Newsletter

February 1994 - Number 1

About the FLS Newsletter

What is the FLS Newsletter?

The FLS Newsletter is a free occasional publication of the Center for the Study of the Family in Economic Development at RAND. The newsletter will serve as a mechanism for disseminating information about surveys fielded by RAND in developing countries. These country surveys cover Malaysia (released to the public), Indonesia (fieldwork stage), Guatemala (development stage), and Bangladesh (development stage). This issue focuses on the Malaysian Family Life Surveys (MFLS-1, 1976-1977 and MFLS-2, 1988-1989). The FLS Newsletter will provide data updates and clarifications, and a bibliography of research based on the MFLS data. These topic areas will be regular sections of the newsletter and will incorporate equivalent entries from other surveys as they become publicly available. Future issues will discuss the other surveys in greater detail.

Who gets the FLS Newsletter?

The FLS Newsletter is meant to serve all FLS data users and interested parties. The newsletter's initial mailing list consists of individuals and institutions who have ordered the Second Malaysian Family Life Survey (MFLS-2) data, who have expressed an interest in those data, or who are on the mailing list for RAND Labor and Population's Working Paper Series.

To receive subsequent issues of the FLS Newsletter and/or RAND Labor and Population Working Papers, please fill out the enclosed form and return it to RAND. To make the process as painless as possible, postage and the return address have been preprinted on the subscription form. If you know of MFLS data users or institutions who have not received a copy of this initial newsletter, please encourage them to contact us so we can add them to our mailing list.

In this first issue, we have included an announcement letter for the FLS Newsletter as an insert. Please post this announcement in an appropriate place so that other people will see it and contact RAND. An invitation to subscribe to the FLS Newsletter will be reissued when new databases become available.

What databases will be covered by the FLS Newsletter?

The FLS Newsletter will cover public-release versions of developing-country survey data collected by projects and centers within the RAND Labor and Population. There are currently two available databases, MFLS-1 and MFLS-2. Other survey data currently in the field or in the planning stages will be covered in detail as those survey data are collected and then released to the general demographic research community.

The Indonesian Family Life Survey (IFLS), which is currently in the field, is surveying approximately 7,000 households across 13 provinces in Indonesia. Lembaga Demografi at the University of Indonesia is carrying out the fieldwork. Public release is projected for late 1995.

Work will soon begin on two additional surveys, which will be fielded in 1995 and which are projected for release in 1997. One will collect data on 6,000 households in the Matlab area of Bangladesh. The other will collect data on 3,000 households in 60 rural communities in Guatemala. Those desiring more information about these three surveys should contact the survey project leaders at RAND: Paul Gertler (IFLS), Omar Rahman (Bangladesh), and Anne Pebley (Guatemala). Future issues of the FLS Newsletter will include updates on the progress of each of these surveys with respect to development, fieldwork, public release, user support, and subsequent research.

About the MFLS Data and Documentation

What are the MFLSs?

The MFLSs are a pair of surveys with partially overlapping samples designed by RAND and administered in Peninsular Malaysia in 1976-1977 (MFLS-1) and 1988-1989 (MFLS-2). Fieldwork for MFLS-1 was carried out by Survey Research Malaysia, Sdn. Bhd., and for MFLS-2 by the National Population and Family Development Board of Malaysia. Each survey collected detailed current and retrospective information on family structure, fertility, economic status, education/training, transfers, migration, and many other topics.

MFLS-1 data description

MFLS-1 was fielded in 1976-1977. The survey collected current and life history data on a representative sample of 1,262 households in Peninsular Malaysia that contained an ever-married woman (EMW) aged 50 or less. Both the EMW and her current husband were interviewed. 1,194 women were married at the time of MFLS-1, with 1,136 husbands present of whom 1,048 were interviewed. Community data were gathered for the 52 primary sampling units (PSUs) that made up the MFLS-1 enumeration areas.

