Abstract

The Labor Force Survey is a nationwide survey of households conducted regularly to gather data on the demographic and socio-economic characteristics of the population. It is primarily geared towards the estimation of the levels of employment in the country.

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

Geographic coverage

National coverage, the sample design has been drawn in such a way that accurate lower level classification would be possible. The 73 provinces, 14 cities of the Philippines are covered.

Analysis unit

• Person/ individual

Universe

The survey covered all persons 10 years old and over. Persons who reside in institutions are not covered.

Kind of data

Sample survey data [ssd]

Sampling procedure

The sampling design of the Labor Force Survey adopts that of the Integrated Survey of Households (ISH), which uses a stratified two-stage sampling design. It is prepared by the NEDA Technical Committee on Survey Design and first implemented in 1984. It is the same sampling design used in the ISH modules starting in 1986.

The urban and rural areas of each province are the principal domains of the survey. In addition, the urban and rural areas of cities with a population of 150,000 or more as of 1980 are also made domains of the survey. These cities are the four cities in Metro Manila (Manila, Quezon City, Pasay and Caloocan); and the cities of Angeles, Olongapo,, Bacolod, Iloilo, Cebu, Zamboanga, Butuan, Cagayan de Oro, Davao, and Iligan.

The rest of Metro Manila, i.e., Pasig, Makati and the 11 other municipalities, are treated as three separate domains. In the case of Makati, six exclusive villages are identified and samples are selected using a different scheme. These villages are Forbes Park, Bel-Air, Dasmarinas, San Lorenzo, Urdaneta and Magallanes.

Sampling Units and Sampling Frame The primary sampling units (PSUs) under the sample design are the barangays and the households within each sample barangay comprise the secondary sampling units (SSUs). The frame from which the sample barangays are drawn is obtained from the 1980 Census of Population and Housing (CPH). Hence, all the approximately 40,000 barangays covered in the 1980 CPH are part of the primary sampling frame. The sampling frame for the SSUs, that is, the households, is prepared by listing all households in each of the selected sample barangays. The listing operation is conducted regularly in the sample barangays to update the secondary sampling frame from where the sample households are selected.

Sample Size and Sampling Fraction The size of the sample is envisioned to meet the demand for fairly adequate statistics at the domain level. Taking this need into account and considering cost constraints as well, the decision reached is for a national sample of about 20,000 households. In general, the sample design results in self-weighting samples within domains, with a uniform sampling fraction of 1:400 for urban and 1:600 for rural areas. However, special areas are assigned different sampling fractions so as to obtain "adequate" samples for each. Special areas refer to the urban and rural areas of a province or large city which are small relative to their counterparts.

Selection of Samples For the purpose of selecting PSUs, the barangay in each domain are arranged by population size (as of the 1980 Census of Population) in descending order and then grouped into strata of approximately equal sizes. Four independent PSUs are drawn with probability proportional to size with complete replacement.

Secondary sampling units are selected systematiclally with a random start.

Sampling deviation

Replacement of non-responding or transferred sample households is allowed although it is still possible to have non-response cases due to critical peace and order situation or inaccessibility of the selected sample households. If there are unenumerated barangays or sample households, non-response adjustments are utilized.

Mode of data collection

Face-to-face [f2f]

Research instrument

The items of information presented in the April 1991 Quarterly Labor Force Survey questionnaire were derived from a structured questionnaire covering the demographic and economic characteristics of individuals. The demographic characteristics include age, sex, relationship to household head, marital status, and highest grade completed. The economic characteristics include employment status, occupation, industry, nomal working hours, total hours worked, class of worker, etc.

Cleaning operations

Data processing involves two stages: manual processing and machine processing. Manual processing refers 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 is a very complex operation and may happen that accomplished questionnaires may have some omissions and implausible or inconsistent entries. Editing is 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 are usually in such media as tapes and diskettes. Machine editing is preferred to ensure correctness of encoded information. Except for sample completeness check and verification of geographic identification which are the responsibility of the subject matter division, some imputations and corrections of entries are done mechanically.

