The Central Statistics Office (CSO) conducted its eighth Household Budget Survey (HBS) from July 2006 to June 2007 in the Republic of Mauritius. Previous surveys were conducted in 1961/62, 1975, 1980/81, 1986/87, 1991/92, 1996/97 and 2001/02. The first Household Budget Survey (HBS) was taken in 1961/62 under the guidance of Mr. Wolf Scott, an expert from the International Labour Office (ILO) and the second in 1975 with the assistance of an Indian Expert, Mr D.S Ramaratnam. As from the third round in 1980/81, HBS have been carried out regularly every five years by local staff.
The Mauritius Household Budget Survey (HBS) was carried out between July 1, 2006 and June 30, 2007 by the Central Statistics Office of Mauritius. The survey was conducted among a sample of 6,720 private households representative of all households in the islands of Mauritius and Rodrigues, with 560 households surveyed each month. The main objective of the survey was to obtain up-to-date information on the consumption pattern of the Mauritian population mainly with a view to update the basket of goods and services used for the computation of the monthly Consumer Price Index (CPI). The HBS also provides data on the distribution of household income and expenditure and is widely used, in that respect, for poverty analysis.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
The scope of the study includes:
- Household: demographic and educational characteristics of household members, activity status of each household member, dwelling, fuel used for cooking, household possessions, regular expenses, and irregular expenses over the past 12 months.
- Expenditure: goods and services obtained free or at reduced prices, own consumption goods from house garden, consumption of own produced goods, sale of own produced goods from house garden, and goods consumed from own shop, tabagie, store, type of outlets.
- Income: employee income, income from self-employment, property income, like withdrawal from savings, inheritances, transfer income.
The HBS 2006/07 covered all private non-institutional households in the Republic of Mauritius, i.e. in both the islands of Mauritius and Rodrigues. Households of non-residents as well as institutional households such as hotels, hospitals, boarding houses and prisons were excluded.
Producers and sponsors
Ministry of Finance and Economic Empowerment
Central Information System Division
Ministry of Information Technology
Government Printing Office
A sample of 6,720 households, out of an estimated total 335,000 households, was selected for the HBS 2006/07. The sample comprised two separate samples, one of 6,240 households (out of 325,000) for the Island of Mauritius and another of 480 households (out of 10,000) for Rodrigues. As the number of households in Rodrigues was smaller, a larger sampling fraction was used in order to generate reliable estimates.
Each sample was selected through a two-stage design with probability proportional to size. At the first stage, clusters (comprising around 100 households) were selected with probability proportional to size; this was followed by selection of households within these selected clusters.
The HBS 2006-07 spanned over 12 months to account for seasonal fluctuations in consumption. At the same time, the fieldwork was spread to a more manageable size. Each month, 560 households were surveyed, of which 520 in the Island of Mauritius and 40 in Rodrigues.
The sampling procedures are more fully described in "Mauritius Household Budget Survey 2006-07 - Methodological Report" pp.27-30.
Dates of Data Collection
Data Collection Mode
The Household Budget Survey involved the collection of a huge amount of data. Therefore, to ensure the quality of the data, field supervision and control were vital.
The Chief Supervisor was responsible for the whole field operation. He was appointed for a period of 14 months as from June 2006. His main responsibilities were:
(a) to be responsible for fieldwork in the islands of Mauritius and Rodrigues,
(b) to train field staff,
(c) to ensure that the Senior Supervisors performed their duties satisfactorily and to deal with unforeseen problems,
(d) to control the quality of the fieldwork.
These duties involved about 60 hours of work each month. The Chief Supervisor received a fee of Rs 13,000 and a travelling allowance of Rs 6,000 per month.
Three Senior Supervisors were appointed for 14 months as from June 2006 to assist the Chief Supervisor in monitoring the implementation of field procedures and instructions. Each Senior Supervisor was in charge of 4-5 Supervisors and 20-25 Interviewers for each survey month. The main duties of the Senior Supervisors were:
(a) to conduct/attend briefing and training sessions,
(b) to ensure that all field staff under their responsibility performed their duties as required and within established time limits,
(c) to control the quality and timeliness of the work of their respective Supervisors,
(d) to settle difficult cases.
These duties required about 75 hours of fieldwork per month. Each Senior Supervisor received a fee of Rs12,000 and a travelling allowance of Rs5,000 per month.
