Hi hi hi hi w w w w however for some child sampling

This preview shows page 81 - 83 out of 128 pages.

We have textbook solutions for you!
The document you are viewing contains questions related to this textbook.
Introduction to Probability and Statistics
The document you are viewing contains questions related to this textbook.
Chapter 3 / