This is the fourth Labor Force Survey of Tonga. The first one was conducted in 1990. Earlier surveys were conducted in 1990, 1993/94, and 2003 and the results of those surveys were published by the Statistics Department.
The objective of the LFS survey is providing information on not only well-known employment and unemployment as well as providing comprehensive information on other standard indicators characterizing the country labour market. It covers those age 10 and over in the whole Kingdom. Information includes age, sex, activity, current and usual employment status, hours worked and wages and in addition included a seperate Food Insecurity Experiences Survey (FIES) questionniare module at the Household Level.
The conceptual framework used in this labour force survey in Tonga aligns closely with the standards and guidelines set out in Resolutions of International Conferences of Labour Statistician.
Kind of Data
Sample survey data [ssd]
Version 01: Clean, labelled and de-identified version of the Master file.
This version of the dataset contains Internation Labour Orgnization (ILO)'s variables required to compute some ILO indicators.
There are six statistical regions known as Division's in Tonga namely Tongatapu urban area, Tongatapu rural area, Vava'u, Ha'pai, Eua and the Niuas.Tongatapu Urban refers to the capital Nuku'alofa is the urban area while the other five divisions are rural areas. Each Division is subdivided into political districts, each district into villages and each village into census enumeration areas known as Census Blocks. The sample for the 2018 Labour Force Survey (LFS) was designed to cover at least 2500 employed population aged 10 years and over from all the regions. This was made mainly to have sufficient cases to provide information on the employed population.
Population living in private households in Tonga. The labour force questionnaire is directed to the population aged 10 and above. Disability short set of questions is directed to all individuals age 2 and above and the food insecurity experience scale is directed to the head of household.
Producers and sponsors
Authoring entity/Primary investigators
Tonga Statistics Department (TSD)
Technical Assistance Sampling Methodolgy
International Labour Organisation
Government of Tonga
International Labour Organisation
2018 Tonga Labour force survey aimed at estimating all the main ILO indicators at the island group level (geographical stratas).
The sampling strategy is based on a two stages stratified random survey.
1. Computation of the survey parameters:
Total sample size per strata, number of households to interview in each Primary Sampling Unit (PSU = census block) and number of PSUs to select
The stratification of the survey is the geographical breakdown by island group (6 stratas Tongatapu urban, Tongatapu rural, Vava'u, Ha'apai, 'Eua, Niuas)
2. The selection strategy is a 2 stages random survey where:
Random selection of census blocks within each
- Census blocks are randomly selected in first place, using probability proportional to size
- 15 households per block are randomly selected using uniform probability
3. The sampling frame used to select PSUs (census blocks) and household is the 2016 Tonga population census.
The computation of sample size required the use of:
- Tonga 2015 HIES dataset (labour force section)
- Tonga 2016 population census (distribution of households across the stratas)
The resource variable used to compute the sample size is the labour force participation rate from the 2015 HIES.
The use of the 2015 labour force section of the Tonga HIES allows the computation of the design effect of the labour force participation rate within each strata. The design effect and sampling errors of the labour force participation rate estimated from the 2015 HIES in combination with the 2016 household population distribution allow to predict the minimum sample size required (per strata) to get a robust estimate from the 2018 LFS.
Total sample size: 2685 households
Geographical stratification: 6 island groups
Selection process: 2 stages random survey where census blocks are selected using Probability Proportional to Size (Primary Sampling Unit) in the first place and households are randomly selected within each selected blocks (15 households per block)
Non response: a 10% increase of the sample happened in all stratas to account for non-response
Sampling frame: the household listing from the 2016 population census was used as a sampling frame and the 2015 labour force section of the HIES was used to compute the sample size (using labour force participation rate.
Deviations from the Sample Design
No major deviation from the original sample has taken place.
A total, 2,685 households were selected for the sample. Of these existing households, 2,584 were successfully interviewed, giving a household response rate of 96.2%.
Response rates were higher in urban areas than in the rural area of Tongatapu.
The household weight variable is called "hh_weight" while that of the person dataset is called "person_weight".
The weighting process started with the computation of the sampling weight at the household level.
The household weights were computed using the inverse of the factor of the probability of selection of the PSU and the probability of selection of the household within the selected PSU.
It appeared clearly that the population does not match the population structure at the youngest age groups.
This is why this first set of household weights was adjusted to match the population age structure (using the Stata command sreweight).
