The purpose of the HIES survey is to obtain information on the income, consumption pattern, incidence of poverty, and saving propensities for different groups of people in FSM. This information will be used to guide policy makers in framing socio-economic developmental policies and in initiating financial measures for improving economic conditions of the people. The 2005 FSM HIES asked income of all persons 15 years and over. It referred to income received during the calendar year 2004, and includes both cash and in-kind income. The survey has five primary objectives, namely to:
1) Rebase the FSM Consumer Price Index (CPI);
2) Provide data on the distribution of income and expenditures throughout the FSM;
3) Provide data for national accounts, particularly regarding income from home production activities and the consumption of goods and services derived form home production activities;
4) Provide nutritional information and food consumption patterns for the FSM families; and
5) Provide data for hardship study.
Kind of Data
Sample survey data [ssd]
Version 2.0 - edited data; not anonymized; basis for public reports
consumption/consumer behaviour [1.1]
income, property and investment/saving [1.5]
community, urban and rural life [13.1]
Four states of the FSM: Yap, Chuuk, Pohnpei, and Kosrae
Unit of Analysis
- Expenditure items
The survey universe covered all persons living in their place of usual residence at the time of the survey. Income data were collected from persons aged 15 years and over while expenditure data were obtained from all household members at a household level. Persons living in institutions, such as school dormitories, hospital wards, hostels, prisons, as well as those whose usual residence were somewhere else were excluded from the survey.
Producers and sponsors
Authoring entity/Primary investigators
FSM Divison of Statistics
Office of Statistics, Budget and Economic Management, Overseas Development Assistance and Compact Management (SBOC)
US Census Bureau, International Programs Center
Technical assistance in overall project oversight
FSM Divison of Statistics
Technical assistance in sampling methodology/selection and data analysis
Secretariat of the Pacific Community
Technical assistance in data collection and data processing
Office of Insular Areas, US Department of the Interior
OIA, US DOI
The 2005 FSM Household Income and Expenditure Survey (HIES) used a sampling frame based on updated information on Enumeration Districts (ED) and household listing from the 2000 FSM Census. Based on this sampling frame, the four states of FSM were then classified as the domains of the survey. Each of the states was further divided into 3 strata, except for Kosrae which was not divided at all because it doesn't possess any outer islands and it has relatively good access to goods and services. The entire island was therefore classified under stratum 1. Each stratum was defined as follows:
1) State center and immediate surrounding areas:
- High 'living standard' and has immediate access to goods and services.
2) Areas surrounding state center (rest of main island):
- Medium 'living standard' and sometime limited access to goods and services
3) Outer islands:
- Low 'living standard' and rare access to goods and services.
Within each stratum, the HIES used a two-stage stratified sampling approach from which the sample was selected independently. First, enumeration districts (EDs) were drawn from each stratum using Proportion Probability to Size (PPS) sampling. Thus, the larger the ED size, the higher its probability of selection. About 69 EDs out of a total of 373 EDs were selected nationwide for the survey. Generally, one enumerator is assigned to each ED. Second, 20 households were systematically selected from an updated household listing for each of the selected EDs using a random start to come up with a total sample size of 1,380 households, or roughly 8.4 percent of all households in the state. Although it offered a fairly good representation of the total households in the nation, the final sample size showed a reduction of nearly 180 households from the 1,560 households, or 10 percent, initially selected for the survey.
Detailed information on the changes made to the sample size can be found in the next section under "Deviations from Sample Design."
Deviations from the Sample Design
The original plan to sample 1,560 households, or about 9.5 percent of all households in the nation was eventually reduced to 1,380 households, or about 8.4 percent of all households. The reduction of the sample size was due to fuel unavailability for transportation and uncertainty of field trip schedules to some of the selected outer islands. Dropping some of these islands from the sample was not expected to impact significantly on the accuracy of the survey results because independent weighting took place within each stratum, where islands were considered to be sufficiently homogenous.
Original Sample Size: 1,560 Households
Original Sampling fraction 9.5%
Final Sample Size: 1,380 Households
Final Sampling fraction 8.4%
The response rate for the final sample size of 1,380 households is 100 percent. The majority of households originally selected for the survey did respond to the survey. Households which have moved to other unselected areas or elsewhere and those who refused to respond were replaced with nearby households that were willing to participate in the survey.