Response rate

The response rate for April 1991 LFS was 99.86 percent. The non-response rate of 0.14 percent was due to crticial peace and order situation or inaccessibility of the selected sample or sample households.

Sampling error estimates

Standard Error (SE) and Coefficient of Variation (CV) for the selected variables of the Labor Force Survey (LFS) for April 1991 survey round was computed using the statistical package IMPS. The selected variables referred to include the employment, unemployment and labor force population levels and rates.

A sampling error is usually measured in terms of the standard error for a particular statistic. A standard error is a measure of dispersion of an estimate from the expected value.

The SE can be used to calculate confidence intervals within which the true value for the population can be estimated, while the CV is a measure of relative variability that is commonly used to assess the precision of survey estimates.

The CV is defined as the ratio of the standard error and the estimate. An estimate with CV value of less than 10 percent is considered precise.

Abstract

The Labor Force Survey is a nationwide survey of households conducted regularly to gather data on the demographic and socio-economic characteristics of the population. It is primarily geared towards the estimation of the levels of employment in the country.

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

Geographic coverage

National coverage, the sample design has been drawn in such a way that accurate lower level classification would be possible. The 74 provinces, 24 cities and eight key municipalities are covered.

Analysis unit

• Person/ individual

Universe

The survey covered all persons 10 years old and over. Persons who reside in institutions are not covered.

Kind of data

Sample survey data [ssd]

Sampling procedure

The sampling design of the Labor Force Survey adopts that of the Integrated Survey of Households (ISH), which uses a stratified two-stage sampling design. It is prepared by the NEDA Technical Committee on Survey Design and first implemented in 1984. It is the same sampling design used in the ISH modules starting in 1986.

The urban and rural areas of each province are the principal domains of the survey. In addition, the urban and rural areas of cities with a population of 150,000 or more as of 1990 are also made domains of the survey with urban and rural dimensions. These include the four cities and five municipalities of Metro Manila (Manila, Quezon City, Pasay and Caloocan; Valenzuela, Paranaque, Pasig, Marikina and Makati), and other key cities such as Baguio, Angeles, Cabanatuan, Olongapo, Batangas, Lipa, Lucena, San Pablo, Bacolod, Iloilo, Cebu, Mandaue, Zamboanga, Butuan, Cagayan de Oro, Davao, General Santos, and Iligan and key municipalities such as San Fernando, Pampanga and Tarlac, Tarlac.

The rest of Metro Manila, i.e., the remaining municipalities are treated as separate domains. In the case of Makati, six exclusive villages are identified and samples are selected using a different scheme. These villages are Forbes Park, Bel-Air, Dasmarinas, San Lorenzo, Urdaneta and Magallanes.

Because of the creation of the Autonomous Region of Muslim Mindanao (ARMM), this, defining its areas of coverage, Marawi City and Cotabato Cfity are likewises treated as domains.

Sampling Units and Sampling Frame The primary sampling units (PSUs) under the sample design are the barangays and the households within each sample barangay comprise the secondary sampling units (SSUs). The frame from which the sample barangays are drawn is obtained from the 1990 Census of Population and Housing (CPH). Hence, all the approximately 40,000 barangays covered in the 1990 CPH are part of the primary sampling frame. The sampling frame for the SSUs, that is, the households, is prepared by listing all households in each of the selected sample barangays. The listing operation is conducted regularly in the sample barangays to update the secondary sampling frame from where the sample households are selected.

Sample Size and Sampling Fraction The size of the sample is envisioned to meet the demand for fairly adequate statistics at the domain level. Taking this need into account and considering cost constraints as well, the decision reached is for a national sample of about 26,000 households. In general, the sample design results in self-weighting samples within domains, with a uniform sampling fraction of 1:400 for urban and 1:600 for rural areas. However, special areas are assigned different sampling fractions so as to obtain "adequate" samples for each. Special areas refer to the urban and rural areas of a province or large city which are small relative to their counterparts.