MFLS-2 data description

MFLS-2 was fielded in 1988-1989 and collected current data and full life history data for four samples:

  1. The PANEL Sample - Women who were the primary respondents to the MFLS-1 in 1976-1977. Of the 1,262 MFLS-1 households, 926 were reinterviewed in MFLS-2, a follow-up rate of 72 percent. Within these Panel households, 889 women completed the female life history questionnaire. Interviews were also completed with 728 current husbands of Panel women; 644 of these husbands had been interviewed in MFLS-1.
  2. The CHILDREN Sample - Children aged 18 or older in 1988 of the women interviewed as primary respondents for MFLS-1. Interviews were conducted with one child, selected at random, inside the Panel household and two children, selected at random, living elsewhere in Peninsular Malaysia. There are 1,136 respondents in the Children Sample, of whom 1,096 completed the life history questionnaires. Among the 528 married Children Sample respondents, 494 spouses were also interviewed.
  3. The NEW Sample - A new sample of households with a woman aged 18-49 (regardless of her marital status) or an ever-married woman under age 18. There are 2,184 respondents in the MFLS-2 New Sample, 1,767 of whom are currently married. Interviews were also completed for 1,642 spouses.
  4. The SENIOR Sample - A new sample of households with a person (male or female) aged 50 or older. There are 1,357 respondents in the Senior Sample. Spouses were not interviewed in this sample.

Households for the New and Senior samples were located in 398 enumeration blocks (EBs) in Peninsular Malaysia, chosen with probability proportional to size. Within Peninsular Malaysia, there are three main ethnic groups--Malays, who make up about 55 percent of the population; Chinese, who make up 30 percent; and Indians, who make up 10 percent. In each EB, Indians were oversampled to provide sufficient sample sizes for analyses within each ethnic group. Many households are covered by both the New and Senior samples (e.g., a household that contains a woman aged 30 and her father aged 60).

MFLS-2 collected community-level data for the 398 EBs covered in the New and Senior samples and for the original 52 PSUs covered in MFLS-1.

MFLS documentation

Below is a list of the documentation that accompanies the MFLS-1 and MFLS-2 databases. An order form is provided for each document.

MFLS-1 documentation

R-2351-AID, The Malaysian Family Life Survey: Summary Report, William P. Butz and Julie DaVanzo, RAND, 1978.

R-2351/1-AID, The Malaysian Family Life Survey: Appendix A, Questionnaires and Interviewer Instructions, William P. Butz, Julie DaVanzo, Dorothy Z. Fernandez, Robert Jones, and Nyle Spoelstra, RAND, 1978.

R-2351/3-AID, The Malaysian Family Life Survey: Appendix C, Field and Technical Report, Robert Jones and Nyle Spoelstra, RAND, 1978.

R-2351/4-AID, The Malaysian Family Life Survey: Appendix D, Description of Sample Communities, Fahmi Omar, RAND, 1978.

R-2351/5-AID, The Malaysian Family Life Survey: Appendix E, Master Codebook, Terry Fain and Tan Poh Kheong, RAND, 1982.

MR-111-NICHD, The First Malaysian Family Life Survey: Documentation for Subfiles, Christine Peterson and Nancy Campbell, RAND, 1993. The MFLS-1 data were recently restructured from a hierarchical format to a subfile format (matching MFLS-2 data structure), with separate data files for each card and record type. MR-111 describes the structure of the reformatted MFLS-1 data.

MFLS-2 documentation

MR-106-NICHD/NIA, The Second Malaysian Family Life Survey: Overview and Technical Report, John Haaga, Julie DaVanzo, Christine Peterson, Tey Nai Peng, and Tan Boon Ann, RAND, 1993.

MR-107-NICHD/NIA, The Second Malaysian Family Life Survey: Survey Instruments, Julie DaVanzo, John Haaga, Tey Nai Peng, Ellen Starbird, and Christine Peterson, with the staff of the Population Studies Center of the National Population and Family Development Board of Malaysia, RAND, 1993. This document describes the questionnaire development process, notes differences in questions between MFLS-1 and MFLS-2, and includes the interviewer instruction manual.

MR-107/1-NICHD/NIA, The Second Malaysian Family Life Survey: Questionnaires, Julie DaVanzo, John Haaga, Tey Nai Peng, Ellen Starbird, and Christine Peterson, RAND, 1993. This document contains the actual questionnaires used in the field.

MR-108-NICHD/NIA, The Second Malaysian Family Life Survey: Codebook, Christine Peterson, Jeffrey Sine, and Deborah Wesley, RAND, 1993.