Weights for the survey were calculated at the stratum level for each household. An estimate of the household population at the time of the survey was derived using population projections and average household size. These values were divided by the responding sample size for the stratum in order to produce the weight for the household.
One outlier was identified during the data editing phase and as a result the weight for this household was set to 1. The weights for the remaining households within that stratum were adjusted accordingly.
Dates of Data Collection (YYYY/MM/DD)
Drop-off Wk1 diaries for the 1st 10 HHs
Enumeration, pick-up of Wk1 diaries and dropping off diaries for Wk2
Drop-off Wk1 diaries for last 10 HHs, pick up of Wk2 Diaries for 1st 10 HHs
Enumeration, pick-up of Wk1 diaries and dropping off diaries for Wk2
Pick up of Wk2 Diaries for last 10 HHs
Mode of data collection
Supervision of the HIES data collection process, at its lowest level, occurred in the field with field supervisors overseeing the enumerator's interviewing process and monitoring their daily progress. These individuals worked under the direction of the State Survey Coordinator and the National Coordinator who trained and help them in any difficult situation. It is the responsibility of each field supervisor to work closely with his/her assigned enumerators as well as the State Survey Coordinator to ensure that the survey runs smoothly and efficiently without undue difficulties.
A field supervisor monitored a crew of about 3 to 5 enumerators previously assigned to similar number of Enumeration Districts (EDs). A total of about 15 field supervisors were contracted out for the 2005 FSM HIES. Of these 5 each were assigned to Chuuk and Pohnpei, 3 to Yap, and 2 to Kosrae. All field supervisors must accompany their enumerators for the first few interviews on the first few days to ensure that they understood the interviewing process and completing their assigned tasks completely and accurately. In this regard, the most important responsibility for a field supervisor was to make sure that all assigned tasks were completed accurately and on schedule. A field supervisor was also expected to keep the State Survey Coordinator informed of potential delays or problems. Specific details of each supervisor's tasks and responsibilities include:
1) Attending the training session for supervisors to learn enumeration procedures.
2) Learning the enumerators' job.
3 ) Participate in the training of your enumerators.
4) Supervise the Field Exercises during training.
5) Preparing and distributing enumerator assignments.
6) Conducting on-the-job training (as required).
7) Collecting enumerators' completed work.
8) Performing field reconciliation (if required) for a Block or ED.
9) Re-canvassing a Block or ED, if necessary.
10) Conducting first reviews, daily reviews, and final reviews of each enumerators' work.
11) Monitoring and reporting the progress of your staff using progress reports.
12) Visiting and enumerating households that refuse to cooperate with an enumerator.
13) Releasing enumerators who are unable to perform satisfactorily.
14) Reassigning work, when necessary. Those who have completed their assignment(s) may finish up the work of those unable to perform a satisfactory job.
15) Supervising and controlling the progress of enumeration in your assigned areas to ensure that work is completed on schedule.
16) Prepare a procedural history report for Phases I and II of the survey operation.
