21.06.2014 Views

Questionnaire Dwelling Unit-Level and Person Pair-Level Sampling ...

Questionnaire Dwelling Unit-Level and Person Pair-Level Sampling ...

Questionnaire Dwelling Unit-Level and Person Pair-Level Sampling ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

ecause, even though the parent-child pair relationship had been established, it often was unclear<br />

whether there was a second "parent" in the household. Other domains had very little uncertainty.<br />

The counts of the number of children in the parent-focus parent-child domain were almost<br />

always definitively determined.<br />

Table 6.10 Amount of Imputation Required for Multiplicities in Various <strong>Pair</strong> Domains:<br />

2006<br />

<strong>Pair</strong> Domain Multiplicity Missing Cases<br />

Parent-Child (12-14), Child Focus Number of parents 74<br />

Parent-Child (12-14), Parent Focus Number of children 0<br />

Parent-Child (15-17), Child Focus Number of parents 70<br />

Parent-Child (15-17), Parent Focus Number of children 0<br />

Parent-Child (12-17), Child Focus Number of parents 144<br />

Parent-Child (12-17), Parent Focus Number of children 0<br />

Parent-Child (12-20), Child Focus Number of parents 166<br />

Parent-Child (12-20), Parent Focus Number of children 4<br />

Sibling (12-14)-Sibling (15-17), Older Sibling<br />

Focus<br />

Sibling (12-14)-Sibling (15-17), Younger Sibling<br />

Focus<br />

Sibling (12-17)-Sibling (18-25), Older Sibling<br />

Focus<br />

Sibling (12-17)-Sibling (18-25), Younger Sibling<br />

Focus<br />

Number of younger<br />

siblings<br />

Number of older siblings 8<br />

Number of younger<br />

siblings<br />

6.3.3 Creation of Imputation-Revised Multiplicity Variables<br />

Number of older siblings 18<br />

In many cases where the pair relationships were not defined, multiplicity counts also<br />

were not defined. In addition, there were a h<strong>and</strong>ful of cases where multiplicity counts were not<br />

determined even when the pair relationship was known. In all of these cases, imputation was<br />

required to determine the multiplicity count. As with the pair relationship imputation, missing<br />

multiplicities were imputed using the PMN method described in Appendix N. In this section, the<br />

application of PMN to the imputation of multiplicities is described. Since only the multiplicity in<br />

the second stage was imputed for each pair, the imputation was univariate in the sense that no<br />

sequential models were necessary. However, in some cases, several variables were associated<br />

with a single model, as described below.<br />

6.3.3.1 Setup for Model Building<br />

Multiplicity counts were defined only within the relevant domain, which, in turn,<br />

depended upon the pair relationship. For the sibling-sibling pairs, four separate imputations were<br />

4<br />

4<br />

47

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!