The stability and growth of a country’s economy hinges on its ability to produce goods and services for both domestic and international use. Labor represents an important factor of production, hence, the improvement of the quality of the labor force and efforts to make it more productive and responsive to growth are necessary for the development of the economy. A clear knowledge and understanding of the size, composition and other characteristics of the segment of the population is a big step in this direction. A continuing supply of data on labor force is indispensable to national as well as regional planning.

The Labor Force Survey (LFS) aims to provide a quantitative framework for the preparation of plans and formulation of policies affecting the labor market.

Specifically, the survey is designed to provide statistics on levels and trends of employment, unemployment and underemployment for the country, as a whole, and for each of the administrative regions, including provinces and key cities.

Starting July 1987, the LFS uses a new questionnaire design and adopts modifications in the concepts and definitions for measuring labor force and employment characteristics. The design is based on a past week reference period and new concept of availability and looking for work is adopted.

The July 1996 round of the Labor Force Survey (LFS) adopted a new sampling design constructed from the listings of the recently concluded 1995 Census of Population. The number of sample households increased from 26,000 to an expanded sample of 41,000 households deemed sufficient to provide a more precise and reliable estimates at the provincial/key city levels.

This section presents the important concepts used in the LFS. Concepts and definitions mentioned in previous Integrated Survey of Households (ISH) series are in most cases the same as those in this one.

Barangay

A barangay is the smallest political subdivision in the country, several of which compromise one city or municipality. For purposes of enumeration in the LFS, a barangay is considered the basic geographic enumeration area.

Household

A household is an aggregate of persons, generally but not necessarily bound by ties of kinship, who live together under the same roof and eat together or share in common the household food. Members comprise the head of the household, relatives living with him, and other persons who share the community life for reasons of work or other consideration. A person who lives alone is considered a separate household.

Reference Period

Employment Status Concepts

In the Labor Force or Economically Active Population.This refers to persons 15 years old and over who are either employed or unemployed in accordance with the definitions described below.

Employed. Employed persons include all those who, during the reference period are 15 years and over as of their last birthday and are reported either:

a.   At work. Those who do any work even for one hour during the reference period for pay or profit, or work without pay on the farm or business enterprise operated by a member of the same household related by blood, marriage or adoption; or

b.   With a job but not at work. Those who have a job or business but are not at work because of temporary illness/injury, vacation or other reasons. Likewise, persons who expect to report for work or to start operation of a farm or business enterprise within two weeks from the date of the enumerator’s visit are considered employed.

Underemployed. Underemployed persons include all employed persons who expressed the desire to have additional hours of work in their present job or an additional job, or to have a new job with longer working hours. Visibly underemployed persons are those who work for less than 40 hours during the reference period and want additional hours of work.

Unemployed Unemployed persons include all those who, during the reference period are 15 years old and over as of their last birthday who have no job/business and actively looking for work. Also considered as unemployed are persons without a job or business who are reported not looking for work because of their belief that no work was available or because of temporary illness/disability, bad weather, pending job application or waiting for job interview.

Employment Population Ratio. Employment Population Ratio is the proportion of employed persons to the total population 15 years old and over.

Persons Not in the Labor Force. Persons 15 years old and over who are neither employed nor unemployed according to the definitions mentioned. Those not in the labor force are those persons who are not looking for work because of reasons such as housekeeping, schooling, etc. Examples are housewives, students, disabled or retired persons.

