 2010 and 2011 National Intimate Partner and Sexual Violence Surveys (NISVS), a national random-digit-dial telephone survey of the noninstitutionalized U.S. English- and Spanish-speaking population aged 18 and older. Some of the tables in this report that rely on data from the 2011 NISVS are disaggregated by race and ethnicity. In this CDC report, Hispanics may be of any race or two or more races, and only whites and blacks are defined as non-Hispanic.

To highlight issues pertinent to the safety of youth in states across the nation, IWPR compiled data from the CDC’s Youth Risk Behavior Surveillance System (YRBSS), which includes a national school-based survey conducted by the CDC and state, territorial, tribal, and local surveys conducted by state, territorial, and local education and health agencies and tribal governments. The YRBSS includes both high school and middle school surveys that monitor health-risk behaviors contributing to the leading causes of death and disability among youth and adults. IWPR analyzed YRBSS data for high school students by state using the CDC’s Youth Online Interactive Data Tables for 2013, the most recent data available.

 

Work & Family

To analyze the status of women in the states, IWPR selected indicators that prior research and experience have shown illuminate issues that are integral to women’s lives and that allow for comparisons between each state and the United States as a whole. Th e data in IWPR’s Status of Women in the States reports come from federal government agencies and other sources; data in this report also rely on analysis from organizations such as AARP, Child Care Aware of America, the National Partnership for Women & Families, the National Institute for Early Education Research, and Tax Credits for Working Families.

Calculating the Composite Index

The four components of the work and family composite index—paid leave legislation, elder and dependent care, child care, and the gender gap in parents’ labor force participation rates—were each weighted equally for a total value of the index of 8. With the exception of the gap in parental labor force participation, each component includes more than one indicator; the indicators were individually scored and weighted. The paid leave, unemployment insurance, dependent care credit for adults, dependent care credit refundability, and large size of dependent care credit indicators were scored on a simple yes/no basis. For the indicator based on the costs of center-based infant care as a proportion of women’s
median annual earnings, the state with the lowest proportional costs got the highest score and was the
reference point for the other states’ scores. Th e resulting values were summed for each state to create the four component scores and the composite index score. The states were ranked from the highest score (first place) to the lowest score (last place) of the composite index. Each state’s score was compared with the ideal composite index score to determine the state’s grade.

PAID LEAVE LEGISLATION: Th is component is based on three indicators—statewide Temporary
Disability Insurance for all workers, statewide paid family care leave, and paid sick days (Gault et al. 2014; National Partnership for Women & Families 2014a and b). Th e indicators were weighted equally, with a score of 0.67 for a statewide law and a score of 0.33 for a local law (the maximum score in any leave area for a state without a statewide law, irrespective of the number of localities with separate laws, is 0.33). As a practical matter, only paid sick days have been implemented at the local as well as the state level. Th e maximum score on this component is 2.0, which only California attained. Forty states had 0.0 scores.

ELDER AND DEPENDENT CARE: This component is based on three indicators: the availability of unemployment insurance benefits to someone who had to leave their job because of family care responsibilities based on a U.S. Department of Labor (2014) report and a study of states’ unemployment insurance systems conducted by AARP (Ben-Ishai, McHugh, and Ujvari 2015); dependent care tax credits that can be applied to elder or adult dependent care expenses (Tax Credits for Working Families 2015); and nurse delegation of Longterm Support Services (LTSS; Reinhardt et al. 2014). They are each weighted equally within this index, with a maximum score of 0.67 for each of the three indicators and a maximum total of 2.0 for this component.

Unemployment Insurance is scored on a yes/no basis: a state with a law, regulation, or policy interpretation allowing benefits receives a score of 0.67; other states receive a 0.

The dependent care tax credit indicator has three subcomponents: half of the value of the indicator is given to states where dependent care credits are available for the care of dependent adults, on a yes/no basis; 25 percent
of the value of the indicator is given to states where the tax credit is refundable (yes/no basis); and another 25 percent to states where the value of the tax credit is at least $500 (yes/no basis). The maximum value of the indicator is 0.67, the weight of this indicator in the elder and dependent care component of the work and family composite index.

Nurse delegation of LTSS: 16 tasks are considered for nurse delegation; the score is determined by dividing the number of tasks delegated in a state by the total number of possible tasks to be delegated (16), to a maximum value of 1.0. Th is score is then multiplied by 0.67, the weight of this indicator in the elder and dependent care component of the work and family composite index.