MR-109-NICHD/NIA, The Second Malaysian Family Life Survey: User's Guide, Christine Peterson, RAND, 1993. This document explains how the various MFLS-2 subfiles are related and how users can link information among subfiles. It also discusses how to identify different types of respondents and how to link MFLS-1 and MFLS-2 data.

Other documents that may be of interest to MFLS users include:

N-2157-AID, The Accuracy of Retrospective Data from the Malaysian Family Life Survey, John Haaga, RAND, 1986. This document assesses the quality of the MFLS-1 retrospective fertility-related data.

MR-110-NICHD, The Second Malaysian Family Life Survey: Quality of Retrospective Data for the New Sample, Jeffrey Sine and Christine Peterson, RAND, 1993. This document assesses the quality of the retrospective data for the MFLS-2 New Sample on marital status, fertility, infant and fetal mortality, birthweight, contraception, breastfeeding, and education.

Where can I get the MFLS data and documentation?

Data and documentation for MFLS-1 and MFLS-2 can be ordered from RAND. The MFLS-2 data can also be ordered from the Inter-University Consortium for Political and Social Research (ICPSR). The MFLS-1 data have been sent to the ICPSR archive but have not yet been formally released by ICPSR.

Individuals at institutions that are ICPSR members may contact their ICPSR representative to order the MFLS-2 data and check on the status of the MFLS-1 data. All ICPSR data are available free of charge to ICPSR members. We encourage ICPSR members to order the MFLS-2 data (and the MFLS-1 data when they become available) from ICPSR.

Individuals and institutions not associated with ICPSR may order the MFLS-2 data and the MFLS-1 data from RAND. Requests for the MFLS data and documentation should be made to:

Distribution Services
RAND
1700 Main Street
P.O. Box 2138
Santa Monica, CA 90407-2138
Phone: 310-451-7002
FAX: 310-451-6915

Distribution Services will send you an order form and will handle all questions regarding billing and shipping. Questions about the content and use of the MFLS data are discussed below under MFLS Support. The charge for each MFLS data set, including documentation, is $180 (to cover the costs of processing the order). Shipping charges are extra and depend on how the data are being sent. The MFLS data are free to Malaysian citizens, although shipping charges may be assessed for priority overseas delivery.

About MFLS Support and Updates

MFLS support

The MFLS database is complex, especially the "panel" component that links the 1976-1977 survey with the 1988-1989 survey. While we have tried to include extensive detail in the MFLS documentation, especially the codebook and user's guide for MFLS-2, there will continue to be questions about variables and the relationships between people and their responses. In some cases ready answers may be available; in other cases more investigation may be needed. In either case, we encourage users to contact us with perplexing problems or questions. Just as we hope to reduce the amount of user "head scratching," we also hope to learn more about the MFLS database and pass that information on to other users.

While users may always write or call RAND with questions, we have found that questions can be handled most effectively via electronic mail or fax. These media are timely and provide written records of questions and answers. Users who wish to send questions via electronic mail can address their questions to:

mfls-supp@rand.org

Faxes should be sent to Christine Peterson at 310-393-4818. Phone calls or letters regarding MFLS data questions may be directed to Christine Peterson (ext. 6534) or Julie DaVanzo (ext. 7516) at RAND (310-393-0411).

Data updates

Invariably after a database is released, some data errors and oddities are discovered. In this section of the FLS Newsletter, we present any known data errors and oddities that have emerged after the public release of a database and since the last issue of the newsletter.

MFLS-1 updates

During comparisons of the MFLS-1 and MFLS-2 pregnancy histories, we discovered that the following revisions should be made to the MFLS-1 pregnancy history data :

Revised format for MFLS-1 data

The MFLS-1 data are now available in a subfile format similar to the MFLS-2 data. We have found that the subfile format is more user-friendly. The data are already broken up into files of the same record type. Users can then work with only those subfiles that interest them.

The reformatted MFLS-1 data are described in the following publication:

MR-111-NICHD, The First Malaysian Family Life Survey: Documentation for Subfiles, Christine Peterson and Nancy Campbell, RAND, 1993.

The reformatted data are available from RAND and will be available from ICPSR.