Determination of Employment Status.The employment status of persons 15 years and over is determined on the basis of answers to a series of inter-related questions, which are described below:

a.   "Did____ do any work at all even for only one hour during the past week?" This question is asked to identify the employed persons. "Work at all" for purposes of this survey means that a person reported to his place of work and performed his duties/activities for at least one hour during the reference week. If a person reported that he did some work, not counting chores around the house, he is still considered in the employed category although most of his time was devoted to household chores. All persons not identified by the above question as employed are asked the following questions.

b.   "Although ____ did not work, did ____ have a job or business during the past week?" Some persons may not have work at all during the past week but may actually have jobs or businesses which they are temporarily not reporting to, as in the following cases: an employee on strike; a person temporarily laid off due to non-economic reasons like machine breakdown; a person with a new job to begin within two weeks from the date of interview; regular and temporary teachers, excluding substitutes, during summer vacation who still receive pay and who expect to go back to their jobs in the next school year. These persons are considered employed even though they are not actually at work.

c.   "Did ____ look for work at any time during the past week?" This question is asked to determine who among those who had no job/business had really done something to look for work. If a person looked for work, he or she is classified as unemployed, otherwise, the next question asked is to determine whether a person should be classified as unemployed or not in the labor force

d.   "Why did ____ not look for work?" This question seeks to determine if the main reason for not looking for work is valid (see definition of unemployed) in which case the person is considered unemployed.

If the answer to this question is schooling, housekeeping, too young/old or retired/permanent disability or other reasons not considered valid, then the person is excluded from the labor force.

Philippine Concept on Unemployment

The Philippine Concept considered a person unemployed if he has no job/business during the reference week and is actively looking for work. Also considered as unemployed are persons without a job/business who are reported not looking for work because of the belief that no work available or because of temporary illness/disability, bad weather, pending job application or waiting for job interview.

ILO Concept on Unemployment

The ILO Concept on employment states that a person is unemployed if he has no job or business during the reference week and is reported available and actively looking for work. Also considered as unemployed are persons who do not have job/business and are available for work but did not look for work because of the belief that no work is available, because of temporary illness/disability, bad weather, awaiting results of job application or awaiting for rehire/job recall.

Work

Work means something a person does during the past week, for pay in cash or in kind, in any establishment, office, farm, private home or for profit or without pay on a family farm or enterprise. It also includes what a farm operator or member of the operator’s family does on the farm operated by another household on exchange labor arrangement.

In addition to the above, any activity that a person does during the past week in relation to minor activities in home gardening, raising of crops, fruits, hogs, poultry etc., fishing for home consumption and manufacturing for own use are also considered work. However, there must be some harvest in the case of home gardening, raising of crops, fruits and nuts and gathering of wild fruits and vegetables; animals disposed of (sold, consumed, bartered or given away) or some catch in fishing in order that these activities will be considered work.

Occupation and Industry

The data on occupation and industry relate to the job held by employed persons during the past week. Occupation refers to the specific kind of work a person does while industry refers to the nature or character of the business or enterprise or the place wherein a person works. Persons employed with two or more jobs are reported in the job at which they worked the greatest number of hours during the past week.

The new Philippine Standard Occupation Classification (1992 PSOC) and Philippine Standard Industry Classification (1994 PSIC) codes were used starting January 2001

Class of Worker

Employed persons are classified according to seven categories, namely:

Worked for private household. These are employed persons working in a private household for pay, in cash or in kind. Examples are domestic helper, household cook, gardener, and family driver.

Worked for private establishment. These are persons working in a private establishment for pay, in cash or in kind. This category includes not only persons working for a private industry but also those working for a religious group, missionary, unions, and non-profit organizations. Examples of persons working for a private establishment are public transport drivers who do not own the vehicle but drive them on boundary basis, persons working in public works projects on private contractors, dock hands or stevedores, cargo handlers in railroad stations or piers, etc.

Worked for government/ government corporation. These are persons working for the government or a government corporation or any of its instrumentalities. This category of worker includes the following workers: chaplains in the Armed Forces of the Philippines, Filipinos working in embassies, legation, chancelleries or consulates of foreign government in the Philippines and those working in international organizations of Sovereign States of Governments like the United Nations (UN), World Health Organization (WHO), etc.

Self-employed. These are persons who operate their own businesses or trades and do not employ paid workers in the conduct of their economic activities. This category includes workers who worked purely on commission basis and who may not have regular working hours.

