Mauritius CSD + do files

Support/B - Country Survey Details/MUS/CMPHS/1. Introduction to Mauritius CMPHS.md

@@ -0,0 +1,83 @@
+# Introduction to Mauritius Continuous Multi-Purpose Household Survey (CMPHS)
+
+- [What is the CMPHS?](#what-is-the-cmphs)
+- [What does the CMPHS cover?](#what-does-the-cmphs-cover)
+- [Where can the data be found?](#where-can-the-data-be-found)
+- [What is the sampling procedure?](#what-is-the-sampling-procedure)
+- [What is the geographic significance level?](#what-is-the-geographic-significance-level)
+- [Other noteworthy aspects](#other-noteworthy-aspects)
+
+## What is the CMPHS?
+
+The Continuous Multi-Purpose Household Survey (CMPHS) is a household survey conducted in Mauritius by the Central Statistics Office (CSO), now Statistics Mauritius. It is the main source of labour market statistics in the country, including labour force, employment, and unemployment. The first CMPHS was conducted in 1999, and from 2001 onwards it was conducted annually.
+
+## What does the CMPHS cover?
+
+The Mauritius CMPHS covers demographic characteristics, labour force, education and training, health, household conditions, environment, and individual income. The harmonized years and sample sizes used in the GLD harmonization are as follows:
+
+| Year | # Persons | # Households |
+|---|---:|---:|
+| 2001 | 24,539 | 6,300 |
+| 2002 | 24,552 | 6,300 |
+| 2003 | 24,393 | 6,480 |
+| 2004 | 32,626 | 8,640 |
+| 2005 | 41,894 | 11,280 |
+| 2006 | 41,559 | 11,280 |
+| 2007 | 41,845 | 11,280 |
+| 2008 | 41,121 | 11,280 |
+| 2009 | 40,385 | 11,280 |
+| 2010 | 41,103 | 11,280 |
+| 2011 | 41,671 | 11,280 |
+| 2012 | 41,699 | 11,280 |
+
+## Where can the data be found?
+
+The datasets are not publicly accessible. The World Bank has been granted access to the datasets. If you work or are part of the World Bank Group, kindly contact the Jobs Group with a formal request for access through gld@worldbank.org.
+
+## What is the sampling procedure?
+
+The Mauritius CMPHS uses a stratified two-stage sampling design. In the first stage, primary sampling units (PSUs) are selected with probability proportional to size. In the second stage, a fixed number of households is selected within each PSU.
+
+There were changes in the sample design over time. Before 2005, around 525 households were surveyed each month, resulting in about 6,300 households per year. These rounds were based on the 1999 master sample and covered private non-institutional households in urban, semi-urban, and rural areas in the islands of Mauritius and Rodrigues.
+
+From 2005 onwards, the annual sample size increased to 11,280 households. These rounds followed a 2-2-2 rotation sampling design, where half of the households interviewed in one quarter were re-interviewed in the following quarter in order to support quarterly estimates. 
+
+In the microdata available to the GLD team, however, only the first interview of each household appears to be available. As a result, the data support annual analysis, but not the intended quarter-to-quarter follow-up of households.
+
+## What is the geographic significance level?
+
+The official reports detail estimates by area of residence, namely urban and rural areas.
+
+## Other noteworthy aspects
+
+### Survey structure
+
+The CMPHS is designed as a continuous survey with quarterly fieldwork. However, the microdata available for GLD appear to retain only one interview per household. This is important for users because the survey design allows repeated interviews within the year, but the available data do not support the construction of quarterly household panels.
+
+### Weight variable
+The datasets received by the World Bank do not contain a household or individual weight variable. This applies to all harmonized CMPHS rounds currently available to the GLD team.
+
+Users should therefore be careful when interpreting levels and distributions from the data, as unweighted estimates are not expected to be consistent with official estimates. 
+
+### Industry classification
+
+The Mauritius CMPHS uses the National Standard Industrial Classification of Economic Activities (NSIC) to code industry. NSIC Revision 1, used from 2001 to 2010, is compatible with ISIC Revision 3 at the 2-digit level, while NSIC Revision 2, used in 2011 and 2012, is compatible with ISIC Revision 4 at the 2-digit level.
+
+| Year | Classification in survey | ISIC revision compatibility | Level of compatibility |
+|:----:|:------------------------:|:---------------------------:|:----------------------:|
+| 2001-2010 | NSIC Revision 1 | ISIC Revision 3 | 2 digits |
+| 2011-2012 | NSIC Revision 2 | ISIC Revision 4 | 2 digits |
+
+
+### Occupation classification
+
+The Mauritius CMPHS uses the National Standard Classification of Occupation (NASCO) to code occupation. The version used from 2001 to 2010 is compatible with ISCO 1988 at the 2-digit level, while NASCO-08, used in 2011 and 2012, is compatible with ISCO-08 at the 2-digit level.
+
+| Year | Classification in survey | ISCO revision compatibility | Level of compatibility |
+|:----:|:------------------------:|:---------------------------:|:----------------------:|
+| 2001-2010 | NASCO-99 | ISCO 1988 | 2 digits |
+| 2011-2012 | NASCO-08 | ISCO 2008 | 2 digits |
+
+### Education
+
+Information on educational attainment in the Mauritius CMPHS is recorded by year level, and the main education variable does not identify attainment beyond secondary schooling. In the GLD harmonization, post-secondary attainment is coded using a separate variable on field of study or qualification. More details are provided in a [separate document here](Education.md).

