Survey ID Number
Demographic and Health Survey 1987
Estimates of Sampling Error
Sampling errors, on the other hand, can be evaluated statistically. The sample of women selected in the 'IIDHS is only one of many samples of the same size that could have been drawn from the population using the same design. Each sample would have yielded slightly different results from the sample actually selected. The variability observed among all possible samples constitutes sampling error, which can be estimated from survey results (though not measured exact/y).
Sampling error is usually measured in terms of the "standard error" (SE) of a particular statistic (mean, percentage, etc.), which is the square root of the variance of the statistic across all possible samples of equal size and design. The standard error can be used to calculate confidence intervals within which one can be reasonably sure the true value of the variable for the whole population falls. For example, for any given statistic calculated from a sample survey, the value of that same statistic as measured in 95 percent of all possible samples of identical size and design will fall within a range of plus or minus two times the standard error of that statistic.
If simple random sampling had been used to select women for the TTDHS, it would have been possible to use straightforward formulas for calculating sampling errors. However, the TTDHS sample design used two stages and clusters of households, and it was necessary to use more complex formulas. Therefore, the computer package CLUSTERS, developed for the World Fertility Survey, was used to compute sampling errors.
In addition to the standard errors, CLUSTERS computes the design effect (DEFT) for each estimate, which is defined as the ratio between the standard error using the given sample design, and the standard error that would result if a simple random sample had been used. A DEFT value of 1 indicates that the sample design is as efficient as a simple random sample; a value greater than 1 indicates that the increase in the sampling error is due to the use of a more complex and less statistically efficient design.
Sampling errors are presented in Table B.1 of the Final Report for 35 variables considered to be of primary interest. Results are presented for the whole country, for urban and rural areas, and for three age groups. For each variable, the type of statistic (mean, proportion) and the base population (e.g., all women, women in union) are given in Table B.1. Table B.2 presents the value of the statistac, R; its standard error, SE; the actual number of cases, N; the DEFT value; and the relative standard error, SE/R for each variable. In addition to these indicators, the 95 percent confidence limits for the statistic, R-2SD and R+2SD, are presented.
In general, the sampling errors for the country as a whole are small, which means that the TTDHS results are reliable. For example, in the whole sample, the survey found that women average 2.059 children ever born; the standard error of this estimate is .037. Therefore, to obtain the 95 percent confidence limit, one adds and subtracts twice the standard error to the sample estimate, i.e., 2.05 + .074. There is a 95 percent chance that the true average number of children ever born to all women 15-49 in Trinidad and Tobago is between 1.985 and 2.134. This same calculation can be performed for all other variables listed.