CHILD CARE: Th is component is based on three indicators: the costs of infant center care as a proportion of the median annual earnings of women; the percent of four-year-olds enrolled in state Pre-K, Preschool Special Education, and State and Federal Head Start programs; and the number of quality indicators met by the state’s Pre-K programs. Th e costs of center-based infant care are based on the National Association of Child Care Resource and Referral Agencies’ (NACCRRA) January 2014 survey of Child Care Resource and Referral State Networks, and in some states it is based on the most
recently available state market rate survey (Child Care Aware of America 2014a). Median annual earnings for women who work full-time year-round were calculated based on American Community Survey data (IWPR 2015a). The percent of four-year-olds enrolled in state Pre-K, Preschool Special Education, and State and Federal Head Start programs and the number of quality measures implemented by a state’s Pre-K programs are based on the National Institute for Early Education Research (Barnett et al. 2013). Th e cost of infant care indicator has a maximum value of 0.5; the enrollment in state Pre-K, Preschool Special Education, and State and Federal Head Start programs and the quality of Pre-K indicators each have a maximum value of 0.75. The total value of this component is a maximum of 2.0.

The annual costs of infant care as a proportion of women’s median annual earnings for full-time work:
This indicator is scored by taking 1.0 minus the cost-to-earnings ratio of a state by the calculated value for the state with the best (lowest) cost-to-earnings ratio; the best state has a value of 1.0. Th e score is then multiplied by 0.5, the weight of this indicator in the child care component of the work and family composite index.

The proportion of four-year-olds in publicly funded Pre-K, Preschool Special Education, and State and Federal Head Start programs: Th e score of this indicator is the percent of four-year-old children in publicly funded programs divided by 100 percent; the maximum score of this indicator is 1.00 for 100 percent enrollment. The score is then multiplied with 0.75, the weight of this indicator in the child care component of the work and family composite index.

Th e quality of Pre-K education: Th e score of this indicator is based on NIEER’s assessment of states on
ten indicators of the quality of Pre-K provision; the score is 0 for states that do not have any programs or
practices rated by the NIEER, 0.2 if one or two criteria are met, 0.4 for three of four criteria, 0.6 for five or six criteria, 0.8 for seven or eight criteria, and 1.0 for nine or ten criteria. The score is then multiplied with 0.75, the weight of this indicator in the child care component of the work and family composite index.

GENDER GAP IN PARENTS’ LABOR FORCE PARTICIPATION RATES: Th is indicator is calculated for women and men age 16 and older with children under the age of six. To score this indicator, mothers’ participation rates (divided by 100 percent) are subtracted from fathers’ participation rates (divided by 100 percent) in each state. To give the best-performing state the highest score, a state’s differential is subtracted from 1. The score is then multiplied by 2. The total value of this component is a maximum of 2.0, if a state were to have equal labor force participation rates for mothers
and fathers. The data on labor force participation rates of parents aged 16 and older with children under age six are based on microdata analysis of the American Community Survey 2013.

COUNTING BREADWINNER MOTHERS: For the data on breadwinner mothers, IWPR analyzed American Community Survey microdata, combining three years of data (2011, 2012, and 2013) to ensure sufficient sample sizes. IWPR constructed a multi-year file by selecting the 2011, 2012, and 2013 datasets, adjusting dollar values to their 2013 equivalents using
the Consumer Price Index for All Urban Consumers and averaging the sample weights to represent the
average population during the three year period. Female breadwinners are defined as single mothers who are the main householder, irrespective of earnings or cohabitation, and as married mothers who earn at least 40 percent of the couple’s earnings. Single mothers are defined as women who are never married, divorced, separated, or widowed, or where the husband is absent. All households with children under 18 who are related to the main householder by blood, adoption, or marriage are included in the denominator for the analysis of the share of households with female breadwinner mothers. IWPR used personal weights to obtain nationally representative statistics for person-level analyses. Weights included with the IPUMS ACS for person-level data adjust for the mixed geographic sampling rates, nonresponses, and individual sampling probabilities. Estimates from IPUMS ACS samples may not be consistent with summary table ACS estimates due to the additional sampling error and the fact that, over time, the Census Bureau changes the definitions and classifications for some variables. Th e IPUMS project provides harmonized data to maximize comparability over time; updates and corrections to the microdata released by the Census Bureau and IPUMS may result in minor variation in future analyses.

 

Political Participation

Calculating the Composite Index

This composite index reflects four areas of political participation: voter registration; voter turnout; women in elected office, including state legislatures, statewide elected office, and positions in the U.S. Congress; and institutional resources available to women, including a commission for women, a campaign training for women, a women’s PAC, and a state chapter of the National Women’s Political Caucus.

