The Household Income and Expenditure Survey (HIES) 2015 is the third of its kind. Previous HIES was conducted in 2001 and 2009. TSD adopted the Pacific standard HIES methodology developed by the Statistics for Development Division (SDD) of the Pacific Community (SPC) in 2011 to conduct Tonga 2015/2016 HIES. These standards, were designed to feed high-quality data to HIES data end users. The field collection spread over a 12 months period (October 2015 to October 2016 and divided to 16 Rounds) to cover seasonality of income and expenditure. This was the first time ever that HIES is conducted in such a way in Tonga also new in this HIES is the inclusion of Deprivation Section. Previous HIES was conducted over four rounds – once every quarter.
Household Income and Expenditure Survey (HIES) collects a wealth of information on household income and expenditure, such as source of income by industry, HH expenditure on goods and services, and income and expenditure associated with subsistence production and consumption. In addition to this, HIES collects information on sectoral and thematic areas, such as education, health, labour force, primary activities, transport, information and communication, transfers and remittances, food expenditure (acquisition) and gender.
The Pacific Islands regionally standardized HIES instruments and procedures were adopted by Tonga Statistics Department (TSD) for the 2015/2016 HIES. These standards, were designed to feed high-quality data to HIES data end users for: deriving expenditure weights and other useful data for the revision of the CPI, supplementing the data available for use in compiling official estimates of various components in the System of NA, supplementing the data available for production of the balance of payments; and gathering information on poverty lines and the incidence of poverty in Tonga.
The 2015/2016 HIES was conducted to update the 2009 HIES data and aimed to estimate the total amount HH spent and earnt over the past 12 months at the national and island group level (total expenditure and income).
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Individuals and Households.
Version 01: Cleaned, labelled and anonymized version of the Master file.
This data set consists of the aggregated income and expenditures reported during the survey from either modules or diary.
The scope of the Household Income and Expenditure Survey includes:
-INDIVIDUAL: Demographic Information, Labour Force Status, Education Status, Health Status, Communication Status, Household members who left the household
-HOUSEHOLD:Housing characteristics, Housing tenure expenditure, Utilities and Communication, Land and Home, Household goods and assets, Vehicles and accessories, Private travel details, Household Services Expenditure, Cash contribution to special occassions, Provision of financial support, Household Assets insurance and taxes, Personal Insurance, Deprivation.
Household Income and Expenditure Survey (HIES) covered all persons who were considered to be usual residents of private dwellings (must have been living in Tonga for a period of 12-months, or have intention to live in Tonga for a period of 12-months in order to be included in the survey).
Producers and sponsors
Tonga Statistics Department
Government of Tonga
Statistics for Development Division
Pacific Community (SPC)
Technical assistance in questionnaire design, sampling methodology/selection, data collection, data processing and data analysis
Government of Tonga
Provision of technical assistance through out the survey
The 2 stages sample method used in the 2015 Household Income and Expenditure Survey (HIES) aims to select randomly: Census blocks (with probability proportional to size - each blocks will have a different probability of selection) - based on the 2011 population census; and households (HHs) (within each selected blocks all the HHs have the same probability of selection) - based on a update of the HH listing conducted by the HIES field team.
To make the probability of selection more even, some small census blocks were merged and some large census blocks split. There's more detail in the methodological report on this process.
Before each round, the field teams updated the HH listing in each randomly selected block (stage 1 sample selection) and, after the listing was updated, the team randomly selected 18 HHs, which were 12 HHs as primary target HHs to interview; and 6 HHs in case that a primary selected HH cannot participate and a replacement is needed (e.g., refusal, absence, etc.) - this is done in order to achieve a high response rate.
Deviations from the Sample Design
Two-stage selection is used in Tongatapu (urban and rural), Vava’u, Ha’apai and ‘Eua, with the selection of census blocks (merged and split) in the first stage and the selection of households (HHs) in the second stage. HHs in Ongo Niua were selected directly from the updated HH listing (one-stage).