Selection of Samples For the purpose of selecting PSUs, the barangay in each domain are arranged by population size (as of the 1990 Census of Population) in descending order and then grouped into strata of approximately equal sizes. Four independent PSUs are drawn with probability proportional to size with complete replacement.

Secondary sampling units are selected systematiclally with a random start.

Sampling deviation

Replacement of non-responding or transferred sample households is allowed although it is still possible to have non-response cases due to critical peace and order situation or inaccessibility of the selected sample households. If there are unenumerated barangays or sample households, non-response adjustments are utilized.

Mode of data collection

Face-to-face [f2f]

Research instrument

The items of information presented in the July 1991 Quarterly Labor Force Survey questionnaire were derived from a structured questionnaire covering the demographic and economic characteristics of individuals. The demographic characteristics include age, sex, relationship to household head, marital status, and highest grade completed. The economic characteristics include employment status, occupation, industry, nomal working hours, total hours worked, class of worker, etc.

Cleaning operations

Data processing involves two stages: manual processing and machine processing. Manual processing refers 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 is a very complex operation and may happen that accomplished questionnaires may have some omissions and implausible or inconsistent entries. Editing is 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 are usually in such media as tapes and diskettes. Machine editing is preferred to ensure correctness of encoded information. Except for sample completeness check and verification of geographic identification which are the responsibility of the subject matter division, some imputations and corrections of entries are done mechanically.

Response rate

The response rate for January 1992 LFS was 99.94 percent. The non-response rate of 0.06 percent was due to crticial peace and order situation or inaccessibility of the selected sample or sample households.

Sampling error estimates

Standard Error (SE) and Coefficient of Variation (CV) for the selected variables of the Labor Force Survey (LFS) for July 1991 survey round was computed using the statistical package IMPS. The selected variables referred to include the employment, unemployment and labor force population levels and rates.

A sampling error is usually measured in terms of the standard error for a particular statistic. A standard error is a measure of dispersion of an estimate from the expected value.

The SE can be used to calculate confidence intervals within which the true value for the population can be estimated, while the CV is a measure of relative variability that is commonly used to assess the precision of survey estimates.

The CV is defined as the ratio of the standard error and the estimate. An estimate with CV value of less than 10 percent is considered precise.

Abstract

The Labor Force Survey is a nationwide survey of households conducted regularly to gather data on the demographic and socio-economic characteristics of the population. It is primarily geared towards the estimation of the levels of employment in the country.

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

Geographic coverage

National coverage, the sample design has been drawn in such a way that accurate lower level classification would be possible. The 74 provinces, 24 cities and eight key municipalities are covered.

Analysis unit

• Person/ individual

Universe

The survey covered all persons 10 years old and over. Persons who reside in institutions are not covered.

Kind of data

Sample survey data [ssd]

Sampling procedure

The sampling design of the Labor Force Survey adopts that of the Integrated Survey of Households (ISH), which uses a stratified two-stage sampling design. It is prepared by the NEDA Technical Committee on Survey Design and first implemented in 1984. It is the same sampling design used in the ISH modules starting in 1986.

The urban and rural areas of each province are the principal domains of the survey. In addition, the urban and rural areas of cities with a population of 150,000 or more as of 1990 are also made domains of the survey with urban and rural dimensions. These include the four cities and five municipalities of Metro Manila (Manila, Quezon City, Pasay and Caloocan; Valenzuela, Paranaque, Pasig, Marikina and Makati), and other key cities such as Baguio, Angeles, Cabanatuan, Olongapo, Batangas, Lipa, Lucena, San Pablo, Bacolod, Iloilo, Cebu, Mandaue, Zamboanga, Butuan, Cagayan de Oro, Davao, General Santos, and Iligan and key municipalities such as San Fernando, Pampanga and Tarlac, Tarlac.

The rest of Metro Manila, i.e., the remaining municipalities are treated as separate domains. In the case of Makati, six exclusive villages are identified and samples are selected using a different scheme. These villages are Forbes Park, Bel-Air, Dasmarinas, San Lorenzo, Urdaneta and Magallanes.