Employers. These are persons who employ one or more paid employees in the operation of their businesses or trades. Thus, domestic helpers, family drivers and other household helpers who assist in the family-operated business, regardless of time spent in this activity, are not hired employees in the enterprise/business. A farm or business proprietor who is assisted purely by such domestic help is not also considered an employer.

Worked with pay on own family-operated farm or business.These are members of the family who receive cash or fixed share of the produce as payment for their services in a farm or business operated by another member living in the same household.

Worked without pay on own family-operated farm or business.These are members of the family who assist another member in the operation of the family farm or business enterprise and who do not receive any wage or salary for their work. The room and board and any cash allowance given as incentives are not counted as compensation for these family workers.

Number of Hours Worked

Number of hours worked refers to the total number of hours a person actually worked in all the jobs/businesses that he held. It includes the duration or the period the person was occupied in his work, including overtime, but excluding hours paid but not worked. The normal working hours per day is the usual or prescribed working hours of a person in his primary job/business, which is considered a full day’s work.

Averages

The averages shown in this report are arithmetic means.

Rounding of Estimates

Individual figures are independently rounded to the nearest thousands; hence, group totals may not always be equal to the sum of the individual figures.

Comparability with Related Data

The information presented herein is obtained from sample households. Differences observed among corresponding figures obtained from a complete count or another independent survey using the same schedules and instructions are due to sampling variations and other biases not attributable to sampling. Due to the difference in primary sampling units, the employment data obtained from household surveys may differ from employment data based on reports from establishment surveys.

Population Coverage

The LFS has as its target population, all households or members of households living in them nationwide. A household is defined to be an aggregate of persons, generally but not necessarily bound by ties of kinship, which live together under the same roof and eat together or share in common the household food. Household membership comprise the head of the household, relatives living within him, and other persons who share the community life for reasons of work or other consideration. A person who lives alone is considered a separate household.

Excluded in the target population are the households in the least accessible barangays (LABs). A barangay is classified as LAB if: (a) it requires more eight hours walk from the last vehicle station; and/or, (b) the frequency of transportation is less than three times a week and the cost of a one-way trip is more than five hundred pesos. A total of 350 barangays were classified LABs. This accounts for only 0.83% of the total number of barangays in the country. The total number of households in these areas account for only 0.38% of the total number of households.

Sampling Design

The LFS used the sampling design of the 2003 Master Sample (MS) for household surveys starting in July 2003.

• Domains

  The 2003 MS considers the country’s 17 administrative regions as defined in Executive Orders (EO) 36 and 131 as its sampling domain. A domain is referred to as a subdivision of the country in which estimates with adequate level of precision is generated. It must be noted that while there is demand for data at the provincial level (and to some extent municipal and barangay levels), these were not treated as domain because of its large number (more than 80) and the large resource requirement that goes along with it. Below are the 17 administrative regions of the country:

• National Capital Region

• Cordillera Administrative Region

• Region I - Ilocos

• Region II - Cagayan Valley

• Region III - Central Luzon

• Region IVA - CALABARZON

• Region IVB - MIMAROPA

• Region V - Bicol

• Region VI - Western Visayas

• Region VII - Central Visayas

• Region VIII - Eastern Visayas

• Region IX - Zamboanga Peninsula

• Region X - Northern Mindanao

• Region XI - Davao

• Region XII - SOCCSKSARGEN

• Region XIII - Caraga

• Autonomous Region in Muslim Mindanao

• Sampling Frame

  As in most household surveys, the 2003 MS made use of an area sample design. For this purpose, the Enumeration Area Reference File (EARF) of the 2000 Census of Population and Housing (CPH) was utilized as sampling frame. The EARF contains the number of households by enumeration area (EA) in each barangay.

  This frame was used to form the primary sampling units (PSUs). With consideration of the period for which the 2003 MS will be in use, the PSUs were formed/defined as a barangay or a combination of barangays with at least 500 households.