Type of Research Instrument
Questionaires and forms used for the 2005 FSM HIES consisted of 1) HIES Questionnaire and 2) Weekly Diaries. The HIES Questionnaire were provided to enumerators and should be filled out during the first visit to the household. Its main objective was to collect housing information, basic demographic information about members of the household, and general household expenditures over the previous year. On the other hand, the weekly diaries, was an attempt to record household expenditure on a daily basis over the course of a 2 week period. Both the HIES questionnaire and the weekly diary were developed and modeled after similar forms from the 1998 FSM HIES Survey and the 2004 Palau HIES Survey. Dr. Micheal Levin from the US Census Bureau, International Program Center (IPC), Ms. Brihmer Johnson of the FSM Division of Statistics and Mr. Glenn McKinlay, statistics advisor to FSM Division of Statistics, provided crucial inputs to the overall design of these forms. All questionnaires and diaries used during the HIES were printed in English so it was extremely important that field interviewers understand the instuctions and questions contained within. Testing of the questionnaire were carried out by FSM Division of Statistics staffs who conducted "real" interviews with certain households in their neighborhood as well as having their own household be interviewed by a different office staff. Specific sections for both the HIES questionnaire and the weekly diaries are outlined below:
I. HIES Questionnaire
1) General Household characteristics
2 ) Individual Person Characteristics
3) General Expenditure Listings - 12 Months Recall Period
II. 2 Week Daily Diaries
1) Daily Expenditure Diary - Day1 (Mon) thru Day7 (Sun)
2) Home Produced Items
3) Gifts Given Away
4) Gifts Received
5) Unusual Expenses for Special Events
FSM Divison of Statistics
Department of Economic Affairs, FSM National Government
FSM Division of Statistics
College of Micronesia Students
College of Micronesia (COM)
Data editing of the 2005 FSM HIES data occurred over several instances during the data processing phase of the project and afterwards prior to putting together the final report. After a two weeks office review and call backs right after the enumeration phase, the initial phase of data editing took place on July 18, 2005 when the data processing phase of the survey commenced. Training for editing and coding took place on the same day along with the signing of contracts for 10 office clerks recruited to carry out this phase of the survey. As part of their contract, these individuals were also hired to key in the data at a later time. One of their primary responsibily was to match geographic ids for questionnaire with corresponding diaries and ensure consistencies and valid entries accordingly. No computer consistency edit checks were run against the data during the keying/verification process since the programs for these processes were not available at the time. All data quality checks and edits were done at the US Bureau of Census. Further edits were applied to the data during the data analysis and report writing process.
There were five types of checks performed: Structural check, Verification check, Consistency check, Macro Editing check, Data Quality assessment. Edit lists were also produced for health module, income and expenditure questionnaire which needed to be checked against the questionnaires. On the edit list, corrections of errors were made by crossing out incorrect or missing values and entering the correct values in red. Missing amounts that were also missing on the questionnaire will need to be estimated using estimates from questionnaires in the same Enumeration District (ED) batch. For the diaries, the batch files were concatenated for each state and exported to tab delimited files. These files were imported into Excel and the unit price for each item was calculated using quantities and weights where possible. Records for each item were then filtered out and check for outlier unit price values (both large and small values as well as missing values). Values for missing amounts were imputed from estimated using average prices from the items within the same ED.
The office operations manual used for editing and coding the questionnaires and diaries is provided under "Technical Documents/Data Processing Documents/Office Editing & Coding."
The CSPro software was used for the data entry of the survey. The forms were keyed at the state branch offices with the exception of Chuuk. Due to the frequent power outages in Chuuk and shortage of staffs, its forms were shipped to Pohnpei for keying. Each state recruited or utilized some of their enumerators for the data processing. The number of Office Clerks contracted depended on the size of the work or number of selected Enumeration Districts (ED) in each state. Yap had 14 EDs, so two office clerks were recruited. Pohnpei and Chuuk, each had 23 (EDs), so six were recruited. Two of these office clerks came from Chuuk. Kosrae had only 7 (EDs) so only one was recruited and the only one who was not previously an enumerator.
Data entry for the HIES started on the 22nd of August 2005 and should continue on for 3 weeks ending on September 9. Training on the use of the data entry applications was included in this timetable. Instead, by the end of this contractual period none of the states have completed their assigned ED batches. Yap has completed about 62% of their batches, kosrae 57%, pohnpei only 39% while chuuk has not even keyed yet. Because of this and the upcoming keying verification, it was decided to extend the keyer's contracts for another two (2) weeks, starting on the 12th and ending on the 23th of the same month. Meanwhile the FSM Divison of Statistics began to to check the completed batches keyed, to ensure that both the questionnaire and diary totaled to 20 and 40 respectively. While doing this, some geographic id inconsistencies were corrected.
The actual keying in of the data involved using two separate CSPro data entry screen; one for HIES Questionnaire, and the other for the HIES Weekly Diaries. At different stages of the keying process, the completed ED batches were concatenated and sent over to the main office for consolidation. Since Chuuk and Pohnpei forms were keyed in Pohnpei, only the data from Yap and Kosrae were sent over from the states for overall concatanation.