-2018 Labour Force Survey population:
Ongo Niua: 1,326
Dates of Data Collection (YYYY/MM/DD)
Main data collection
Mode of data collection
Computer Assisted Personal Interview [capi]
4 permanent staff of the Tongan Statistics Department were assigned as Headquarters for administration supervision of each Island Division. Twelve field supervisors were assigned to a team of roughly three enumerators to manage and supervise during the field work.
8 teams/supervisors were assigned for Tongatapu to manage 32 Enumerators while the other four supervisors were assigned to managing each of the other four statistical regions, including Vavau with 2 supervisors and 8 enumerators, Ha'apai 1 supervisor and 5 enumerators, Eua 1 supervisor managing 4 enumerators Ongo Niua was managed by two supervisors also assiged to Tongatapu along with 3 enumerators.
Type of Research Instrument
The 2018 Tonga Labour Force Survey questionnaire included 15 sections:
SECTION B: INDIVIDUAL CHARACTERISTICS
SECTION C: EDUCATION (AGE 3+)
SECTIONS B & C: EMPLOYMENT IDENTIFICATION AND TEMPORARY ABSENCE (AGE 10+)
SECTION D: AGRICULTURE WORK AND MARKET DESTINATION
SECTION E1: MAIN EMPLOYMENT CHARACTERISTICS
SECTION E2: SECOND PAID JOB/ BUSINESS ACTIVITY CHARACTERISTICS
SECTION F: INCOME FROM EMPLOYMENT
SECTION G: WORKING TIME
SECTION H: JOB SEARCH
SECTION I: PREVIOUS WORK EXPERIENCE
SECTION J: MAIN ACTIVITY
SECTION K: OWN USE PRODUCTION WORK
FOOD INSECURITY EXPERIENCES
GPS + PHOTO
The questionniares were developed and administered in English and were translated into Tongan language. The questionnaire is provided as external resources.
The draft questionnaire was pre-tested during the supervisors training and during the enumerators training and it was finally tested during the pilot test. The pilot testing was undertaken on the 27th of May to the 1st of June 2018 in Tongatapu Urban and Rural areas. The questionnaire was revised rigorously in accordance to the feedback received from each test. At the same time, a field operations manual for supervisors and enumerators was prepared and modified accordingly for field operators to use as a reference during the field work.
Tonga Statistics Department
The World Bank Survey Solutions software was used for Data Processing, STATA software was used for data cleaning, tabulation tabulation and analysis.
Editing and tabulation of the data will be undertaken in February/March 2019 in collaboration with SPC and ILO.
Data entry was made using Survey Solutions. A Computer-Assisted Personal Interview (CAPI) technology developed by the World Bank. The LFS survey was conducted on low-cost Android (Samsung) tablets.
Estimates of Sampling Error
Sampling errors were computed and are presented in the final report.
The sampling error were computed using the survey set package in Stata. The Finite Population Correction was included in the sample design (optional in svy set Stata command) as follow:
- Fpc 1: total number of census blocks within the strata (variable toteas)
- Fpc 2:
Here is a list of some LF indicators presented with sampling error
Labour force population: 2.2%
Employment - population in employment: 2.2%
Labour force participation rate (%): 1.7%
Unemployment rate (%): 13.5%
Composite rate of labour underutilization (%): 7.3%
Youth unemployment rate (%): 18.2%
Informal employment rate (%): 2.7%
Average monthly wages - employees (TOP): 12%.
Labour force population: 28,203 => 30,804
Employment - population in employment: 27,341 => 29,855
Labour force participation rate (%): 45.2% => 48.2%
Unemployment rate (%): 2.2% => 3.9%
Composite rate of labour underutilization (%): 16% => 21.4%
Youth unemployment rate (%): 5.7% => 12.1%
Informal employment rate (%): 44.3% => 49.4%
Average monthly wages - employees (TOP): 1,174 => 1,904.
"Tonga Statistics Department, Labour Force Survey 2018 (LFS 2018), version 01 of the licensed dataset (January 2019), provided by the Pacific Microdata Library. https://microdata.pacificdata.org/index.php/home"
Disclaimer and copyrights
The user of the data acknowledges that the Tonga National Statistics Department as original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses of the data.
Version 01 (January 2019): This is the first attempt at documenting the 2018 Togan Labour Force Survey (LFS). Done by Tonga Statistics Department (TSD), at Nuku'alofa, Tonga.
Version 02 (April 2019): This is the review of the existing documentation of the 2018 Togan Labour Force Survey (LFS). Done by Statistics for Development Division (SDD), at Noumea New Caledonia.