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.

almost all cases since only one spouse-spouse pair could have been selected that included that<br />

pair member. If the true multiplicity count exceeded 1, then the multiplicity count was set to 1. 20<br />

Note that other spouse-spouse pairs in the household (one spouse's parents, for example) would<br />

have been of interest in the household counts discussed in subsequent sections.<br />

Table 6.9<br />

<strong>Pair</strong><br />

Relationship<br />

Multiplicity Counts for Each <strong>Pair</strong> Member<br />

Focus<br />

Member Direct Count Indirect Count<br />

Parent-Child Child From child: number of parents From parent: self + spouse/partner<br />

Parent-Child Parent From parent: number of children<br />

in appropriate age range<br />

Sibling-Sibling<br />

Sibling-Sibling<br />

Older<br />

sibling<br />

Younger<br />

sibling<br />

From older sibling: number of<br />

siblings in younger age range<br />

From younger sibling: number of<br />

siblings in older age range<br />

From child: self + number of siblings<br />

in the appropriate age range<br />

From younger sibling: self + number<br />

of siblings in younger age range<br />

From older sibling: self + number of<br />

siblings in older age range<br />

6.3.2 Determining the Final Multiplicity Count<br />

Once the counts were determined for each pair member, it was necessary to resolve<br />

differences between these counts across pair members. In most cases, the direct <strong>and</strong> indirect<br />

counts agreed, with no bad relationship codes for either pair member, resulting in an easy<br />

determination of the final multiplicity count. An easy determination was usually possible if one<br />

pair member had bad relationship codes or had a count of 0, which meant that the final<br />

multiplicity count came from the pair member with good data. 21 For some cases, both pair<br />

members had bad relationship codes, which meant that the final multiplicity was left to<br />

imputation. Some of the remainder of cases could be reconciled <strong>and</strong> some could not. In the cases<br />

where reconciliation was possible, many of the disagreements between the pair members were<br />

resolved by going to the screener. The method used to reconcile differing counts depended upon<br />

the domain. In addition to the screener, for the parent-child domains, the FIPE3 variable was<br />

used to help reconcile differences. Detailed rules for reconciling differences between pair<br />

members are provided in Appendix R.<br />

If reconciliation between the counts from the two pair members in the household <strong>and</strong> the<br />

screener was not possible, upper <strong>and</strong> lower bounds within which the imputed value had to reside<br />

were determined from the counts for each pair member <strong>and</strong> the counts for the screener. The<br />

amount of imputation required for the multiplicity counts is shown in Table 6.10 for the 2006<br />

survey year. From this table, it is apparent that the greatest degree of uncertainty came with the<br />

determination of the number of parents in the child-focus parent-child domains. This occurred<br />

20 In rare cases, it was possible for a respondent to have two or more spouses. Determining the appropriate<br />

multiplicity count in these cases required knowledge of which spouse was the focus, which would be arbitrary.<br />

Because having multiple spouses was an extremely rare occurrence, <strong>and</strong> because of the complexity of determining<br />

the appropriate multiplicity count, these situations were not accounted for.<br />

21 There were some provisions to this rule. If the bad relationship codes were only within the relevant age<br />

ranges, then the count from the good side was used only if the age ranges in the good side matched the screener.<br />

46

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

Saved successfully!

Ooh no, something went wrong!