With 40 years of experience in developed and emerging markets, Euromonitor International’s research methodology offers a unique combination of specialist industry knowledge and in-country research expertise.
Euromonitor International’s demographic data have been compiled following an extensive research programme. National Statistical offices around the world were approached by a team of Euromonitor International’s researchers to obtain the most detailed contemporary data available on population age structure. The major sources of demographic data are population censuses, micro-censuses, surveys and registers. The research material is assembled into a uniform database by applying strict category definitions to standardise the primary data. Data arrangement, availability and definitions varied considerably country by country. Thus painstaking regrouping and adding up of population cohorts had to be carried out to ensure cross-country comparability as well as correct regional aggregations. For some countries, the quality and availability of the primary data was not sufficient to fit it into this tightly structured database. In such cases, United Nations Population Division single year age group estimates were used.
Single year age group data was not available for some countries and years. In such cases, Sprague multiplier techniques were employed to break down five-year age groups to single year age groups. Where data was missing between census or micro census results, Hermit interpolation was used to fill in these gaps. On a number of occasions, structural breaks were discovered in the time series provided by National Statistical Offices. In such cases, Euromonitor International’s demographers used advanced data manipulation techniques to revise the out-of-date series and bring them in line with the latest census information.
We sought to include population forecasts of National Statistical Offices whenever such data was provided. In most cases, however, we had to resort to our own forecasts, as national data was either not available or the level of detail was not sufficient. Our forecasts were produced using the cohort component method. In the cohort-component method, the components of population change (fertility, mortality and net migration) are projected separately for each cohort (persons born in a given year). The base population is advanced each year by using projected survival rates and net migration by single year of age and sex. Each year, a new birth cohort is added to the population by applying the projected fertility rates. The age cohorts determine time interval. The time interval for which an estimate can be made is the length of time it takes all the members of an age cohort (say, 18) to pass on to the next age grouping (the 19 year-old group). Projecting population by the cohort-component method involved a number of steps, each of them utilising the demographer’s expert opinion. First, a component projection requires a population properly distributed by sex and age to serve as the base population from the starting date of the projection. Second, assumptions were made regarding the level and patterns of fertility. The level of fertility (total fertility rate) was projected and then a pattern of fertility (age-specific fertility rates) was assumed. A similar process was used to project mortality, where general level of mortality by sex was projected, then age-specific mortality rates assumed. Finally, levels and patterns of net migration were estimated. The projected levels and patterns of the components of population change were applied to the base population to yield the projected population for a given year.