logo5.gif (16944 bytes)

Methodology

[Home] [Participants] [Investigators] [Publications] [Methodology]

This section reviews details about the sampling procedures, the demographic characteristics of the sample, the data collection and measures used in the Children in the Community study. Click here to view the manual for the Transitions Study.

Sampling Procedures, Retrieval Rates, and Maintenance

The Original Sample                                      Where do they live now? See a map.

During the early and middle 70's, Dr. Leonard Kogan, Director of the Center for Social Research at City University of New York, and Dr. Shirley Jenkins of Columbia University School of Social Work co-directed a national study designed to develop indicators of the physical health and the social, emotional, and cognitive functioning of children in census-based geographic areas. As an outcome of that study, an index based on census and health data was developed which was labeled Disorganized Poverty or DIPOV. The letters in DIPOV form an acronym based on the initial letters of the five indicators: D for Dependency (proportion of children under 18 in families receiving Aid to Families with Dependent Children); I for Incomplete Families (proportion of children under 18 not living with both parents); P for Premature Births (proportion of live births weighing under 2501 grams); O for Out-of-Wedlock births (proportion of live births born to unmarried mothers); and V for Venereal Disease rate among persons under age 20.

The DIPOV index was assumed to reflect a larger set of needs and problems of children upon which health and other service planning could be based. However, it was clearly seen as important that this assumption be tested empirically. For this reason, a study was planned which would survey families with children in counties that differ markedly on the DIPOV index.

1975 Site Selection and Sampling Procedures

The DIPOV indices for the 62 counties of New York state in 1970, 1971, and 1972 were examined with the purpose of selecting two counties for study, one with a high DIPOV value and one with a low value. The five counties of New York City were arbitrarily excluded as atypical of the national scene. Albany County was identified on the DIPOV scale as one of the worst counties and adjacent Saratoga county as one of the best, and these were chosen for study by means of a sample survey (Kogan, Smith, & Jenkins, 1977).

Primary Sampling Units were created from Enumeration Districts and Block Groups, divisions defined by the Census Bureau, and which when taken together comprise the entire area and population of the target counties. Enumeration Districts average about 250 housing units, and Block Groups are combinations of contiguous blocks having a combined population of about 1,000.

Population data from the 1970 census for these Primary Sampling Units were updated for 1975 after consultation with local officials, and adjustments were made for new residential construction. After correction, the Primary Sampling Units in each county were stratified by urban-rural status, proportion white, and median income. A systematic sample of Primary Sampling Units in each county was then drawn with probability proportional to the number of households, and probabilities equal for members of all strata.

Each selected Primary Sampling Unit was subdivided for a second-stage sample. Segments were constructed in Enumeration Districts by the use of aerial photographs and survey maps, and block divisions with Block Groups were obtained from census publications. Segments of blocks were then selected with probability proportional to size (number of households), and each was surveyed in the field with a proportion of the households being selected according to the predetermined sampling ratio. Address lists were compiled in this process, and interviewers were sent to the selected addresses. Those households with at least one child between the ages of 1 and 10 years were qualified for the study and, when possible, an interview was obtained. In each qualified household, the interviewer, by use of a set of Kish tables, randomly selected one child from those in the appropriate age range.

In Albany County, 2,252 households were approached, but 1,749 of these did not contain a child within the study age range. Completed interviews were obtained from 424 of the 503 qualified households, yielding a response rate of 84.3%. In Saratoga County, 2,014 households were screened, but 1,376 were not qualified. The response rate for qualified households was 86.5% (n = 552). The fieldwork for this study required about 6 months, from January to July 1975.

 

1983, 1985-86, 1991-93, and 1995-97 Follow-up Location and Reinterview Rates

During the first follow-up of this sample, done eight years after the original survey, 827 (85%) of the original sample was relocated and 724 families were successfully reinterviewed in their homes. At this time located families were living in 26 different states, and in four foreign countries (we did not attempt to interview families living beyond the 48 contiguous states). Failures to reinterview included the death of the child (2 cases), absolute refusal of the family (4 cases), and, for the most part, difficulties in finding times when both the families and the interviewers could be available in these diverse locations. During the third and fourth waves of interviews we again succeeded in interviewing 724 of the original families, excluding two youths with Down's syndrome whom we have decided not to try to interview in the future, and adding 46 families for whom interviews could not be scheduled in the first follow-up. Altogether 770 of the original 976 families (79%) have been interviewed in the follow-up waves, some 93% of those located and living in this country. In the fifth wave of data collection 751 members of the sample, now young adults, were interviewed.

 

The Sample Supplement of Young Urban Poverty Families

We were not surprised to learn that families who were lost to retrieval from the original sample were not a random group. Families lost to follow-up were poorer, more urban, and, particularly, younger than those located. Children and families also differed significantly on several critical independent and dependent variables. The original data collection did not include the last names of families, since no follow-up was then anticipated. Therefore locating families depended heavily on residential stability, being known to neighbors, having relatives in the area, home ownership, or having an identifiable profession. Most of these are less characteristic of families with children in the 1 to 3 year age group, of those in urban areas, and the poor. For many of these families we could not even determine the last name of the people who had lived at that address 8 years earlier, and in some cases whole neighborhoods had been razed for urban development. Positive identification of families once located was easy, because the study had recorded first names and birthdays of all family members.

In order to improve the representativeness of the study sample on these variables, the William T. Grant Foundation provided funds to enable us to draw a new sample of children in the youngest follow-up age group from urban poverty areas. The sampling procedures were the same as in the original study. After obtaining a listing of each census tract in Albany county on median family income, the proportion of families with children with incomes below the poverty level, and an estimate of the proportion of households including eligible children, 27 urban block groups were identified with a) median family income less than $15,000 and b) more than 20% of families with children below the poverty line. Seven areas were estimated to have fewer than 18% of households with eligible children and were eliminated as available funds did not permit screening a higher number of households per identified eligible family. Eleven eligible areas were identified, one was eliminated as it had been a sampled area for the primary sample. Three areas were randomly selected from these ten, a fourth area was subsequently substituted when one of the selected areas was found to have been depopulated of nearly all eligible children by urban renewal.

Every household in selected tracts was listed. Households were then randomly sampled at a sampling ratio designed on the basis of the proportion of households with eligible children to yield 20 children per area. Each randomly selected household was then independently screened to determine the presence of an eligible child. The independence of this operation was considered essential to avoid omission of rundown or dangerous appearing residences. Screening was done by pairs of study personnel. 661 households were screened and interviews were obtained on 54 of the 82 eligible families (77%). 52 (96%) of these families also participated in the 1986 interviews.

The 1983 sample including this supplement is representative of the original counties with regard to race (92% white as compared to 93% white for the population in 1980). A comparison with families including children in the relevant age ranges from the 1980 survey of the U.S. Bureau of the Census shows a very close match on family income and on maternal education, especially when one takes into account 3 years of salary inflation (see Figures 1.1 and 1.2). Family structure also matched census figures reasonably well, with 75% of our children living with married parents, and 19% living with a mother who was not currently married; the census figures were 79% and 17%, respectively. 2% of the sample and 3% of census children of that age were living with a currently unmarried father.

 

The Sample of Siblings

In 1987 we obtained funding from NIMH for a special sampling of siblings of study youth. Selected siblings were those closest in age to the study children who were between the ages 12 and 18 and whose families lived in the larger New York State area that included the original two counties. Siblings and mothers were interviewed in 180 families.