The table below shows the response rates by strata:
-Tongatapu - urban: 99.8%
-Tongatapu - rura: 99.8%
-Ongo Niua: 99.0%
Tonga 2015 Household Income and Expenditure Survey (HIES) was conducted between October 2015 to October 2016, just before the Tonga 2016 population census (November 2016), and the HIES population aggregates are dated to mid-point of field operation (1st of April 2016). HIES demographic aggregates have to match the population census figures and in this regard, the household weights can be adjusted.
Using the November 2016 census population figures, an estimate of the April 2016 population figures was derived in order to compare with the HIES population estimates.
Dates of Data Collection
Data Collection Mode
There were 10 supervisors managing 10 teams in the field. 5 teams in Tongatapu, 2 teams in Vava'u and 1 team in 'Eua, Ha'apai and Ongo Niua.
The role of the supervisor was to coordinate field data collection and activities, including management of the field teams, maps and listings and data entry and sent completed questionnaires and progress reports to the Household Income and Expenditure Survey (HIES) Manager based in main Office.
Field visits was made by the HIES Manager on a preiodic basis during fieldwork.
After each round, data were sent to the Pacific Community (SPC) for monitoring and data quality assessment.
Data Collection Notes
The pre-test for the survey took place two weeks prior to the training of field staffs, interviewers during the pre test were later become supervisors for the main survey.
A two-week training course was held in Nuku’alofa in which all field workers from the outer islands participated together with those from Tongatapu. Around 40 people were trained in which 29 were immediately used as supervisors and enumerators and the rest were reserved as replacements for those who left during the survey period.
The survey was spread over 16 rounds throughout the 12 month enumeration period (October 2015 to October 2016) in order to take adequate account of seasonal fluctuations in income, expenditure and food acquisition (as a proxy to estimate consumption). Each round lasted for a 3-week period and 120 households (HHs) across Tonga were interviewed each round (12 HHs per team). One team consists of a supervisor and 2 enumerators.
Field teams updated the HH listing for each selected census block - and made the HH selection (second stage) - on day one and first contact was made with the HH on day two or three. Each HH was visited several times during the three-week round in order to conduct module interviews and to check that the two-week expenditure and home production diary was being comprehensively completed. Data entry occurred on the same day or the day after the interview for the modules, while the first diary was entered in week-two and the second week diary is entered in week-three of each round. In-field data entry was conducted in order to make use of the entry validation checks and to return to the HH to verify data where inconsistencies were identified.
Interviews were primarily in Tongan, but English was occassionally used for those who did not speak Tongan.
Tonga Statistics Department
Government of Tonga
The use of a common questionnaire developed by the Statistics for Development Division (SDD) of the Pacific Community (SPC) was adopted by Tonga Statistics Department (TSD) to conduct Household Income and Expenditure Survey (HIES) 2015. Addition to the 4 Modules was a section on Deprivation.
4 modules to collect socio-demographic information, and expenditure and income; and o a two-week diary to collect daily expenditure, gifts received and home produced items Four modules are completed by paper-based personal interview, including:
1. Demographic information – characteristics of household (HH) members, including activity and education profile;
2. Household characteristics and expenditure (Housing characteristics, Housing tenure expenditure, Utilities and communication...etc);
3. Individual expenditure (Education, Health, Clothing, Communication...etc);
4. Individual and HH income (Wages and salaries, Agricultural and Forestry activities, Fishing, gatehring and hunting activities...etc).
Depending on the information being collected, a recall period (ranging from the last 7 days to the last 12 months) is applied to various sections of the questionnaire.
The forms were completed by face-to-face interview, usually with the Household (HH) head providing most of the information, with other HH members being interviewed when necessary. The interviews took place over a 2-week period such that the HH diary, which is completed by the HH on a daily basis for 2 weeks, can be monitored while the module interviews take place.
The HH diary collects information on the HH’s daily expenditure on goods and services; and the harvest, capture, collection or slaughter of primary produce (fruit, vegetables and animals) by intended purpose (home consumption, sale or to give away).
The 4 Modules were published in English but with a Tongan version that was made available to enumerators to help them with the interview. For the diaries they were published in both Tongan and English which ever version that the household find easy to fill in.
