Thursday, February 8, 2018

Missing Values in NHS Individuals PUMF


Question:
For the WAGES variable in the NHS individuals PUMF, why are "not available" and "not applicable" not declared as missing values? The codebook states that:

"The value 8,888,888 stands for not available. The value 9,999,999 stands for not applicable and is applied to all persons aged less than 15 years"

Leaving those values (which constitute over 17% of the cases) subject to calculation would seem to skew the results rather significantly, would they not?


Answer:
B Estimation
Note: Users must refrain from publishing unweighted estimates and from conducting analyses based on unweighted data from the file because the unweighted results do not represent the population but
only describe the sample. They must also make sure to exclude values of study variables that are not applicable or not available from their calculations because those values might be considered as valid observed values by the statistical software when they are not. For example, values such as 9,999,999 or 8,888,888 for a numeric (or quantitative) variable would be interpreted as valid observed values but should be considered as nominal values indicating these values are not usable in estimation.

Example 4:

We want to estimate the average total income of women aged 15 years and over living in Ontario who
have an income. In the calculation of the numerator, WEIGHT is multiplied by the value of the 'total income' variable for individuals with an income (where TOTINC ^= 8,888,888, TOTINC ^= 9,999,999, TOTINC ^= 0) whose gender is female (SEX = 1) and who are aged 15 or over (AGEGRP ≥ 6, AGEGRP ^= 88) in the province of Ontario (PR = 35); the results are then totalled. To estimate the average, the numerator (or estimated total income) is divided by the sum of WEIGHT for individuals satisfying the same conditions on TOTINC, SEX, AGEGRP and PR.
The result obtained is: $179,154,359,345 / 5,072,260 = $35,320, which means the average total income of women aged 15 and over living in Ontario who have an income is around $35,320.
B.2.b.3 Estimator of a ratio
A ratio can be defined as the division of two amounts, which could be two totals or two averages

Example these counts are included
8888 Not available (unweighted) 13,676 - (Weighted) 483,013
9999 Not applicable Canadian citizens by birth and non-permanent residents (unweighted) 697,600 – (Weighted) 26,060,226