Because of the creation of the Autonomous Region of Muslim Mindanao (ARMM), this, defining its areas of coverage, Marawi City and Cotabato Cfity are likewises treated as domains.

Sampling Units and Sampling Frame The primary sampling units (PSUs) under the sample design are the barangays and the households within each sample barangay comprise the secondary sampling units (SSUs). The frame from which the sample barangays are drawn is obtained from the 1990 Census of Population and Housing (CPH). Hence, all the approximately 40,000 barangays covered in the 1990 CPH are part of the primary sampling frame. The sampling frame for the SSUs, that is, the households, is prepared by listing all households in each of the selected sample barangays. The listing operation is conducted regularly in the sample barangays to update the secondary sampling frame from where the sample households are selected.

Sample Size and Sampling Fraction The size of the sample is envisioned to meet the demand for fairly adequate statistics at the domain level. Taking this need into account and considering cost constraints as well, the decision reached is for a national sample of about 26,000 households. In general, the sample design results in self-weighting samples within domains, with a uniform sampling fraction of 1:400 for urban and 1:600 for rural areas. However, special areas are assigned different sampling fractions so as to obtain "adequate" samples for each. Special areas refer to the urban and rural areas of a province or large city which are small relative to their counterparts.

Selection of Samples For the purpose of selecting PSUs, the barangay in each domain are arranged by population size (as of the 1990 Census of Population) in descending order and then grouped into strata of approximately equal sizes. Four independent PSUs are drawn with probability proportional to size with complete replacement.

Secondary sampling units are selected systematiclally with a random start.

Sampling deviation

Replacement of non-responding or transferred sample households is allowed although it is still possible to have non-response cases due to critical peace and order situation or inaccessibility of the selected sample households. If there are unenumerated barangays or sample households, non-response adjustments are utilized.

Mode of data collection

Face-to-face [f2f]

Research instrument

The items of information presented in the October 1991 Quarterly Labor Force Survey questionnaire were derived from a structured questionnaire covering the demographic and economic characteristics of individuals. The demographic characteristics include age, sex, relationship to household head, marital status, and highest grade completed. The economic characteristics include employment status, occupation, industry, nomal working hours, total hours worked, class of worker, etc.

Cleaning operations

Data processing involves two stages: manual processing and machine processing. Manual processing refers 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 is a very complex operation and may happen that accomplished questionnaires may have some omissions and implausible or inconsistent entries. Editing is 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 are usually in such media as tapes and diskettes. Machine editing is preferred to ensure correctness of encoded information. Except for sample completeness check and verification of geographic identification which are the responsibility of the subject matter division, some imputations and corrections of entries are done mechanically.

Response rate

The response rate for October 1991 LFS was 99.81 percent. The non-response rate of 0.19 percent was due to crticial peace and order situation or inaccessibility of the selected sample or sample households.

Sampling error estimates

Standard Error (SE) and Coefficient of Variation (CV) for the selected variables of the Labor Force Survey (LFS) for October1991 survey round was computed using the statistical package IMPS. The selected variables referred to include the employment, unemployment and labor force population levels and rates.

A sampling error is usually measured in terms of the standard error for a particular statistic. A standard error is a measure of dispersion of an estimate from the expected value.

The SE can be used to calculate confidence intervals within which the true value for the population can be estimated, while the CV is a measure of relative variability that is commonly used to assess the precision of survey estimates.

The CV is defined as the ratio of the standard error and the estimate. An estimate with CV value of less than 10 percent is considered precise.

Abstract

The Labor Force Survey (LFS) is a nationwide survey of households conducted quarterly to gather data on the demographic and socio-economic characteristics of the population. It is primarily geared towards the estimation of the levels of employment and unemployment in the country. One of the objectives of the Labor Force Survey is 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 of the country, as a whole, and for the 17 administrative regions.

A total national sample of 42,768 sample housholds (rounds with Batanes sample) or 42,576 sample households (rounds without Batanes sample) will be alloted per survey round deemed sufficient to provide more precise and reliable estimates at the national and regional levels only.

