This appendix provides information on the sources of data used in this publication, and the limitations of each source. The major surveys used in this publication are:
These surveys provide the most comprehensive, current estimates from large, nationally-representative samples. They are presented first in this appendix.
For other topics in the area of women and disability, smaller or more specialized surveys and databases provide invaluable information. Other surveys and databases used in this publication include, in alphabetical order:
The following summaries describe the surveys, their sampling formats, the size of the respondent bases, and definitions of terms used in the surveys, including how disability is measured. Information is also presented on databases that served as data sources for this publication. More details can be found in the original publications.
The Survey of Income and Program Participation (SIPP) is a multi-panel, longitudinal survey conducted by the U.S. Census Bureau. The SIPP covers the civilian, non-institutionalized population of residents living in the United States, and collects data on source and amount of income, labor force information, program participation and eligibility data, and general demographic characteristics. The SIPP also includes disability supplements that ask questions to determine individuals' disability status. Historical background and more detailed information on the SIPP can be found on the Internet at <http://www.sipp.census.gov/sipp>.
Survey design and sampling -- The data in this publication are based on a number of overlapping waves and panels of the SIPP. The survey design is a continuous series of national panels in which the same households are interviewed every four months for periods ranging from 2 1/2 to 4 years. A cycle of four interviews covering the entire sample and using the same questionnaire is called a wave. Interviews are conducted by personal visit and by follow-up telephone calls. All household members who are 15 years and older are interviewed, if possible, and proxy response is permitted when individuals are not available for interviewing.
Respondents -- Sample size ranged from approximately 34,000-40,000 households per data file. (The 1991-92 and 1994-95 SIPP analyses each constitute a data file.) A rough estimate of the number of individuals interviewed per file would be 85,000 to 100,000, based on an estimated average of 2.5 individuals per household.
Definitions -- The questions that have been asked in the disability supplements of the SIPP were designed to be consistent with the definition of disability set forth in the Americans with Disabilities Act (ADA). People 15 years of age and older were considered to have a disability if they met the following criteria:
Functional limitations are defined from the questions asked about the difficulty in performing basic activities such as seeing, hearing, having one's speech understood, walking, carrying or lifting 10 pounds or walking up a flight of stairs. Activities of daily living (ADLs) covered in the survey include getting around inside the home, getting in and out of bed or a chair, bathing, dressing, eating and toileting. Instrumental activities of daily living (IADLs) covered in the survey include going outside the home, keeping track of money or bills, preparing meals, doing light housework and using the telephone.
The National Health Interview Survey (NHIS) is a principal source of information on the health of the civilian non-institutionalized population of the United States. It has been conducted annually by the National Center for Health Statistics (NCHS) since 1957. Each year, the survey consists of a basic set of questions on health, socioeconomic and demographic items as well as one or more special questionnaires to obtain more detailed information on major current health issues. A special questionnaire on mental health was conducted in 1989 and a special questionnaire on disability, the NHIS-D, was conducted in 1994-95. The NHIS also provides information about activity limitations and chronic conditions.
Survey design and sampling -- The NHIS is conducted according to a multistage probability design, permitting continuous sampling of the civilian non-institutionalized population living in the United States. Each weekly sample is representative of the target population and is additive with other weekly samples. Sampling is done throughout the year, preventing seasonal bias. Information is obtained about health and other characteristics of each member of the household. The usual sample size is approximately 48,000 households or about 122,000 individuals each year.
Respondents -- The interviewed sample for 1994 was 116,179 individuals (45,705 households) and for 1995 was 102,467 individuals (39,239 households). Response rates were 94.1 percent in 1994, and 93.8% in 1995.
Definitions -- The NHIS defines chronic condition as one that has lasted for three months or more, or one that is on the NCHS list of chronic conditions regardless of onset. Disability refers to the state of being limited, due to a chronic mental or physical health condition, in the type or amount of activities. The NHIS has three measures of disability: (1) activity limitation in major activity, (2) work limitation, and (3) need for personal assistance with activities of daily living (ADLs).