The keying verification was schedule to begin on Septemeber 19th during the last week before the September 23th contract deadline. However, for Yap and Kosrae which had finished keying early, they actually started the keying verification a week early beginning September 12, 2005. The goal for this phase of the survey was to do a 50% verification of all the ED batches previously keyed in to correct data entry errors. Instead, the actual verification for the nation averaged about 37% of the total batches (Yap about 36%, Chuuk and Pohnpei each 35% and Kosrae 57% respectively). Except for Kosrae, Yap retained only one of its two contractors while chuuk and pohnpei depending on performance of the clerks, retained 4 of their office clerks.
Estimates of Sampling Error
No sampling error analysis of the survey was calculated.
The questionnaire design of the 2005 HIES vary from that of the 1998 HIES rendering comparison of the data to the 2005 HIES limited. However, when the data permits, comparisons were made.
Assistant Director, FSM Division of Statistics
Office of Statistics, Budget & Economic Management, Overseas Development Assistance and Compact Management
Section 108 of the "Statistics and Census Act of 1988" guarantees confidentiality of all information collected during the 2005 FSM HIES. Disclosure of any data related to the HIES Survey must get full approval from the FSM Division of Statistics. Any Person provided with such data shall not:
(1) Use the information furnished under the provisions of this chapter for any purpose other than for the purpose of conducting a census or sample survey, gathering statistical data, or for compiling and disseminating tabulations or other statistical materialsr;
(2) Make any publication whereby the data furnished by any particular establishment or individual under this chapter can be identified; or
(3) Permit anyone other than sworn employees of the FSM Division of Statistics to examine the individual reports, except in the prosecution of alleged violations of this Act.
Similar provsions are also provided under Public Law 5-77 which prohibits any FSM Division of Statistics employees, including contactual workers from revealing any information contained in any questionnaire or report obtained from respondents to anyone except authorized survey staff. An oath of confidentiality places restraint upon all survey staff and contractors to guard and keep personal records confidential. Any printed publications issued by the FSM Division of Statistics contains information only in summary form, and are prepared for the benefit of the public.
The 2005 FSM HIES dataset is not specifically covered under any data access policy except for related provisions stipulated in section 107 of Chapter 1 of the "Statistics and Census Act of 1988", entitled "Dissemination of Statistical Data, Restrictions on Use" which specifically states that:
"The Chief [of the National Statistics Office] may furnish copies of tabulations and other statistical materials which do not disclose the information reported by, or on behalf of, any particular respondent, and may make special statistical compilations or surveys, for departments, offices, and agencies of the National Government, and State and local governments." Likewise, any statistical information and reports furnished to the public should also be strictly anonymized to ensure its confidentiality.
Because the dataset itself has not been fully anonymized, persons wishing to access it should contact the FSM Division of Statistics directly and should abide by the following terms and conditions:
1. The data and other materials will not be redistributed or sold to other individuals, institutions, or organizations without the written agreement of the FSM Division of Statistics.
2. The data will be used for statistical and scientific research purposes only. They will be used solely for reporting of aggregated information, and not for investigation of specific individuals or organizations.
3. No attempt will be made to re-identify respondents, and no use will be made of the identity of any person or establishment discovered inadvertently. Any such discovery would immediately be reported to the FSM Division of Statistis.
4. No attempt will be made to produce links among datasets provided by the FSM Division of Statistis , or among data from the FSM Division of Statistics and other datasets that could identify individuals or organizations.
5. Any books, articles, conference papers, theses, dissertations, reports, or other publications that employ data obtained from the FSM Division of Statistics will cite the source of data in accordance with the Citation Requirement provided with each dataset.
6. An electronic copy of all reports and publications based on the requested data will be sent to the FSM Division of Statistis .
The original collector of the data, the FSM Division of Statistis , and the relevant funding agencies bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
Refrencing this dataset and associated materials should be cited as follows:
"FSM Divison of Statistics, 2005 FSM Household Income and Expenditure Survey, Version 2.0, http://www.spc.int/prism/data-catalogue"
Disclaimer and copyrights
The data contained herein is provided "as is" without any expressed or implied warranty or responsibility. The FSM Division of Statistics accepts no responsibility for the use of such data nor for any interpretations or inferences based upon such uses.