The survey involves the collection of data on demographic and socio-economic characteristics of the population in general. The reporting unit is the household which means that statistics emanating from this survey refers to the characteristics of the population residing in private households. Persons who belongs to the institutional population are not within the scope of the survey.

Geographic coverage

The sample was selected to allow separate estimates for the national level, and regional levels only (17 administrative regions).

National Capital Region (NCR) Cordillera Administrative Region (CAR) Region I - Ilocos Region Region II - Cagayan Valley Region III - Central Luzon Region IV-A - CALABARZON Region IV-B - MIMAROPA Region V - Bicol Region Region VI - Western Visayas Region VII - Central Visayas Region VIII - Eastern Visayas Region IX - Zamboanga Peninsula Region X - Northern Mindanao Region XI - Davao Region Region XII - SOCCSKSARGEN Caraga Autonomous Region in Muslim Mindanao (ARMM)

Analysis unit

Individual or Person Persons 15 years old and over

Universe

The survey covered all household members of the sample households.

Considered as members of a household are:

a. Persons who are present at the time of visit, whose usual place of residence is the sample household regardless of their length of stay in the household;

b. Persons who are present at the time of visit, whose usual place of residence is outside the sample household but have stayed temporarily with the sample household, for at least 30 days;

c. Persons who are present at the time of visit, whose usual place of residence is outside the sample household but have stayed with the sample household even for less than 30 days, provided that they have been away from their usual place of residence for 30 days or more;

d. Persons who are not present at the time of visit, but are expected to return within 30 days from date of departure to their usual place of residence, which is the sample household; and

e. The following family members who are away at the time of visit:

 1 . Overseas contract workers (OCWs);

 2. Other overseas Filipino workers who have been away for not more than five years from the date of departure, and are expected to be back within five years from the date of last departure;

 3. Employees in Philippine embassies, consulates and other missions; and

 4. Students abroad/tourists who have been away for one year or less and are expected to be back within a year from the date of departure. This category also includes those attending training abroad, medical treatment and missionaries.

Kind of data

Sample survey data [ssd]

Sampling procedure

The 2013 Master Sample for household-based survey:

In order to be more effecient in the conduct of household based-surveys, the PSA designed a master sample consists of randomly assigned and selected set of geographic areas with non-overlapping and discernable boundaries known as the primary sampling units (PSUs). The primary sampling unit (PSU) can be (1) the whole barangay, or (2) a portion of a large barangay, or (3) combinations of small barangays.

Provinces and Highly Urbanized Ciities as Sampling Domain

To provide sub-national or provincial level statistics with precise estimates the 2013 MS has 117 major domains as follows: 81 provinces (including the newly created province Davao Occidental); 33 highly urbanized cities (HUC) (including 16 citiies in the National Capital Region; and 3 other areas (Pateros, Isabela City, and Cotabato City).

Primary Sampling Units

In the 2013 MS design each sampling diomain(i.e., province/HUC is divided into exhaustive and non-overlapping area segments known as Primary Sampling Units (PSUs) with about 100 to 400 households. Thus, a PSU can be a barangay/Enumeration Area (EA) or a portion of a large barangay or two or more adjacent small barangays/EAs. The PSUs are then ordered according to the following: (1) North-South/West-East Geographic location; (2)decreasing proporion of households with overseas workers; and (3) decreasing wealth index.

Replicates

From the ordered list of PSUs, all possible systematic samples of 6 PSUs will be drawn to form a replicate for the most of the province domain i.e., 75 out of 81 provinces while all possible systemmatic sample of 8 PSUs will be drawn to form a replicate for most of the HUCs, i.e., 31 of 33 HUCs. Small province domains such as Guimaras, Siquijor, Camiguin, Apayao, and Dinagat Isalnds will have 3 PSUs per replicate. Batanes with very less PSUs formed will have 3 PSUs per replicate but will be covered twice a year only (i.e., January and July rounds only). For other HUCs, San Juan City and Lucena City will have 3 and 5 PSUs per replicate, while the other urban areas, Pateros, and City of Isabela will also have 3 PSUs per replicate while Cotabato City will be allocated with 5 PSUs per replicate.