National Health Interview Survey Supplement on Mental Health (NHIS-MH): In 1989, a special supplement on mental health was included as part of the NHIS. The purposes of the supplement were to update previous estimates of the prevalence of serious mental illness and to examine the use of mental health services and disability program participation of the population with serious mental illness. Information was collected on 113,231 people, a response rate of 96.8% of those who were interviewed on the basic NHIS questionnaire that year or 91.9% of the total NHIS sample for the year. Given that mental illness often carries great stigma and that the survey was not designed to diagnose mental disorders, the data from this survey are likely to underestimate the actual prevalence of mental disorders.
National Health Interview Survey on Disability (NHIS-D): In 1994-95, a special questionnaire on disability, the NHIS-D, was included as a topical module on the National Health Interview Survey. The survey was conducted in two phases. Phase I of the NHIS-D was designed to identify a wide range of children and adults with chronic conditions, impairments, disabilities and elevated service needs. Information was collected on 107,469 individuals, of whom 30,032 were children. The response rate for the NHIS-D, Phase I, was 87%. Phase II of the NHIS-D was designed as a follow-up questionnaire with individuals who were identified in Phase I. Data from Phase II is not yet generally available. See functional activity and functional disability in the Glossary for definitions of those terms as used in the NHIS-D.
The Current Population Survey (CPS) is a monthly survey conducted by the U.S. Census Bureau that collects labor force data on the non-institutionalized civilian population. The data presented in this publication are from the March Income Supplement, in which questions related to labor force participation and income are asked of all members of the household 16 years of age and older.
Survey design and sampling -- The sample of approximately 60,000 is selected to be representative of the entire population of the United States, and numbers are "weighted" or adjusted to independent population estimates based on the results of the decennial Census. These weights take into account age, gender, sex, race, Hispanic origin and state of residence.
Respondents -- About 60,000 households were eligible to participate in the survey, representing about 1 in every 1,600 households in the country.
Definitions -- Work disability is the only disability measured by the CPS. People are classified as having a work disability if they:
The Decennial Census ("The Census") has been conducted every 10 years since 1790. Data in this publication are based on the 1990 census. The Census seeks to simultaneously enumerate all individuals in the United States. Census forms are mailed out and interviewers dispatched to residential addresses, and the Census questionnaire includes questions about both the condition of the housing unit and its occupants.
The "short form" of the 1990 Census asked seven population and seven housing questions; it was intended to be completed by all households (100% sample). The "long form" of the Census contained all of the short form questions and many other questions and was distributed to a sample of the United States population. Questions about disability are contained in the long form of the Census. Data from the decennial Census are used to weight the sample results of the SIPP and the CPS on age, sex, race, and Hispanic/non-Hispanic categories.
Survey design and sampling -- The short form of the decennial Census, which is used to weight other surveys, does not sample from the population, but rather seeks to enumerate the entire population of the United States. About 106 million housing units received the short form. The long form of the Census, which includes disability questions, was distributed to 17.7 million housing units. The Census samples according geographical areas, and smaller, less populated areas are sampled more heavily than densely populated urban areas. The Census also conducts a Post-Enumeration Survey which is used to estimate the degree to which the Census over- or under-counts the population.
Respondents -- The total population count of the 1990 Census was 247.8 million people. Researchers estimate that the Census undercounted the population by 4 - 5.3 million people, based on the results of the Post-Enumeration Survey (Hogan, 1993). The under-count is proportionally larger for certain groups of people (particularly ethnic minorities, and people with low income or low educational level) and for certain geographical areas (Barrett, 1994).
Definitions -- Disability, according to the 1990 Census, is defined by answers to the following questions. The work disability question is the same as in the 1980 Census. People were asked whether they had a physical, mental or other health condition that limited the amount or kind of work they could do, prevented them from working, or prevented them from using public transportation. In 1990, a new question about functional disability asked whether any person in the household had a health condition that limited the ability to go outside the home alone or caused difficulty in taking care of personal needs such as bathing, dressing, or getting around inside the home (Barrett, 1994).
