Newsletter

November 1995 - Number 2

Data Updates

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

MFLS-2 Updates

Use of the MFLS-2 data both inside and outside of RAND has led to a few revisions of the public-release data. Some are based on corroborative information, and some are best guesses to fix inconsistent responses.

The following revision should be made to the MF21ROST file for the Panel and Children (PC) sample:

CASE= 635, SPLIT=0, PERSON=301: Make PERSON=6,

Make MFLS1=301

The following revisions should be made to the MF22MIG file for the New and Senior (NS) sample:

CASE=3250, SPLIT=0, PERSON=1, MIGNUM=3: Make NMOVES=5

CASE=3462, SPLIT=0, PERSON=1, MIGNUM>=1: Make NMOVES=0

CASE= 479, SPLIT=0, PERSON=1, MIGNUM=3: Make NMOVES=3

The following revisions are best guesses as to how to fix cases in the MF25INC and MF25OTH household economy files where either the unit of pay given or the amount given appears to be incorrect. These revisions are based on cross-checking with other relevant data within the MFLS-2 database and are not based on reinspection of the hard copy record of the interview.

Suggested revisions to MF25INC file for the New and Senior (NS) sample:

CASE=7435, SPLIT=0, PERSON=1, SEQNO=1: Make PAYUNIT=4

CASE=5954, SPLIT=0, PERSON=1, SEQNO=1: Make NETPAY=120

CASE=3631, SPLIT=0, PERSON=1, SEQNO=1: Make PAYUNIT=3

CASE=4952, SPLIT=0, PERSON=2, SEQNO=2: Make PAYUNIT=5

CASE=3417, SPLIT=0, PERSON=1, SEQNO=1: Make PAYUNIT=4

Suggested revisions to MF25INC file for the Panel and Children (PC) sample:

CASE=1858, SPLIT=0, PERSON=1, SEQNO=1: Make NETPAY=80

CASE=1550 ,SPLIT=2, PERSON=1, SEQNO=1: Make PAYUNIT=4

Suggested revisions to MF25OTH file for the New and Senior (NS) sample:

CASE=3910, SPLIT=0, PERSON=2, SEQNO=1: Make UNITDIV=5

CASE=3830, SPLIT=0, PERSON=2, SEQNO=2: Make UNITREL=5

CASE=5863, SPLIT=0, PERSON=2, SEQNO=1: Make UNITREL=5

CASE=3113, SPLIT=0, PERSON=7, SEQNO=2: Make UNITREL=5

CASE=7450, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=3917, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=3718, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=7226, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=7169, SPLIT=0, PERSON=1, SEQNO=1: Make AMTREL=200

CASE=3340, SPLIT=0, PERSON=2, SEQNO=1: Make UNITDIV=4

CASE=3899, SPLIT=0, PERSON=2, SEQNO=1: Make UNITDIV=4

CASE=4944, SPLIT=0, PERSON=2, SEQNO=1: Make UNITDIV=4

CASE=4944, SPLIT=0, PERSON=2, SEQNO=2: Make UNITDIV=4

CASE=4354, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=6

CASE=3536, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=5251, SPLIT=0, PERSON=1, SEQNO=1: Make AMTREL=1200

CASE=5829, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=5863, SPLIT=0, PERSON=2, SEQNO=1: Make UNITREL=5

CASE=7141, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=7451, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=8044, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=3329, SPLIT=0, PERSON=1, SEQNO=1: Make AMTREL=960

CASE=4505, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=4

CASE=8195, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=5549, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=8285, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=3647, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=7639, SPLIT=0, PERSON=1, SEQNO=2: Make UNITREL=5

CASE=3830, SPLIT=0, PERSON=2, SEQNO=1: Make UNITREL=5

CASE=3903, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=5058, SPLIT=0, PERSON=3, SEQNO=2: Make AMTREL=60

CASE=5537, SPLIT=0, PERSON=1, SEQNO=1: Make AMTREL=1000

Suggested revisions to MF25OTH file for the Panel and Children (PC) sample:

CASE=1225, SPLIT=0, PERSON=1, SEQNO=1: Make UNITDIV=5

CASE= 242, SPLIT=0, PERSON=2, SEQNO=2: Make UNITPROP=4

CASE=1919, SPLIT=0, PERSON=1, SEQNO=1: Make UNITPROP=4

CASE= 942, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=2004, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE= 444, SPLIT=0, PERSON=1, SEQNO=1: Make AMTREL=800

CASE= 810, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE= 839, SPLIT=0, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=1733, SPLIT=2, PERSON=1, SEQNO=1: Make UNITINH=6

CASE= 86, SPLIT=1, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=1578, SPLIT=1, PERSON=1, SEQNO=1: Make UNITREL=5

CASE=1891, SPLIT=1, PERSON=2, SEQNO=1: Make AMTREL=480

The following revisions are best guesses as to how to fix cases in the MF23WORK male work history and the MF25INC household economy income file where the occupation codes appear to be in error. There are also a couple of suggested revisions for the wage variable in MF23WORK and the EMPSTAT variable in MF25INC. These revisions are based on crosschecking with other relevant data within the MFLS-2 database and are not based on reinspection of the hard copy record of the interview.