Support/B - Country Survey Details/MUS/CMPHS/Education.md

@@ -0,0 +1,69 @@
+# Education in the Mauritius CMPHS
+
+## Overview
+
+In the Mauritius CMPHS, educational attainment is primarily recorded by year level. The raw education attainment variable does not identify levels beyond secondary schooling in a way that is sufficient for GLD harmonization.
+
+To identify post-secondary attainment, the harmonization relies on a separate variable that records the respondent's qualification. This variable follows the National Standard Classification of Education (NSCED), Mauritius' adaptation of the International Standard Classification of Education (ISCED), and is used to distinguish higher levels of education.
+
+## Classification used in the qualification variable
+
+The qualification variable is based on NSCED for most years and on NSCED-97 in 2011 and 2012, corresponding to the adaptations of ISCED 1976 and ISCED 1997, respectively.
+
+| Year | Classification in survey | International compatibility |
+|:----:|:------------------------:|:---------------------------:|
+| 2001-2010 | NSCED | ISCED 1976 |
+| 2011-2012 | NSCED-97 | ISCED 1997 |
+
+## Mapping of qualification codes to GLD education categories
+
+The harmonization relies on the first digit of the qualification code to identify the broad education level.
+
+The qualification variable is used only to identify post-secondary attainment. Codes `0` to `3` are not used to construct `educat7`, since levels up to secondary are already identified from the main raw education variable. The qualification variable is used only for codes `4` to `8`, which distinguish post-secondary and university education.
+
+The table below summarizes the interpretation used in the GLD harmonization.
+
+| Code | ISCED level | Description | Qualification  | GLD harmonized category |
+|:----:|:-----------:|:------------|:----------------------|:------------------------|
+| 0 | 0 | Pre-primary |  | Not used in harmonization |
+| 1 | 1 | Primary | Certificate of Primary Education | Not used in harmonization |
+| 2 | 2 | Lower secondary | Pre-Voc (Yr I - IIII) | Not used in harmonization |
+| 3 | 3 | Upper secondary | National Trade Certificate (NTC) Level 3, Certificate | Not used in harmonization |
+| 4 | 4 | Post-secondary | NTC Level 2, Certificate | Higher than secondary but not university |
+| 5 | 5 | First stage of tertiary education | Diploma | Higher than secondary but not university |
+| 6 | 5 | First stage of tertiary education | Degree (BSc) | University incomplete or complete |
+| 7 | 5 | First stage of tertiary education | Master degree | University incomplete or complete |
+| 8 | 6 | Second stage of tertiary education | MPhil/PhD | University incomplete or complete |
+
+In the GLD harmonization, codes `4` and `5` are grouped as `Higher than secondary but not university`, while codes `6` to `8` are grouped as `University incomplete or complete`.
+
+## Sample coding logic
+
+Below is an illustration of the coding logic used to harmonize the qualification information into GLD education categories.
+
+```stata
+*<_educat7_>
+	gen byte educat7=.
+	replace educat7=1 if educationlevel==1 | educy==0
+	replace educat7=2 if educationlevel>=11 & educationlevel<=16
+	replace educat7=3 if educationlevel==17
+	replace educat7=4 if educationlevel>=21 & educationlevel<=26
+	replace educat7=5 if educationlevel>=27 & educationlevel<=31
+
+	* Use qualification variable for post-secondary attainment
+	gen qual1d = int(qualification/1000)
+	replace educat7=6 if inlist(qual1d, 4, 5)
+	replace educat7=7 if inlist(qual1d, 6, 7, 8)
+	drop qual1d
+
+	label var educat7 "Level of education 1"
+	la de lbleducat7 1 "No education" 2 "Primary incomplete" 3 "Primary complete" 4 "Secondary incomplete" 5 "Secondary complete" 6 "Higher than secondary but not university" 7 "University incomplete or complete"
+	label values educat7 lbleducat7
+*</_educat7_>
+
+```
+## Field of study information
+Users interested in the post-secondary field of study information can refer to the variable educat_orig, which retains the original education coding from the survey. This variable contains both the raw education information up to secondary level and the more detailed qualification information for post-secondary education. These can be distinguished by the number of digits: the raw education information is stored in two digits, except in 2012 when it is stored in three digits, while the post-secondary qualification information is stored in three digits, except in 2012 when it is stored in four digits.
+
+However, the meaning of the detailed qualification or field of study codes is not fully documented in the materials available to the GLD team, and these codes are not fully compatible with ISCED categories. Their detailed interpretation is therefore beyond the scope of the GLD harmonization, and users interested in this information would need to determine the meaning of the raw codes separately.
+