Epidemiological Catchment Area Survey (ECA) was a multi-site, epidemiological and health services research study that assessed the prevalence and incidence of mental disorders, as well as use of mental health services. There were five sites for the study, and five universities that carried out the research: New Haven, CT (Yale); Baltimore, MD (Johns Hopkins); St. Louis, MO (Washington University); Durham, NC (Duke); and Los Angeles, CA (UCLA).
Survey design and sampling -- Population sizes in the five sites ranged from 270,000 to 420,000 with different mixes of urban, rural and suburban locations as well as ethnic and age compositions. In each area, a probability sample of households was selected, and one adult, age 18 or over, was interviewed about the people in the household. The following groups were oversampled: the elderly in New Haven and Durham, African-Americans in St. Louis, and Hispanic people in Los Angeles.
Respondents -- A total of 18,571 people were interviewed, with a range of 3,004 to 5,034 completed interviews at each site.
Definitions -- The ECA diagnosed mental disorders according to the diagnostic criteria outlined in the Diagnostic and Statistical Manual of Mental Disorders Third Edition (DSM-III) of the American Psychological Association.
The National Alliance for Caregiving Survey was designed as a telephone survey and conducted between August and September of 1996. A total of 1,509 English-speaking family caregivers participated in interviews. Samples were generated by randomly calling U.S. households to identify caregivers. Blacks, Hispanics and Asian caregivers were oversampled to ensure representation proportional to the ethnic mix in U.S. telephone households. Given the relatively small sample size, there may be limitations to generalizing to the U.S. population. Further information on the survey and sampling issues is available in the report of the National Alliance for Caregiving (1997).
National Comorbidity Survey (NCS): Mandated by Congress and conducted in 1990-92, the NCS was the first survey to administer a structured psychiatric interview to a nationally representative sample. The survey was designed to study the prevalence and correlates of psychiatric disorders and the use of services by people with those disorders. The study also focused on the comorbidity of substance use disorders and non-substance psychiatric disorders.
Survey design, sampling, and respondents -- A household survey of over 8,000 respondents, ages 15 to 54 years, was conducted in 48 states, based on a stratified, multistage probability sample. The study also included other supplemental samples.
Definitions -- Psychiatric diagnoses were assigned based on a structured psychiatric interview (a modified version of the Composite International Diagnostic Interview, which provides diagnoses according to the Diagnostic and Statistical Manual of Mental Disorders -- Third Edition, Revised, DSM-III-R, of the American Psychological Association). The study did not measure disability, but rather the presence or absence of psychiatric disorders.
The National Hospital Discharge Survey (NHDS) is a continuing nationwide survey of hospitals, focusing on short-stays rather than long-term care. On the basis of patient records, the survey estimates the causes of illness and injury nationwide.
Survey design and sampling -- Hospitals are selected to have patient records abstracted and analyzed. The survey uses a stratified sampling methodology, based on hospital size, number of discharges, primary diagnoses, patient age and gender group, and other factors. Statistical adjustments are made for non-responding hospitals and missing abstracts.
Respondents -- In 1995, 525 hospitals were selected, 512 met the inclusion criteria, 466 participated, and 263,000 medical records were abstracted.
Definitions -- Among other purposes, the survey provides national estimates on causes of short-term disabilities that result from illness and injury.
The National Medical Expenditures Survey (NMES) is a national probability sample of households done for the Agency for Health Care Policy and Research (AHCPR). It surveyed the civilian, non-institutionalized population of the United States, as well as medical providers, about medical service use and expenditures, sources of payments for health care, and health insurance coverage. It measured disability at two levels: activity limitations and activities of daily living (ADL) limitations.
Sampling -- The NMES has a stratified multistage area probability sample design. The survey oversamples for poor and low-income families, elderly people, individuals with functional limitations, Blacks, and Hispanics. Participants were interviewed five times between February 1987 and July 1988.