Newly available MFLS-1 data

We recently discovered several data items collected in MFLS-1 that were not part of the original release of the MFLS-1 data in 1978. While not critical to the use of MFLS-1, these data do provide additional information that some users might find helpful. These new files are part of the reformatted MFLS-1 database discussed above.

The first data set contains a group of recall test questions included in Round 2 to check the reliability of responses to a small set of questions. These questions, found on page 143 of the MFLS-1 questionnaires (R-2351/1), cover place of birth, residences at age 15, whether enrolled in school at age 15, breastfeeding of the first child, jobs since age 15, date when began first job.

The second data set contains information updating open breastfeeding and amenorrhea intervals that existed at the end of Rounds 1 and 2.

The Male Migration History (MF10) data file has been revised. The original MF10 data did not include all the recorded migration history records, due to an error in keypunch instructions. As part of the MFLS-2 project, most of the original MFLS-1 questionnaires were located and the complete migration history data (date and place) from the MF10 were reentered. All the above data are described in MR-111, cited above.

MFLS-2 updates

During comparisons of the MFLS-1 and MFLS-2 pregnancy histories, we discovered that the following revisions should be made to the MF22PREG file for the Panel and Child (PC) Sample:

User Questions

This regular section of the FLS Newsletter incorporates questions of general interest raised by data users. This section will deal largely with questions that clarify existing variables and relationships among variables, respondents, and subfiles in the publicly released databases.

The Second Malaysian Family Life Survey: User's Guide (or MFLS-2 User's Guide), MR-109-NICHD/NIA, provides basic information on linking the different subfiles of the MFLS-2 database; on identifying different samples, households, and individuals within the subfiles; and on the interrelatedness of data across the various subfiles. The User Questions section of the FLS Newsletter will cover topics not discussed in the MFLS-2 User's Guide or any user's guide produced for future databases.

How do I link the MF25 wage information for the current job to the current job in MF22WORK/MF23WORK to fill in a value for EARNEND (last wage on job)?
The SEQNO in MF25INC has no relation to JOB_NUM in MF23WORK. Therefore, to link an activity listed in MF25INC with its counterpart in MF22WORK/MF23WORK, use only those records in MF23WORK where YRSWORK=88 (i.e., current activity) and match them to MF25INC using CASE SPLIT PERSON OCC (OCCODE in MF25INC can be renamed to OCC for matching). There are a few people who had an activity in MF25INC that they did not report in MF22WORK/MF23WORK and vice versa. These omissions are the result of respondent error--the person just did not give the information. There are also some cases where the same occupation code appears for what in fact are two different activities. This happens mainly with agricultural occupations and occurs because the difference in activities is at the three-digit occupation code level, but only a two-digit code was entered. For those odd cases where more than one current activity has the same occupation code, the records will have to be checked by hand to ensure proper linkage between MF22WORK/MF23WORK and MF25INC. The full-time/part-time information proves quite helpful in distinguishing which records match between the two file types.
How do I convert the MF25 wage information in NETPAY and PAYUNIT to monthly wage, which is the unit of EARNEND in MF22WORK/MF23WORK?
Computing EARNEND from NETPAY and PAYUNIT is fairly easy, since current earnings are reported as monthly income for 90 percent of the activities in MF25INC. Annual income is reported for another 5 percent. If the WEEKS value in MF25INC reflects a full year of work, simply divide NETPAY by 12; if not, use WEEKS to figure out how many months the person did work and divide NETPAY by that figure. If a weekly rate is given, multiply NETPAY by 4.3 to get a monthly wage rate. Converting daily rates into monthly can be tricky and the analyst must decide on how to proceed. One could assume X number of days in a month and multiply by X; or one could use HOURS to get an estimate of how many days per week a person might work--for employees, 8 hours a day is common; for self-employed people, it could run from 8 to 12 hours (or more)--and then multiply that by 4.3 to get a monthly rate. For the few hourly rate cases, one could use HOURS*NETPAY to get pay/week and then multiply by 4.3 to get a monthly estimate. Lump sum and unknown units are left to the analyst to estimate.
How do I construct an URBAN/RURAL variable for the location in 1988?

