The 2015 FIES is a sample survey designed to provide income and expenditure data that are representative of the country and its 17 regions. It used four replicates of the 2003 Master Sample (MS) created for household surveys on the basis of the 2000 Census of Population and Housing. The 2003 MS has been designed to produce the sample size needed for large surveys, like the FIES. To facilitate subsampling, the 2003 MS has been designed to readily produce four replicate samples from the full set of sampled PSUs.
Starting 2012 FIES, the survey adopted the 2009 Philippine Classification of Individual Consumption According to Purpose (PCOICOP). The 2009 PCOICOP is the first standard classification of individual consumption expenditure in the country.
The 2015 FIES enumeration was conducted twice – the first visit was done in July 2015 with the first semester January to June as the reference period; the second visit was made in January 2016 with the second semester of 2015, that is, July to December 2015 as reference period. The same set of questions is asked for both visits.
The number of families for the 2015 FIES was estimated using the household population projections-based on the household population counts from the 2010 Census of Population and Housing (CPH).
The set of samples selected for the 2015 FIES is only one of the possible sets of samples of equal size that have been selected from the same population using the same sampling design. Estimates derived from each of these sets of samples would differ from one another. Sampling error is a measure of the variability of the estimates among all possible sets of samples. It is usually measures in terms of the standard errors for a particular statistic.
The standard error can be used to calculate confidence intervals within which the true value for the population can reasonably be assumed to fall. For example, for any given statistic calculated from a sample survey, the value of that statistic will fall within a range of plus or minus two times the standard error of that statistic in 95 percent of all possible samples of the same size and design.