• Sample Size

  The 2003 MS consists of a sample of 2,835 PSUs of which 330 were certainty PSUs and 2,505 were non-certainty PSUs. The entire MS was divided into four sub-samples or independent replicates, such as a quarter sample contains one fourth of the PSUs found in one replicate; a half sample contains one-half of the PSUs in two replicates.

  This frame was used to form the primary sampling units (PSUs). With consideration of the period for which the 2003 MS will be in use, the PSUs were formed/defined as a barangay or a combination of barangays with at least 500 households.

• Stratification

  The 2003 MS considers the 17 regions of the country as the primary strata. Within each region, further stratification was performed using geographic groupings such as provinces, highly urbanized cities (HUCs), and independent component cities (ICCs). Within each of these substrata formed within regions, the PSUs were further stratified, to the extent possible, using the proportion of strong houses (PSTRONG), indicator of engagement in agriculture of the area (PAGRI), and a measure of per capita income as stratification factors (PERCAPITA).

  PSTRONG is defined to be the percentage of housing units occupied by households that are classified as made of strong materials in terms of both the roof and outer walls, based on the data from the 2000 CPH. A roof is considered made of strong material if it is made of either galvanized iron, aluminum, concrete/clay tile, half galvanized-half concrete, or asbestos. The outer wall is considered made of strong material if it is made of concrete, brick, stone, wood, half concrete-half wood, galvanized iron, asbestos, glass.

  AGRI was determined in the following way: initially, an indicator variable was computed at the barangay level. That variable has the value 1 if more than 50 percent of the households in the barangay are engaged in agriculture or fisheries and 0 otherwise, based on the 2000 CPH Barangay Schedule. To obtain a measure at the PSU level, a weighted average of the barangay indicator variable was computed for all the barangays within the PSU, weighted by the total number of households in the barangay. Thus, the l value of AGRI at the PSU level lies between 0 and 1.

  PERCAPITA is defined as the total income of the municipality divided by the total population in that municipality. Note that the PERCAPITA value of the PSUs is the same if the PSUs are in the same municipality. The municipal income used was the 2000 municipal income sourced from the Department of Finance. If the 2000 municipal income was not reported to the BLGF, 2001 income was used. If no 2000 or 2001 municipal income was reported, the median income of the municipal class of the municipality was used.

• Sample Selection

  To have some control over the subsample size, the PSUs were selected with probability proportional to some estimated measure of size. The size measure refers to the total number of households from the 2000 CPH. Because of the wide variation in PSU sizes, PSUs with selection probabilities greater than 1 were identified and were included in the sample as certainty selections.

  At the second stage enumeration areas (EAs) were selected within sampled PSUs, and at the third stage housing units were selected within sampled EAs. Generally, all households in sampled housing units were enumerated, except for few cases when the number of households in a housing unit exceeds three. In which case, a sample of three households was selected with equal probability.

  An EA is defined as an area with discernable boundaries within barangays consisting of about 150 contiguous households. These EAs were identified during the 2000 CPH. A housing unit is a structurally separate and independent place of abode which, by the way it has been constructed, converted, or arranged, is intended for habitation by a household.

  The selection of sample households for the third stage was done systematically from the 1995 POPCEN List of Households.

Estimation Procedures

• Weighting

  In the 2003 Master Sample Design, the probability that a household is included in the sample varies across domains/regions. However, the sampling design is epsem within domain (i.e. equal selection probabilities within region). The initial step in the construction of weights is to determine the unit’s base weight. This is defined as the inverse of its selection probabilities. The base weight is further adjusted to take into account possible nonresponse and possibly to make the estimates conform to some known population totals.

• Base weights

  In general, the base weight assigned to a sampled unit is the inverse of its selection probability. In particular, the base weight is computed as the inverse of equations 1 (NSR) and 2 (SR) below:

  

  

  That is, the base weight for NSR and SR samples are equal to equations 3 and 4, respectively:

  

Note that the last term will equal to one in cases when all households in the sampled housing unit are enumerated. That is, when households per housing unit do not exceed three.