For New and Senior Sample households, merge in STRATUM and/or URBAN by the variables EB EBSECT from the community data (either MF26EB or MF27COMM). STRATUM is a categorical variable that represents "urbanicity." URBAN is a simple indicator, where 1=urban and 2=rural. The MFxxSUM files have the variables EB and EBSECT, and these can be merged onto any MFxx file by using CASE SPLIT PERSON.

For Panel Sample households that have not moved since 1976, use STRATUM and/or URBAN from the community data and match by the variable PSU. Since PSU for the Panel Sample represents the PSU in 1976 and NOT the location of the household in 1988, the MFLS-2 community data are accurate only for Panel households that have not moved.

For Children Sample households and Panel Sample households that have moved since 1976, use the last migration record to get an urban/rural measure for the 1988 location (MF22MIG if there is an MF22 respondent or MF23MIG if there is no MF22 respondent). The migration record contains a variable called TYPE. Values 1-4 represent rural locations, but there is one possible exception: If the DISTRICT is 78 (Wilayah Persekutuan, the Federal District of Kuala Lumpur), some TYPE=1 cases might really be urban since areas called "kampung" exist within Kuala Lumpur, the capital city. One may want to consider such cases to be urban since they have access to urban services, although the kampung area itself may be poor.

Bibliography of FLS Research

This regular section of the FLS Newsletter will include a list of research papers based on RAND Family Life survey databases. Initially, this section will contain mainly papers based on the MFLS-2 data. A bibliography of MFLS-1-based research through 1992 is included as Appendix A in MR-106, The Second Malaysian Family Life Survey: Overview and Technical Report. The completeness of the MFLS bibliography is dependent on the goodwill and assistance of MFLS users. We can list only those papers of which we are aware. The assistance of MFLS users is especially needed to identify working papers, conference papers, and other works not yet published in professional journals. Thus, we encourage MFLS users who wish to include their work in the bibliography to send us the relevant citations. We have included a form to make it easy to respond.

MFLS Bibliography

Journal articles, book chapters, reports, working papers, and conference papers

Bollinger, Lori, 1993, "The Effect of Child Care and Family Planning Costs in Malaysia," presented at the annual meeting of the Population Association of America, Cincinnati, Ohio.

Brien, Mike J., and Lee A. Lillard, 1994, "Education, Marriage and First Conception in Malaysia," Journal of Human Resources (forthcoming); also available as RAND Labor and Population Working Paper 93-16 (RAND/DRU-346).

Chan, Angelique, and Julie DaVanzo, 1993, "Interrelations Among Various Types of Intergenerational Transfers in Peninsular Malaysia," presented at the annual meeting of the Gerontological Society of America, New Orleans, LA.

DaVanzo, Julie, and Angelique Chan, February 1994, "Living Arrangements of Older Malaysians--Who Coresides with Their Adult Children?" Demography, Vol. 31, No.1; also available as RAND Labor and Population Working Paper 93-14 (RAND/RP-284).

DaVanzo, Julie, Jeffrey Sine, Christine Peterson, and John Haaga, 1994, "Reversal of the Decline in Breastfeeding in Peninsular Malaysia? Ethnic and Educational Differentials and Data Quality Issues," Social Biology, Vol. 41, Issue 1-2; also available as RAND Labor and Population Working Paper 93-29 ( RAND/DRU-487-NICHD).

Govindasamy, Pavalavalli, and Julie DaVanzo, 1992, "Ethnic Fertility Differentials in Peninsular Malaysia: The Impact of Government Policies," Population and Development Review, Vol. 18, No. 2, pp. 243-267.

Haaga, John, Christine Peterson, Julie DaVanzo, and S. Mengchee Lee, 1993, Health Status and Family Support of Older Malaysians, RAND, Labor and Population Working Paper 93-17 (RAND/DRU-378-NIA).

Haaga, John, Julie DaVanzo, Christine Peterson, and Tey Nai Peng, 1993, Twelve-Year Follow-Up of Respondents and Their Adult Children in a Panel Survey in Peninsular Malaysia, RAND, Labor and Population Working Paper 93-19 (RAND/RP-358).

Karoly, Lynn, and Nga Vuong, 1993, "Remittances by Migrants: Evidence from the Malaysian Family Life Surveys," presented at the annual meeting of the Population Association of America, Cincinnati, Ohio.

