CSDE Survey of W-2 Families

Codebooks
Questionnaires
Variable Cross-References

Overview

Four surveys of parents were conducted using a sample of 3000 cases selected from Cohort 1 of the CSDE project, that is, W-2 cases opened during the 10-month period from September 1997 to July 1998. The survey cases were selected randomly; they consisted of resident-mother cases from groups stratified according to:

• Prior AFDC experience, comparing those who transitioned from the old AFDC program to W-2 with those who were new to public assistance or had no recent AFDC experience.
• Experimental or Control status, in terms of the child support pass-through experiment.
• W-2 work slot placement, comparing those in grant payment slots with those in nongrant, service-only slots.

Resident W-2 fathers were not included in the survey sample. Mothers in the sample of 3000 who later left W-2 and/or became noncustodial parents were not excluded from the surveys.

After the 3000 sample cases were selected, a "focal child" was randomly selected from among the minor children listed in the case. Once the focal child was selected, an attempt was made to identify the legal father of that child (either through marriage, or through paternity establishment for a nonmarital child) in the KIDS database. In many cases, a legal father could not be identified.

After the sample of 3000 cases was selected, additional information on the characteristics of the case or the resident mother made some cases ineligible for the survey. The final number of survey sample cases was slightly fewer than 3000. These cases were not replaced.

Interviews were attempted with both the mother and the father of the focal child, in years 1999 and 2000.

A User's Guide to the Survey Data is available as a .pdf file. You will need Adobe Acrobat Reader in your computer to open this file.

Return to top of page

Time 1 Mother's Survey

The first survey was conducted in the Spring and early Summer of 1999 with resident W-2 mothers.

In some cases, the mother was no longer on W-2 and/or was no longer the custodial parent of the focal child. These mothers remained in the survey sample. In the cases in which the mother was no longer the custodial parent, an abbreviated survey was used.

A telephone survey first made repeated attempts to contact the mother. If attempts to make contact by telephone failed, personal contact and a personal interview were attempted with mothers living in the central and southern urban areas of the state.

Questions were asked about the current living situation of the mother at the time of the interview; other questions were asked about her economic status in 1998. General questions were asked about other household members, and more detailed questions were asked about the well-being of the focal child. Other questions were asked about the focal child's father.

The variables in this data file are named and identified on the basis of the survey question number.

Survey response rate: 82.1%.

Return to top of page

Time 1 Father's Survey

This survey was conducted in the late Spring and early Summer of 1999 with legal fathers of focal children who were identified as the child's father as of December 1998.

Not all focal children had a legally established father as of December 1998; for those cases there is no Time 1 Father's Survey.

In some cases, the father had become the custodial parent of the focal child. These fathers remained subjects of the survey.

Some fathers were found to be the legal father of two focal children, with two different resident W-2 mothers. In these situations the survey was conducted once, with questions about the focal child repeated. This information is presented in the data as two complete interviews, with identical information for all questions, except those relating to the focal children.

A telephone survey first made repeated attempts to contact the father. If attempts to make contact by telephone failed, personal contact and a personal interview was attempted, for one third of the sample, with fathers living in the central and southern urban a21 December, 2004

General questions were asked about other household members, contact with the focal child and the child's mother, and the economic situation of the father. Some questions were asked about the father's current situation at the time of the interview; other questions were asked about the year 1998.

The variables in this data file generally parallel the set of questions asked of the resident mothers, and are differentiated from the mother's variable names with a prefix of 'd'.

Survey response rate: complete father's sample, 33.4%; the one-third full-effort sample only, 42.9%.

Return to top of page

Time 2 Mother's Survey

This survey was conducted in the Spring and early Summer of 2000 with mothers from the original sample of 3000. An interview was attempted with the mother whether or not she had been interviewed in Time 1.

In many cases, the mother was no longer on W-2 and in some cases she was no longer the custodial parent of the focal child. These mothers remained in the survey sample. In cases in which the mother was no longer the custodial parent, an abbreviated survey was used.

A telephone survey first made repeated attempts to contact the mother. If attempts to make contact by telephone failed, personal contact and a personal interview was attempted with mothers living in the central and southern urban areas of the state.

Questions were asked about the mother's current living situation at the time of the interview; other questions were asked about her financial situation in 1999. General questions were asked about household members, and more detailed questions were asked about the well-being of the focal child. Other questions were asked about the focal child's father.

If the resident mother had been a respondent at Time 1, then several questions were not repeated. For information on these questions, the user must retrieve the information from Time 1 data. Some questions were dropped at Time 2, other questions were added, and some questions were reworded.

The variables in this data file generally parallel questions asked at Time 1, and are differentiated from Time 1 questions by a prefix of 't2'. If a question was reworded, the variable name at Time 2 has the prefix of 't2z'.

Survey response rate: 82.0%.

Return to top of page

Time 2 Father's Survey

This survey was conducted in the late Spring and early Summer of 2000 with legal fathers of focal children, identified as the father as of December 1999.

In some cases, the father had become the custodial parent of the focal child. These fathers remained subjects of the survey.

Not all focal children had a legally established father as of December 1999; for those cases, there is no Time 2 Father's Survey.

In some cases paternity of the focal child had not been legally established until 1999. In these cases there could be a Time 2 Father's Survey, but no Time 1 Father's Survey.

Some fathers were found to be the legal father of two focal children, with two different resident mothers. In these cases the survey was conducted once, with questions about the focal child repeated. This information is presented in the data as two complete interviews, with identical information for all questions except those relating to the focal children.

A telephone survey first made repeated attempts to contact the father. If attempts to make contact by telephone failed, personal contact and a personal interview was attempted, for one third of the sample, with fathers living in the central and southern urban areas of the state (full-effort sample).

Questions were asked about household members, contact with the focal child and the child's mother, and the economic situation of the father. Some questions were asked about the father's current situation at the time of the interview; other questions were asked about his financial situation in 1999.

The variables in this data file generally parallel the set of questions asked of the resident mothers, and are differentiated from the mother's data with a prefix of 'dt2'. If a question was reworded in Time 2, then the prefix is 'dt2z'.

Survey response rate: complete father's sample, 32.8%; the one-third full-effort sample only, 46.3%.

Merging data from the four surveys

For a longitudinal analysis of mother's or father's data, Time 1 and Time 2 data can be merged, using the survey sample identification number.

For an analysis of matched pairs of parents, at either Time 1 or Time 2, the Mother's and Father's Survey responses can also be merged.

Sample SAS programs are available to assist the user in:

*reading the Time 1 Mother's Survey Data
*reading the Time 1 Father's Survey Data
*reading the Time 2 Mother's Survey Data
*reading the Time 2 Father's Survey Data
*merging the Time 1 and Time 2 Mother's Survey Data
*merging the Time 1 and Time 2 Father's Survey Data
*merging the Time 1 Mother's and Father's Survey Data
*merging the Time 2 Mother's and Father's Survey Data

Return to top of page