A.   Nonresponse Weights

Nonresponse Adjustments

All surveys experience some degree of unit or total nonresponse in which a sampled and eligible unit fails to participate in the survey (for example, the unit may refuse to participate, or may never be at home at the times the interviewer calls). Adjustments are made to the base weights to compensate for nonresponse by sampled units eligible for the survey. In essence the adjustment inflates the base weights of "similar" responding units to compensate for each nonrespondent.

The most common form of nonresponse weighting adjustment is a weighting class adjustment and that is the type of adjustment being used for surveys based on the 2003 MS. The full sample of respondents and nonrespondents is divided into a number of weighting classes or cells and nonresponse adjustment factors are computed for each cell c as



The denominator of ^W'c  is the sum of the weights of respondents (indexed r) in cell c. The numerator adds together the sum of the weights for respondents and the sum of the weights for eligible nonrespondents (indexed m for missing) in cell c. Together these two sums in the numerator give the sum of the weights for the total eligible sample (indexed s) in cell c. Thus, the nonresponse weight adjustment ^W'c  is the inverse of the weighted response rate in cell c. Note that the adjustment is applied with eligible units. Ineligible sampled units (e.g., vacant or demolished housing units and units out of scope for a given survey) are excluded.

Population Weighting Adjustments

Generally, weighted sample distributions do not conform to known population distributions (e.g. projected population counts). In particular, sample estimates of population counts generally fall short of true population counts because of noncoverage. Further weighting adjustments --- termed as population weighting adjustments --- may be made to compensate for noncoverage and to make the survey estimates based on the adjusted weights estimates consistent with known population distributions. These weighting adjustments may be made within weighting cells like the nonresponse cells described above. In this case, the adjustments are often termed post stratification adjustments. More broadly, the adjustments may be made using some form of calibration method. The raking adjustments used with the July 2003 LFS are one form of calibration adjustment.

The population weighting adjustments used, with persons as the units of analysis in the LFS, force the weighted sample estimates to conform to population counts on two dimensions separately: one dimension contains the 12 cells created by the cross classification of sex and six 10-year age groups (15-24, 25-34, 35-44, 45-54, 55-64, 65+); the other dimension is region. The reference population counts are the population projections developed from the 1995 base population. An iterative proportional fitting algorithm, due originally to Deming and Stephan (1940), was employed to rake the nonresponse adjusted person weights so that the weighted survey estimates of the national sex/age distribution and of the regional total population distribution produced the corresponding population projection distributions.

For adjusting household level estimates, the reference count of households is obtained by dividing the total projected population by the average household size. This is resorted to in the absence of projected number of households.

Final Survey Weight

The final survey weight assigned to each responding unit is computed as the product of the base weight, the nonresponse adjustment, and the population weighting adjustment, as described above. The final weights should be used in all analyses to produce valid estimates of population parameters. The use of the weights in estimation is described below.

Estimation Procedure

A.    Estimation of population total and ratio of totals

The LFS generates estimates of totals and ratios. The estimation of totals for domains and/or specific subclasses is quite straightforward and simple. Let be the final weight assigned to a responding unit. Then the estimate of the population total for variable y (e.g. total in the labor force) for a specific domain d can be estimated as:



In similar way, estimates of the population total for the variable y can be estimated for specific subclass of the entire population (e.g. households engaged in agriculture or households in rural areas) or in each domain (e.g. province) as:



where in here A refers to the specific subclass. This approach can also be used in estimating the total number of elements in the population that possess a particular attribute of interest by letting if the unit possess the attribute (e.g. employed) and y _i = 0, otherwise.

Estimation of unemployment rate involves estimating the ratio of the population totals of two variables x and y or the ratio of the total economically active population who are unemployed with the total economically active population. In a specific domain, the ratio of population totals can be estimated as:





Similarly, the estimator of the population ratio of totals for specific subclass of the entire population or domain is given as:





This approach in estimating ratios can also be used in estimating population mean as well as a population proportion. In this case, note that In the case of a proportion, let x_i = 1 and let y_i = 1 if the unit possess the attribute (e.g. poor) and y_i = 0, otherwise.