King, Elizabeth, K. H. Anderson, and Y. Wang, 1993, "Linkages Between the Returns to Schooling and the Schooling Choices of Boys and Girls in Peninsular Malaysia," presented at the annual meeting of the Population Association of America, Cincinnati, Ohio.

Klerman, Jacob A., 1993, Heaping in Retrospective Data: Insights from Malaysian Family Life Surveys' Breastfeeding Data, RAND, Labor and Population Working Paper 93-21 (RAND/DRU-384-NICHD).

Lillard, Lee A., and Robert J. Willis, 1994, "Intergenerational Educational Mobility: Effects of Family and State in Malaysia," Journal of Human Resources (forthcoming); also available as RAND Labor and Population Working Paper 93-38 ( RAND/DRU-558-NIA).

Panis, Constantijn W. A., and Lee A. Lillard, 1994, "Health Inputs and Child Mortality: Malaysia," Journal of Health Economics (forthcoming); also available as RAND Labor and Population Working Paper 93-03 (RAND/DRU-191-NICHD).

Panis, Constantijn W. A., and Lee A. Lillard, 1993, Child Mortality in Malaysia, 1950-1988, RAND, Labor and Population Working Paper 93-09 (RAND/DRU-311-NICHD).

Panis, Constantijn W. A., and Lee A. Lillard, 1993, Timing of Child Replacement Effects on Fertility in Malaysia, RAND, Labor and Population Working Paper 93-13 (RAND/DRU-331-NICHD).

Peterson, Christine, and Courtland Reichman, 1993, "Desired Fertility Measures for Wives and Husbands: Are They Consistent and How Well Do They Predict Subsequent Fertility?" paper presented at the annual meeting of the Population Association of America, Cincinnati, Ohio.

Pong, Suet-Ling, 1993, "Preferential Policies and Secondary School Attainment in Peninsular Malaysia," Sociology of Education, Vol. 66.

Sine, Jeffrey, Tey Nai Peng, and Julie Da Vanzo, editors, 1993, Proceedings of the Seminar on the Second Malaysian Family Life Survey, Kuala Lumpur, Malaysia, October 1991, RAND, CF-109-NICHD/NIA/WFHF. This volume contains papers prepared by researchers at the Population Studies Center of the National Population and Family Development Board of Malaysia and at RAND that use the data from the MFLS-2 New and Senior samples. Topics covered include marriage, by Khalipah Mohd. Tom; fertility, by Philomena Ganga; contraceptive use, by Tey Nai Peng; breastfeeding, by Julie Da Vanzo; women's labor force participation, by Rohani Ab. Razak; childcare, by Nor Akmawati Mohd. Zain; data quality, by Jeffrey Sine; educational expenses, by Abdul Manan Abd. Rahman; health status and health care utilization of the older population, by Nazileh Ramli; living arrangements of the older population, by Julie DaVanzo; and household amenities, by Foo Sya Tong.

Smith, James P., and Duncan Thomas, 1993, On the Road: Marriage and Mobility in Malaysia, RAND, Labor and Population Working Paper 93-11 (RAND/DRU-323-NICHD).

Smith, James P., Duncan Thomas, and Lynn A. Karoly, 1992, "Migration in Retrospect: Differences Between Men and Women," unpublished paper, RAND.

Theses and dissertations

Chan, Angelique, Are Intergenerational Transfers in Peninsular Malaysia a Substitute for Coresidence? Masters thesis, University of California, Los Angeles, 1991.

Govindasamy, Pavalavalli, Ethnic Fertility Differentials in Peninsular Malaysia: The Impact of Government Policies, Ph.D. dissertation, Michigan State University, 1991.

LeFranc, Christophe, Demographic Influences on Income Distribution: A Study of Household Survey Data from Peninsular Malaysia, Ph.D. dissertation, University of California, Berkeley, 1993. Available from University Microfilms Inc. (UMI).

Panis, Constantijn W. A., A Simultaneous Treatment of Fertility Timing, Child Mortality, and Health Inputs, Ph.D. dissertation, University of Southern California, 1992.

Setapa, Sabariah, An Analysis of Old Age Support and Security in Malaysia, Masters thesis, University of Southern California, 1993.