DDI_COK_2005_HES_v02_M Scott Pontifex <_value> NADA Version 02 (February 2015). Edited version 01 (SPC Documentation in collaboration with ESCAP (Jan Smit) Mission, June 2008), the initial version of DDI was done by Secretariat of the Pacific Community. The following DDI elements are edited/added, DDI ID, Study ID, Sampling Procedure, Descriptions of Questionnaires, Note on Data Collection, and Data Processing. Household Expenditure Survey 2005-2006 HES 2005-06 COK_2005_HES_v01_M Cook Islands Statistics Office Taggy Tangimetua Michelle Asiake NADA Asian Development Bank TA 6009-REG: Statistics Office Income/Expenditure/Household Survey [hh/ies] The 2005 Household Expenditure Survey (HES) is the third comprehensive survey of household expenditure in the Cook Islands, following on from the 1986 and 1998 surveys. Previously HES were carried out in Rarotonga only. consumption/consumer behaviour [1.1] income, property and investment/saving [1.5] community, urban and rural life [13.1] The Household Expenditure Survey (HIES) is a critical component of the work of the Cook Islands Statistics Office. The survey provides major indicators of the economic health of the nation and the well being of its people. For example Cook Islands residents (as opposed to businesses, government or tourists) consume a significant proportion of goods and services available in the Cook Islands. Changes in consumer and household spending patterns reflect changes in the economy and in individual and social behaviour. It provides the data which is necessary for the periodic revision of weights for the Office's Consumer Prices Index and other indices. The objectives of the HES are as follows: • To construct a revised weighting pattern for the Consumer Price Index • To provide data regarding the distribution of expenditure and wealth within the Cook Islands. • To provide information on subsistence economic activities and other unpaid work in the household. • To provide a gender-specific database to determine the level of men and women’s participation in economic activities. • To provide information for the estimation of the household consumption in the national accounts. • To undertake some poverty analysis and measure relevant Millennium Development Goals (MDGs), where applicable. Cook Islands The 2005 HES aimed to cover the entire country however not all households were covered in the survey due to the cost of enumeration in the outer islands. Remote islands that were not surveyed were represented in the survey by nearby islands. For example original plans were to survey Pukapuka and have this island representing Nassau and Suwarrow. The five cyclones that hit Cook Islands in early 2005 left Pukapuka heavily damaged in turn significantly affecting expenditure patterns. As a result, Pukapuka was not enumerated and the Penrhyn and Manihiki samples were increased to cover all the Northern Group. - Individuals - Households The survey covered all private households that have resided in the Cook Islands for 12 months or more. It did not cover non-private dwellings and institutions. The HES collected a wide range of demographic information, private dwelling information, as well as a fine level of detail on the expenditures of households. Unlike the previous survey the current HES excluded questions on income & time use due to the poor response from respondents. Further health information was collected in this survey. Literacy information was collected for the first time in this survey to gauge the need for further in-depth literacy surveys. Households in the sample were asked to provide information over just a two week period The dates of the survey periods were as follows: 1. Southern Group: 30 May 2005 – 12 June 2005 2. Rarotonga: 20 June 2005 – 17 July 2005 3. Penrhyn: 28 March 2006 – 11 April 2006 4. Manihiki: 26 June 2006 – 10 July 2006 Sampling Frame The frame used for the selection of households for the survey was based on the 2001 Census of Population and Dwellings. This list of dwellings was updated for all Census Districts (CDs) to incorporate any changes that may have occurred between the Census and the time of survey. In the Cook Islands Statistics Office (CISO), CDs are used as census collection units and are defined as small geographic units within Administrative Boundaries. There are roughly 100 households in each CD. Sample Selection The sample strategy used for the survey was a stratified two-stage systematic sample. The first stage involved the stratified selection of islands, and the second stage involved the systematic selection of households within each selected island. Stratification Like the previous HES the plan was to conduct the survey for the three regions in the Cook Islands. Based on the characteristics of the islands, they were grouped (like with like) to form the following five strata: 1. Rarotonga Southern Group 2. Aitutaki 3. Mangaia 4. Atiu, to represent the Ngaputoru islands this includes Mitiaro & Mauke Northern Group 5. Penrhyn, Manihiki to represent Rakahanga, Pukapuka, Nassau and Suwarrow. The original plans were to survey Pukapuka and have this island representing Nassau and Suwarrow. The five cyclones that hit Cook Islands in early 2005 left Pukapuka heavily damaged in turn significantly affecting expenditure patterns. As a result, Pukapuka was not enumerated and the Penrhyn and Manihiki samples were increased to cover all the Northern Group. Face-to-face [f2f] Four types of questionnaires were used for this survey: - HES04/1: Field Control List - this is to indicate households to be contacted by enumerators, and to record all visits made by the enumerator. - HES04/2: Household Control List - provides details of the household members (age, sex, relationship activity status, etc). - HES04/3: Weekly Diary - two books are given to each household. This consists of sections for items bought, home consumption of own produce, gifts of cash and goods given and received by the household. - HES04/4: Expenditure Questionnaire - consist of sixteen sections on various expenditure items. The overall organisation and control of the HES was carried out under the Statistics Act 1966, under the authority of the Government Statistician. The Government Statistician appointed supervisors for each Census District surveyed. In total there were 16 supervisors, 10 of whom were in Rarotonga. Supervisors in turn appointed enumerators for their areas. There were 41 enumerators in Rarotonga and 40 in the outer islands. In Rarotonga, the supervisors for the Northern Group were trained in the conduct of the HES survey and in the administration of the questionnaires. After returning to their home islands, the supervisors trained the enumerators in the administration of the questionnaires. A list of households was provided to each enumerator, which they asked to record their daily expenditure and their annual expenditure. A detailed instruction manual for Supervisors and Enumerators was developed, including the Interviewer’s Control Sheet and Field Book. A household weight was calculated for each EA using a simple formula with inputs including the population of the stratum, number of households in the EA and the number of households in the sample. For details on the calculation of household weights see the survey report, Annex B. The completed HES questionnaire and diaries were returned to the Statistics Office by the supervisors, where they were manually checked by staff and codes allocated for each dairy item. The forms were checked for completeness, validity and consistency to ensure the data was ‘fit for purpose’. Trained data entry operators entered the completed HES forms into an Access database. The database form included range checks and lookup tables to ensure accuracy of the data entry operation. Corrections to the data were made according to the guidelines in the editing and coding manual. Following completion of the data entry, check queries were run to ensure consistency of responses, and removing duplications between diary and household expenditure items. Derived variables, such as household weights and imputed rents, were also calculated and added to the database. A total of 913 households were randomly selected from 4,046 households to participate in the survey. Only 773 were eligible with the overall response rate of 89%. Under Reporting Experience has shown that certain expenditure items are not well represented in HES data. These items include alcohol and tobacco, home consumption of own home grown fruits and vegetables etc, meals taken away from home, and expenditure on items for which children are typically responsible e.g. confectionery, ice cream and beverages. 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.