For instance, in Cagayan with 1008 PSUs formed, a total 1008/6 = 168 possible systematic samples of size 6 or 168 R groups or replicates can be made. The 168 replicates formed are then sorted at random so that the first 4 replicates will be in the first round, next 4 in the second round, and so on.

Sample Allocation Scheme

For each domain, a total of 4 sample replicates will be allotted for each survey round. However, the total number of sample SSUs will be alloted proportionately to the measure of size of the PSU. Thus, a PSU with only 100 households would have less number of samples households than PSUs with 400 households but on the average there will be 12 sample households allotted for each PSU in HUCs and an average of 16 sample households alloted for every PSUs in province domains.

A total national sample of 42,768 sample households (rounds with Batanes sample) or 42,576 sample households (rounds without Batanes sample) will be allotted per survey round.

Mode of data collection

Face-to-face [f2f]

Research instrument

ISH FORM 2 (LFS questionnaire) is a four-page, forty three-column questionnaire that is being used in the quarterly rounds of the Labor Force Survey nationwide. This questionnaire gathers data on the demographic and economic characteristics of the population.

On the first page of the questionnaire, the particulars about the geographic location, design codes and household auxiliary information of the sample household that is being interviewed are to be recorded. Certifications by the enumerator and his supervisor regarding the manner by which the data are collected are likewise to be made on this page.

The inside pages of the questionnaire contain the items to be determined about each member of the sample household. Columns 2 to 10 are for the demographic characteristics; columns 2 to 5A are to be ascertained of all members of the household regardless of age. Columns 6 to 7 are asked for members 5 years old and over, while column 8 is asked for members 5 to 24 years old, column 9 to 10, for 15 years old and over, while columns 11 to 15 are asked for members 5 years old and over. Columns 18 to 42 on the other hand, are the series of items that will be asked of all the members 15 years old and over to determine their labor force and employment characteristics.

Other Relevant Information: The question on hunger experienced by any member of the family because they did not have food to eat, was asked to the head of the household. If in question A, the answer is Yes the frequency of experiencing hunger was asked in question B.

Most of the questions have pre-coded responses. The possible answers with their corresponding codes are printed at the bottom of the page for easy reference. Only the appropriate codes need to be entered in the cells.

Other items, however, require write-in entries such as column 14 (primary occupation) and column 16 (kind of business/industry), etc. For such items, it is required that the enumerator describes the primary occupation or kind of business/industry.

The ISH Form 2 is provided as Technical Documents.

Cleaning operations

Data Processing

Verification and review of questionnaires

The SRs were expected to have verified the completeness of the questionnaires, correctness of the entries, consistency of the entries in the different related items, and the correctness of the codes that were entered in the boxes before the questionnaires were submitted to the Assistant Provincial Supervisor or the Provincial Supervisor. Use the list of sample barangays/EAs provided for coding the region, provinces, municipalities and barangays.

The Assistant Provincial Supervisor or the Provincial Supervisor. upon receipt of the questionnaires reviewed the forms and code the responses for occupation

Abstract

The Labor Force Survey is a nationwide survey of households conducted regularly to gather data on the demographic and socio-economic characteristics of the population. It is primarily geared towards the estimation of the levels of employment in the country.

The Labor Force Survey 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 Philippines, as a whole and for each of the administrative regions, including provinces and key cities.

Geographic coverage

National coverage, the sample design has been drawn in such a way that accurate lower level classification would be possible. The 73 provinces, 14 cities of the Philippines are covered.

Analysis unit

• Person/ individual

Universe

The survey covered all persons 10 years old and over. Persons who reside in institutions are not covered.

Kind of data

Sample survey data [ssd]

Sampling procedure

The sampling design of the Labor Force Survey adopts that of the Integrated Survey of Households (ISH), which uses a stratified two-stage sampling design. It is prepared by the NEDA Technical Committee on Survey Design and first implemented in 1984. It is the same sampling design used in the ISH modules starting in 1986.