To construct this composite index, each of the component indicators was standardized to remove the effects of different units of measurement for each state’s score on the resulting composite index. Each component was standardized by subtracting the mean value for all 50 states from the observed value for a state and dividing the difference by the standard deviation for the United States as a whole. The standardized scores were then given different weights. Voter registration and voter turnout were each given a weight of 1.0. The indicator for women in elected office is itself a composite reflecting different levels of office-holding and was given a weight of 4.0 (in the first two series of reports, published in 1996 and 1998, this indicator was given a weight of 3.0, but since 2000 it has been weighted at 4.0). The last component indicator, women’s institutional resources, is also a composite of scores indicating the presence or absence of each of four resources, and received a weight of 1.0. The resulting weighted, standardized values for each of the four component indicators were summed for each state to create a composite score. The states were ranked from the highest to the lowest score.

To grade the states on this composite index, values for each of the components were set at desired levels to produce an “ideal score.” Women’s voter registration and voter turnout were each set at the value of the highest state for these components; each component of the composite index for women in elected office was set as if 50 percent of elected officials were women; and scores for institutional resources for women assumed that the ideal state had each of the four resources. Each state’s score was then compared with the ideal score to determine its grade.

WOMEN’S VOTER REGISTRATION: This component indicator is the average percent (for the presidential and congressional elections of 2012 and 2010) of all women aged 18 and older (in the civilian noninstitutionalized population) who reported registering, including noncitizens who are ineligible. In 2012, 72.9 percent of U.S. citizen women aged 18 and older reported registering to vote, compared with 67.0 percent of all women aged 18 and older. IWPR selected the larger population base for this indicator because the inability of noncitizens to register accurately reflects the lack of political voice for this population. Source: U.S. Department of Commerce, Bureau of the Census 2010
and 2012, based on the Current Population Survey.

WOMEN’S VOTER TURNOUT: This component indicator is the average percent (for the presidential and congressional elections of 2012 and 2010) of all women aged 18 and older (in the civilian noninstitutionalized population) who reported voting, including noncitizens who are ineligible. In 2012, 63.7 percent of U.S. citizen women aged 18 and older reported voting, compared with 58.5 percent of all women of this age range. IWPR selected the larger population base for this indicator because the lack of voting by noncitizens accurately reflects the lack of political voice for this population. Source: U.S. Department of Commerce, Bureau of the Census 2010 and 2012, based on the Current Population Survey.

WOMEN IN ELECTED OFFICE: Th is index has four components and reflects office-holding at the state and national levels as of January 2015. For each state, the proportion of office-holders who are women was computed for four levels: state representatives; state senators; statewide elected executive officials and U.S. representatives; and U.S. senators and governors. Th e percent values were then converted to scores that ranged from 0 to 1 by dividing the observed value for each state by the highest value for all states. Th e scores were then weighted according to the degree of political influence of the position: state representatives were given a weight of 1.0, state senators were given a weight of 1.25, statewide executive elected officials (except governors) and U.S. representatives were each given a weight of 1.5, and U.S. senators and state governors were each given a weight of 1.75. The resulting weighted scores for the four components were added to yield the total score on this index for each state. Th e highest score of any state for this office-holding index is 4.58. These scores were then used to rank the states on the indicator for women in elected office. Sources: Data were compiled by IWPR from the Center for American Women and Politics 2015a, 2015b, 2015c, 2015d.

WOMEN’S INSTITUTIONAL RESOURCES: Th is index measures the number of institutional resources for women available in the state from a maximum of four, including a commission for women (established by legislation or executive order), a campaign training program for women, a women’s political action committee (PAC), and a state chapter of the National Women’s Political Caucus (NWPC). In order to score the states, each of the four components for this indicator was weighted equally at 0.5 points, for a total of 2.0 points. These scores were then used to rank the states on the indicator for resources available to women. In 2002 and 2004, the institutional resources indicator measured whether a state had a commission for women (established by legislation or executive order) and a legislative caucus for women (organized by women legislators in either or both houses of the state legislature). In earlier years (1996 and 1998) a third resource, a women’s economic agenda project, was also included in this indicator. Sources: Data were compiled by IWPR from the Center for American Women and Politics 2015k, Political and Leadership Resources for Women database; the National Conference of State Legislatures 2014c; and the National Women’s Political Caucus 2015.