B.    Variance Estimation

The calculation of standard errors should take into account the complexity of the design such as stratification and the unequal selection probabilities. Also, since sampling was done without replacement within strata, finite population correction (fpc) factors are appropriate. However, since the sampling fractions in most strata are small, the fpc terms can be ignored. While there are several ways or procedures of computing standard errors, one should choose a procedure that in some ways are considered practical to use given the resources available at NSO.

Consider first estimating the population total for a stratum. Let be the final weight assigned to household belonging to stratum h and is the value of the variable y for the same household. The sample estimate for stratum h   is given as . An estimate of its variance is given as



where
           


is the weighted total for psu a   in stratum h and a _h    is the number of sampled PSUs in the stratum. Note that equation (10) involves computing the totals for each sampled PSU in the stratum and computing the variances between PSU totals. The estimate of the total for domain d  is given as . That is, we simply take the sum of the estimates of the strata totals that fall within the domain d.    Since sampling is done independently across strata within a domain, then the variance of can be estimated as . This method of estimating variances has wide applicability and offers flexibility in computing variances for subclass totals. However, it must be pointed out that all PSUs must be included in the computation of the variances even if they do not contribute to the population total (i.e. Y_ha = 0 ).

Suppose one would like to estimate the ratio of population totals for the variables y  and x  for domain d.   Then the estimated ratio is .  This form of ratio estimate is often times referred to as the combined ratio estimator. In this instance, the Taylor series expansion method (Linearization technique) may be applied in the estimation of the variance of defined as



where are estimated using the procedure earlier described and



It must be noted however that equation (11) is a valid approximation if the quantities, X_ha in the denominator (which often corresponds to sample sizes per stratum) are reasonably uniform in size within strata.

The variance estimation procedures described can easily be implemented using a software package for variance estimation provided that the strata and PSUs are correctly specified and identified in the data file.

Questionnaire Design

The items of information presented in this report were derived from a structured questionnaire covering demographic and economic characteristics of individuals. Refer to Appendix C for detailed information on the items included.

Method of Collection

Personal interview was deemed most applicable for the LFS owing to the complexity of the questionnaire, the details required, and the level of education of respondent in sample households.

NSO Statistical Coordination Officers and Statistical Researchers served as interviewers during the operations. Supervision and monitoring of survey operations were done by the Regional Directors/Provincial Statistics Officers of NSO.

Data Processing

Data processing involved two stages: manual processing and machine processing. Manual processing referred to the manual editing and coding of questionnaires. This was done prior to machine processing, which entailed code validation, consistency checks as well as tabulation.

Enumeration was a very complex operation and it happened that accomplished questionnaires had some omissions and implausible or inconsistent entries. Editing was meant to correct these errors.

For purposes of operational convenience, field editing was done. The interviewers were required to review the entries at the end of each interview. Blank items, which were applicable to the respondents, were verified and filled out. Before being transmitted to the regional office, all questionnaires were edited in the field offices.

Coding, the transformation of information from the questionnaire to machine readable form, was likewise done in the field offices.

Machine processing involved all operations that were done with the use of a computer and/or its accessories, that is, from data encoding to tabulation. Coded data were usually in diskettes or CDs.

Machine editing was preferred to ensure correctness of encoded information. Except for sample completeness check and verification of geographic identification, which were the responsibility of the Income and Employment Statistics Division (IESD), some imputations and corrections of entries were done by the machines.

For this round, preliminary and final tabulations were done at the Central Office.

Publication of Results

Published in this report are data on labor force, which provide details for analytical use at the regional and national levels. Unpublished figures for more detailed cross classification can be obtained from the Income and Employment Statistics Division, Household Statistics Department, NSO.