INTRODUCTION to the VIETNAM LONGITUDINAL SURVEY

Background and Fieldwork

The Vietnam Longitudinal Survey is one of the largest sociological surveys ever conducted in Viet Nam and is the first longitudinal survey, undertaken in Vietnam, to be conducted in the field of sociology. The VLS is part of a long-term collaborative program of research collaboration between the Institute of Sociology and Professor Charles Hirschman from the University of Washington-Seattle. The fieldwork costs are supported by a grant from the National Institute of Child Health and Human Development (from Hirschman at the University of Washington), but the project has also received substantial support and leadership from the Institute of Sociology.

The aims of the VLS project are to: (1) design an area probability sample survey to collect high quality demographic and social data for a representative sample of households and adults in one region of Vietnam, (2) collect baseline data for 1,855 households and all resident adults (ages 15 to 65) in the household with detailed questionnaires on social and economic activities (past and current) and demographic behavior, (3) track all respondents and re-interview them with focused questionnaires annually for five years, (4) construct a clean and organized data file for release to the scholarly community in Vietnam and elsewhere, (5) analyze the impact of the changing household economy on demographic phenomena in Vietnam, and (6) improve professional skills of the staff of the IOS in data collection and analyses.

The idea of conducting this research project emerged as a result of extensive discussion and scholarly exchange between IOS researchers and Charles Hirschman following our first collaborative project, The Vietnam Life History Survey, which was conducted in 1990-91. During the 1994-95 academic year we jointly developed a detailed plan for the VLS based on our experience from the Vietnam Life History Survey. The preparation of questionnaires, the sampling methodology, and other logistical aspects of the project were completed by the early summer 1995. Ten communes in provinces Ha Nam, Nam Dinh, and Ninh Binh were selected for the survey used probability sampling methods. In July 1995, the pretest survey for the VLS was carried out in Dai Xuyen commune which is about 40km south of Hanoi. The principal investigators (Professor Charles Hirschman, Professor Tuong Lai, Professor Pham Bich San) and researchers of the project from the Institute of Sociology directly participated in the pretest. The pretest had helped to refine the survey instruments.

The baseline survey was conducted in the period from September to November 1995 with 1,855 households and 4,464 individuals were interviewed in the first round of the survey. In addition, a large number of in-depth interviews and focus group discussions were conducted. The research team also collected a huge amount of documentation related to socioeconomic and demographic activities of the communes in the sample. The second round was carried out in August-September 1996 with 1,820 households and 4,340 individuals successfully re-interviewed. The third round was carried out in July-August 1997, with 1811 households and 4309 individuals were successfully re-interviewed.
 

Codebooks

Codebooks for the baseline and three follow-up surveys can be accessed and downloaded. These documents describe the contents of the machine-readable data files of the 1995-98 VLS surveys. There are two parts. Part I contains summary lists of variables for 1995 Household and Individual files and 1996 Household and Individual files. Part II contains detailed list of the codes and additional information for each variable. In addition, it is possible to access and download the bilingual questionnaires utilized in the data collection process each year. Finally, a set of appendices exists which provides miscellaneous survey support materials.

In Part I (the List of Variables section) of the codebooks, each variable is given a unique number along with several identifiers. These identifiers include the mnemonic used in the SPSS system file (VARNAME), a short description, the length (number of characters) of the variable, and a reference to the question number in the original questionnaires. A few questions in the printed questionnaire were not coded and thus do not appear as variables in the data files. In the household file, variables were created up for up to 15 persons, however, most of these variables are blank because most households had only 4 to 6 persons. Similarly, the sibling roster and the children table in the individual questionnaires contained more lines than the actual number of siblings/or children of the respondent, and most of the associated variables are blank.

In Part II of the codebooks, the values and associated labels for each variable are presented unless the list would have been very long and redundant (e.g., ages or years). For many variables, there are explanatory notes that explain measurement problems. All cases of "not applicable" were coded as 8, 98, 998, or 9998 depending on the width of the field for the variable in the data file. The same is applied for all "don't knows" responses which were coded as 9, 99, 999, or 9999. The variable fields for persons, siblings, or years that were not relevant for a household or person are left blank.

Regarding the bilingual questionnaires, two questionnaires, an individual and household questionnaire, are available for each year that the survey was conducted. Please note that there are slight differences in the content of questions asked in the surveys across the 4 years of data collection.  All of the questionnaires are displayed in Vietnamese, with an English translation attached to each item in the questionnaires.

The Vietnamese font used in the questionnaires for the 1995 round is from VNKey 1.0, and the Vietnamese font used in the questionnaires for the 1996 round is from Vietware 2.0. Appendices include a series of documents used in the fieldwork and in the coding process. These are Special data collection and coding, Instructions to interviewers, Frequencies of selected variables, Sampling frame and sampling design, 1989 Census data for the ten communes, and the occupational titles and codes. In the original survey, interviewers were instructed to write in the exact occupational titles as reported by the respondents. During the coding process in Seattle, all occupational titles were translated into English and assigned ISCO codes from the 1968 version of the International Standard Classification of Occupations. Appendix 6 contains the three digit ISCO occupational classification and every unique Vietnamese language occupational title (as reported by respondents) and the assigned ISCO code. There are many titles (articulated by Vietnamese respondents) that share the same ISCO code.
 

Data File Administration

The most recent versions of the SPSS system files of the household and individual datafiles are located on \\web.csde.washington.edu\vietnam on CSDE web server system. Suggestions for corrections in the codebook or errors in the data should be sent to:

Charles Hirschman
Department of Sociology,
202 Savery, Box 353340
University of Washington
Seattle, WA 98195-3340
Telephone: (206) 543-5035/ 543-5882
Fax: (206) 543-2516
email: charles@u.washington.edu

GO BACK TO VLS MAIN PAGE
GO TO VIETNAM LIFE HISTORY SURVEY 1991