The urban and rural areas of each province are the principal domains of the survey. In addition, the urban and rural areas of cities with a population of 150,000 or more as of 1980 are also made domains of the survey. These cities are the four cities in Metro Manila (Manila, Quezon City, Pasay and Caloocan); and the cities of Angeles, Olongapo,, Bacolod, Iloilo, Cebu, Zamboanga, Butuan, Cagayan de Oro, Davao, and Iligan.

The rest of Metro Manila, i.e., Pasig, Makati and the 11 other municipalities, are treated as three separate domains. In the case of Makati, six exclusive villages are identified and samples are selected using a different scheme. These villages are Forbes Park, Bel-Air, Dasmarinas, San Lorenzo, Urdaneta and Magallanes.

Sampling Units and Sampling Frame The primary sampling units (PSUs) under the sample design are the barangays and the households within each sample barangay comprise the secondary sampling units (SSUs). The frame from which the sample barangays are drawn is obtained from the 1980 Census of Population and Housing (CPH). Hence, all the approximately 40,000 barangays covered in the 1980 CPH are part of the primary sampling frame. The sampling frame for the SSUs, that is, the households, is prepared by listing all households in each of the selected sample barangays. The listing operation is conducted regularly in the sample barangays to update the secondary sampling frame from where the sample households are selected.

Sample Size and Sampling Fraction The size of the sample is envisioned to meet the demand for fairly adequate statistics at the domain level. Taking this need into account and considering cost constraints as well, the decision reached is for a national sample of about 20,000 households. In general, the sample design results in self-weighting samples within domains, with a uniform sampling fraction of 1:400 for urban and 1:600 for rural areas. However, special areas are assigned different sampling fractions so as to obtain "adequate" samples for each. Special areas refer to the urban and rural areas of a province or large city which are small relative to their counterparts.

Selection of Samples For the purpose of selecting PSUs, the barangay in each domain are arranged by population size (as of the 1980 Census of Population) in descending order and then grouped into strata of approximately equal sizes. Four independent PSUs are drawn with probability proportional to size with complete replacement.

Secondary sampling units are selected systematiclally with a random start.

Sampling deviation

Replacement of non-responding or transferred sample households is allowed although it is still possible to have non-response cases due to critical peace and order situation or inaccessibility of the selected sample households. If there are unenumerated barangays or sample households, non-response adjustments are utilized.

Mode of data collection

Face-to-face [f2f]

Research instrument

The items of information presented in the January 1991 Quarterly Labor Force Survey questionnaire were derived from a structured questionnaire covering the demographic and economic characteristics of individuals. The demographic characteristics include age, sex, relationship to household head, marital status, and highest grade completed. The economic characteristics include employment status, occupation, industry, nomal working hours, total hours worked, class of worker, etc.

Cleaning operations

Data processing involves two stages: manual processing and machine processing. Manual processing refers 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 is a very complex operation and may happen that accomplished questionnaires may have some omissions and implausible or inconsistent entries. Editing is 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 are usually in such media as tapes and diskettes. Machine editing is preferred to ensure correctness of encoded information. Except for sample completeness check and verification of geographic identification which are the responsibility of the subject matter division, some imputations and corrections of entries are done mechanically.

Response rate

The response rate for January 1991 LFS was 99.91 percent. The non-response rate of 0.09 percent was due to crticial peace and order situation or inaccessibility of the selected sample or sample households.

Sampling error estimates

Standard Error (SE) and Coefficient of Variation (CV) for the selected variables of the Labor Force Survey (LFS) for January 1991 survey round was computed using the statistical package IMPS. The selected variables referred to include the employment, unemployment and labor force population levels and rates.

A sampling error is usually measured in terms of the standard error for a particular statistic. A standard error is a measure of dispersion of an estimate from the expected value.

The SE can be used to calculate confidence intervals within which the true value for the population can be estimated, while the CV is a measure of relative variability that is commonly used to assess the precision of survey estimates.

The CV is defined as the ratio of the standard error and the estimate. An estimate with CV value of less than 10 percent is considered precise.