Suggested revisions to MF23WORK for the Panel and Children (PC) sample:

CASE=298,SPLIT=1,PERSON=2,JOB_NUM=1: Make OCC='45'

CASE=324,SPLIT=0,PERSON=3,JOB_NUM=3: Make OCC='59'

CASE=329,SPLIT=0,PERSON=3,JOB_NUM=2: Make OCC='59'

CASE=1215,SPLIT=0,PERSON=4,JOB_NUM=4: Make OCC='85'

CASE=1215,SPLIT=1,PERSON=2,JOB_NUM=3: Make OCC='87'

CASE=1454,SPLIT=0,PERSON=2,JOB_NUM=1: Make OCC='62'

CASE=1919,SPLIT=0,PERSON=2,JOB_NUM=7: Make OCC='98'

CASE=440,SPLIT=0,PERSON=2,JOB_NUM=1: Make

EARNSTRT=100, EARNEND=100

CASE=440,SPLIT=0,PERSON=2,JOB_NUM=2: Make

EARNSTRT=200, EARNEND=200

CASE=1719,SPLIT=1,PERSON=2,JOB_NUM=1: Make EARNEND=350

Suggested revisions to MF23WORK for the New and Senior (NS) sample:

CASE=3307,SPLIT=0,PERSON=2,JOB_NUM=5: Make OCC='41'

CASE=3768,SPLIT=0,PERSON=2,JOB_NUM=3: Make OCC='41'

CASE=5979,SPLIT=0,PERSON=2,JOB_NUM=1: Make OCC='35'

Suggested revisions to MF25INC for the Panel and Children (PC) sample:

CASE= 378,SPLIT=0,PERSON=2,SEQNO=2: Make OCCODE='51'

CASE= 879,SPLIT=2,PERSON=2,SEQNO=2: Make OCCODE='62'

CASE=1029,SPLIT=1,PERSON=2,SEQNO=1: Make OCCODE='98'

CASE=1228,SPLIT=0,PERSON=2,SEQNO=1: Make OCCODE='97'

CASE=1356,SPLIT=0,PERSON=2,SEQNO=1: Make OCCODE='74'

CASE=1671,SPLIT=2,PERSON=1,SEQNO=1: Make OCCODE='96'

CASE=1855,SPLIT=2,PERSON=2,SEQNO=2: Make OCCODE='14'

CASE=2025,SPLIT=1,PERSON=2,SEQNO=2: Make OCCODE='95'

CASE= 947,SPLIT=0,PERSON=2,SEQNO=1: Make EMPSTAT=2

Suggested revisions to MF25INC for the New and Senior (NS) sample:

CASE=3351,SPLIT=0,PERSON=2,SEQNO=2: Make OCCODE='60'

CASE=3351,SPLIT=0,PERSON=2,SEQNO=3: Make OCCODE='45'

CASE=3920,SPLIT=0,PERSON=2,SEQNO=3: Make OCCODE='83'

CASE=4223,SPLIT=0,PERSON=2,SEQNO=1: Make OCCODE='13'

CASE=5160,SPLIT=0,PERSON=2,SEQNO=2: Make OCCODE='45'

CASE=5160,SPLIT=0,PERSON=2,SEQNO=3: Make OCCODE='61'

CASE=5471,SPLIT=0,PERSON=2,SEQNO=2: Make OCCODE='34'

MFLS-1 Individual-Level Dataset
Now at ICPSR

The MFLS-1 Individual-Level Dataset, which has been available from RAND since 1982, has been given to ICPSR for inclusion in their MFLS-1 data archive. The Individual-Level Dataset, created from the original MFLS-1 data, contains one record for each individual who ever appeared in a household roster across the three rounds of the MFLS-1 survey. The total number of individuals is 8,295. Demographic, time allocation, and income and wealth information are available at both the individual and the household level and relate only to the time of each round. Information is drawn primarily from MF1 (the household roster), MF2 and MF3 (the female and female retrospectives), MF4 and MF5 (the female and male time budget questionnaire), and MF6 (income and wealth). The dataset contains a number of generated variables constructed from original MFLS-1 variables. Many of these variables relate to earned and nonearned income sources. Values for all variables are repeated for each round in which the individual is present.

The documentation for the Individual-Level Dataset consists of a codebook that describes variable locations on the file, code values, and the section of the MFLS-1 questionnaire (and main codebook) in which the variable originated. The codebook is a RAND publication: N-1815-AID, The Malaysian Family Life Survey: Codebook for the Individual-Level Dataset, by Terry Fain, 1982.