Respondents -- A screening interview was conducted in 35,600 households. The resulting sample consists of approximately 35,000 people in 14,000 households.
Definitions -- Activity limitation is defined by age group. Adults 18 years and over are considered to have an activity limitation if they answer positively to either of the following questions: 1) Does your health keep you from working at a job, doing work around the house, or going to school? and 2) Are you unable to do certain kinds or amounts of work, housework, or school work because of your health? For school-aged children between the ages of 5 and 17, activity limitations are defined by questions about attending or needing special schools or classes, and limitations in school or other activities because of an impairment or health problem. For young children under 5 years, activity limitation is defined by questions about limitation in play or other activities, due to an impairment or health problem.
National Nursing Home Survey (NNHS): The National Center for Health Statistics has conducted four national nursing home surveys, beginning in 1973. The estimates in this publication of the number of elderly people in nursing homes were based on the results of the most recent (1995) NNHS.
Sampling -- The 1995 survey sampled only nursing homes that provided some level of nursing care, excluding homes that provided only personal care or housing. The sample of 1,500 homes was selected from a sampling frame of 17,500 nursing homes contained in the National Health Provider Inventory.
Respondents -- Information on 8,000 current residents was obtained from approximately 1,400 responding nursing homes. Staff members who were familiar with resident care completed forms, and information was also abstracted from medical records.
Definitions -- For the purpose of this publication, the survey provided estimates of the number of elderly people living in nursing homes.
The National Study of Women with Physical Disabilities was a mail survey of 860 women, 439 with physical disabilities and 421 without physical disabilities. Participants with disabilities were between the ages of 18 and 65 years, had a self-reported physical disability resulting in functional limitations and had no known cognitive impairments, mental health problems or difficulty reading English. They were recruited through independent living centers and announcements in disability publications. These women gave parallel questionnaires (minus the disability questions) to non-disabled friends, so that comparisons could be made between women with and without disabilities. Although the survey was not designed to be nationally representative, it is the first study to survey a relatively large number of women with and without disabilities about their experiences with abuse.
Rehabilitation Services Administration Program Data is collected by the Rehabilitation Services Administration (RSA) from all the states and compiled annually. Information used in this chartbook was made available by RSA from unpublished data analyses.
The Rochester Epidemiology Project (REP) is not a survey but rather an archive of medical records that has been kept since 1907 at the Mayo Clinic in Rochester, MN. This population-based medical database is particularly useful for studying the incidence and prevalence of diseases over relatively long periods of time and for studying the relationships among different diseases. The archives are also helpful in comparing costs of medical treatment for people with a given disease, compared to people who do not have that disease. The REP currently supports nearly 1,000 research projects and provides data to hundreds of other studies each year.
Social Security Administration Program Data is collected by the Social Security Administration (SSA) and published each year in the Annual Statistical Supplement to the Social Security Bulletin. A substantial number of the tables in this publication are based on sampling 1-percent or 10-percent of the files from the administrative records. Due to sampling error, some of these estimates may be different from the results that would have been obtained if all the records had been used. Detailed information on sampling variability may be obtained in the Annual Supplement. (See also the Glossary for definitions of the following terms that are used in the Social Security Administration Program Data: Disabled worker; Old Age, Survivors and Disability Insurance (OASDI); Social Security benefitsp; Social Security Disability Insurance (SSDI); and Supplemental Security Income (SSI).)
U.S. Department of Education, Office of Special Education, Data Analysis System: This data system is the source of program data on special education in the United States. Data is collected from all the states on children and youth, ages 0 to 21 years, who are served under IDEA, the Individuals with Disabilities Education Act. Analyses of these program data are published each year in an annual report to Congress; the Twentieth Annual Report (U.S. Department of Education, Office of Special Education and Rehabilitation, 1998) is the most recent such report. Program data are also published annually in the Condition of Education (U.S. Department of Education, National Center for Educational Statistics, 1997, 1998).