A first set of edits was done on the raw data via questionnaire checks, which was the first set of corrections. A summary of findings includes:
1820 questionnaires retrieved from the field in which 17 households with no diaries were removed. The process for the manual cleaning included checking the following information in the questionnaire: Labour force section (waged job) declared in module 1 and salaries declared in module 4 (income section). Imputation of wages in households: 30011, 40176, 40190, 40097. Electricity connection and payment of electric bill (module 2). Imputation of electricity bill in households: 20290, 20302, 20304, 30051. Households who are using butane for cooking and payment for butane.Ages and relationship to the household head.Check all the government pension in the salary section.
Check the remittances sent by household members who are currently picking fruits overseas (that have to be transferred in wages & salary section).Check if the household members who are in Australia/NZ for fruit picking have declared their resident status in the household accordingly (option 4). After the first clean of the raw data, 1803 valid questionnaires were kept.
Further edits that was done to the raw data set was done using the software STATA.
Data was entered into a laptop by supervisors for each of the 10 teams during the fieldwork in each round from manual questionnaires filled out during face to face interview. Data entry occurred on the same day or the day after the interview for the modules, while the first diary was entered in week-two and the second week diary is entered in week-three of each round. In-field data entry was conducted in order to make use of the entry validation checks and to return to the household (HH) to verify data where inconsistencies were identified.This systems gives the field staff the opportunity to correct the data by manual review and by using the entry system generated error messages.After each round (3 weeks) supervisors submited data entered together with paper questionnaire at the end of each round to HIES Manager who then sends these file to SPC for further checks.
Estimates of Sampling Error
Standard error, the relative sampling errors (RSE) and 95 percentage confidence interval were calculated for the total household expenditure, total household consumption expenditure, total household non consumption expenditure, total household cash expenditure, total household subsistence expenditure, total household expenditure on COICOP division 1, total household net income, total household net cash income, total net wages and salary cash income and total net primary cash income by strata. At the national level the RSE calculated was of good quality however caution should be made at the strata level.
Details of the sampling errors are presented in the sampling errors appendix 2 of the report presented in the external resources.
Non-sampling errors cannot be readily measured, however it is worth noting the issues associated with non-sampling errors, including:
both respondents and interviewers may not entirely understand the information required from the survey, which can result in misinterpretation of the question being asked and the incorrect response; enumerator and respondent fatigue, resulting in underreporting, especially in completion of the household (HH) diary; unwillingness to fully disclose information – especially in a small-island context - such as income and expenditure on some items (e.g., alcohol, tobacco and cash donations); the questionnaire being in English, which could be a second language for both the interviewers and respondents, and the need to complete a written diary (noting that: three-quarters of diaries were in Tongan; HHs were given the opportunity to complete a Tongan written diary; and enumerators could mostly converse in Tongan when required); and the inability to interview HHs members living abroad but remain dependent on the HH (e.g., students living in school dormitories) or are working to support the HH (e.g., seamen living on a ship), but who have not formed another HH outside of Tonga.
"Tonga Statistics Department, Household Income and Expenditure Survey 2015-2016 (HIES 2015-2016), Version 01 of the licensed datasets (February 2020), provided by the Microdata Library. https://microdata.pacificdata.org/index.php/home"
The user of the data acknowledges that the 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.
(c) 2018, Tonga Statistics Department (TSD)
DDI Document ID
Tonga Statistic Department
Government of Tonga
Documentation of the study
Government of Tonga
First review and final edit of the metadata
Statistics for Development Division (SDD)
Secretariat of the Pacific Community
Final review of the study
Date of Metadata Production
DDI Document version
Version 01 (January 2019): First attempt at documenting the 2015-2016 Tongan HIES. This documentation was done by Tonga Statistics Department (TSD) in Tonga.
Version 02 (February 2019): This is an update of the v01 documentation.This documentation was done by Statistics for Devlopment Division (SDD), Pacific Community (SPC) at Noumea, New Caledonia.
Version 03 (February 2020): This is the update of the v02 documentation. This is the documentation of the anonymized version of the Master file. Done by Statistics for Devlopment Division (SDD), Pacific Community (SPC) at Noumea, New Caledonia.