Fourteen Supervisors were appointed during each survey month. A Supervisor had to monitor and control the work of 5 Interviewers. His main duties were:
(a) to attend briefing and training sessions,
(b) to make a reconnaissance of his allocated region with his Interviewers,
(c) to transcribe the data obtained from the listing of households on a data sheet,
(d) to control the quality and timeliness of the work of his Interviewers,
(e) to control the flow of documents from the office to his Interviewers and back.
Each Supervisor furnished some 100 hours of work, spread over eight weeks. The Supervisor was remunerated at the rate of Rs14,500 for a survey month; he was also given a travelling allowance of Rs3,200 or Rs4,000 depending on the spread of regions allocated to him.
Data Collection Notes
Fieldwork was carried out after office hours and especially during weekends when working members of households were deemed more conveniently available. At the same time, it ensured better control over the fieldwork due to the simultaneous presence of interviewers and supervisors on the workplace to promptly solve problems and queries. Five grades of field staff were employed for the survey.
All the field staff was recruited from government employees. The Public Service Commission appointed the Chief Supervisor, the Senior Supervisors and the Co-ordinator on the recommendation of the Director of Statistics. As regards the other grades, the Commission delegated its authority to the Director of Statistics for their recruitment. This was done on the advice of a departmental selection board according to pre-defined criteria as approved by the Commission.
Each Interviewer was allocated eight households and had to maintain contact with them over a series of visits as follows:
(a) During the first contact, about 10 days prior to the beginning of the survey month, the Interviewer met the head of household and explained about the nature and objectives of the survey. During the same visit he/she filled in the HBS 2 schedule, explained the method of completion of the expenditure diaries (HBS 3 schedule) and handed the first week diary to the head. The head of the household was requested to fill in the diary everyday and to include all expenditures made by all members whether at home or away from home.
(b) During the first week of the survey month, the Interviewer regularly visited the household in order to ensure that the diary was properly filled in.
(c) At the end of each week of the survey month, the Interviewer returned to the selected households to collect the completed diaries for the past week and remit diaries for the coming week while at the same time settling queries, if any, on the data provided. More frequent visits were made to those households who could not themselves fill in the diaries.
(d) At the end of the last week of the reference month, the household was interviewed on the HBS5 schedule regarding points of purchase.
(e) Data on income were collected during the first week of the following month at the last stage of the survey after all other schedules have been verified and collected. The income schedule HBS4 was filled in for all members of the household who received any type of income. In the case of income, great care and tact was needed on the part of the interviewers since respondents are usually reluctant to reveal their actual income.
Central Statistics Office
Ministry of Finance and Economic Empowerment
The 2006/07 HBS necessitated the use of five questionnaires to collect all the necessary information from private households. The questionnaires are: (a) HBS 1 – Listing schedule, (b) HBS 2 – Household schedule, (c) HBS 3 – Daily record of the household expenditure, (d) HBS 4 – Income schedule, and (e) HBS 5 – Point of purchase questionnaire.
The HBS 1 - Listing schedule was used to draw a list (frame) of all households in each selected PSU. The following details were collected for each household: (a) name of head of household, (b) size of the household, (c) number of income earners, (d) occupation of main income earner, (e) average monthly household expenditure, and (f) religion.
The above data were used to classify households in socio-economic strata so as to select a representative sample of eight households in each cluster for interview and follow-up.
The HBS 2 - Household schedule was used to collect information on the characteristics of the selected households and their members. Data collected on the HBS 2 schedule covered: (a) demographic and educational characteristics of household members, (b) activity status of each household member, (c) characteristics of the dwelling, (d) fuel used for cooking, (e) household possessions, (f) regular expenses, and (g) irregular expenses over the past 12 months.
The HBS 3 - Daily record of household expenditure schedule was the most important document used for data collection. It was used to record detailed daily household expenditure on every item purchased during the whole survey month. When consolidated, this provided itemwise expenditure for the whole month for each household. Though meant to collect data on expenditure, the following information was also collected: (a) goods and services obtained free or at reduced prices, (b) own consumption goods from house garden, (c) consumption of own produced goods, (d) sale of own produced goods from house garden, and (e) goods consumed from own shop, tabagie, store, etc.
Since expenditure was incurred by any member of the household, a pocket pad was provided to each and every member aged 15 years and above. At the end of each day, the head of the household consolidated the data for all his members onto the diary (HBS 3 schedule).
The form HBS 3B comprised an introductory letter addressed by the Director of Statistics to the head of the selected household, explaining the purpose of the survey and seeking for his co-operation. It also contained a set of instructions with examples for filling in the expenditure diary HBS 3.
The HBS 4 - Income schedule was used to collect data on the income of each income earner of the household. This schedule was completed at the end of the reference month. It served as a check on the total expenditure of the household. Information collected included employment characteristics such as industry, occupation, employment status as well as: (a) employee income, (b) income from self-employment, (c) property income, (d) other receipts( like withdrawal from savings, inheritances, etc), (e) transfer income.
Besides income, information relating to debt repayment was also collected in this schedule.
The HBS 5 - Point of purchase questionnaire was used to collect information on the type of outlets where households usually purchase consumption goods and services and was completed at the end of each survey month. The relevant and important outlets and regions for the collection of price quotations to be used for the computation of monthly Consumer Price Index are derived there from.
Data entry and processing went on in parallel with the field work and was completed by August 2007. The coded schedules were sent to the Central Information Systems Division (CISD) for data capture using the software Integrated Microcomputer Processing System (IMPS). The monthly data files were sent back to the Central Statistics Office (CSO) where automated data consistency checking procedures were run on the input data using the same software. Any errors identified by these programs were corrected by a team at the CSO. Additional consistency checks and cleaning continued until March 2008 to produce a cleaner data set that will be used for detailed analysis.
Estimates of Sampling Error
The HBS 2006-07 estimates were based on a sample of possible observations. Hence, they were subjected to sampling variability and estimates could differ from the figures that would have been produced if information had been collected from all households in the population.
In order to check the representativeness of the sample used for HBS, some demographic data (age, sex, marital status and activity status) and other household characteristics (average household size and average monthly income) were compared with those obtained at the Continuous Multipurpose Survey (CMPHS 2006) which is a survey for estimating Labour Force, Employment and Unemployment.
It can be observed that the HBS results for demographic data and average household size are consistent with the CMPHS results. The differences observed might be accounted for by the difference in the time period to which the figures relate. Different samples at the two surveys may also contribute to the difference.
Before using the survey results to derive the CPI weights, they were checked against estimates derived from other sources such as production, imports, exports and local sales. It is to be noted however, that these data may have different coverage, i.e. they may include consumption by non-private households such as hotels and consumption by small businesses and trade. The survey data on sugar and chicken are less than the sales figure. This is mainly due to the fact that a large quantity of sugar goes in manufacture of sweet products and beverages for the informal sector while an important quantity of chicken is used in fast food. Moreover, the difference between estimated private household consumption and national consumption based on production data of cooking oil and onion can be explained by their use in the preparation of food (briani, mine/riz frit, gateau piments etc.) by vendors in the informal sector.
The reliability of survey data is more fully described in "Mauritius Household Budget Survey 2006-07 - Methodological Report" pp.39-44.
The World Bank Microdata Library
The World Bank
The HBS 2006/07 was conducted according to the provisions of the Statistics Act No. 38 of 2000. The regulations were made by the Minister on 14 April 2006 and were published in the Government Gazette as Government Notice No. 41 of 2006.
The Statistics Act provides for an obligation on the selected households to furnish the required information according to the approved questionnaires and documents. It also lays down strict rules for the CSO and its employees (including temporary ones) to ensure that all information collected are kept strictly confidential. All persons employed for the survey had to make a declaration of secrecy before a magistrate or the Director of Statistics.
Public use files, accessible to all upon request - anonymised files; Fees to be determined.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
Statistics Mauritius. Mauritius Household Budget Survey (HBS) 2006-2007. Ref. MUS_2006_HBS_v01_M. Dataset downloaded from http://statsmauritius.gov.mu on [date].
Disclaimer and copyrights
The user of the data acknowledges that Statistics Mauritius bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
DDI Document ID
Development Data Group
The World Bank
Documentation of the DDI
Date of Metadata Production
DDI Document version
Version 01: Adopted from "DDI_MUS_2006_HBS_v01_M" DDI that was done by metadata producers mentioned in "Metadata Production" section.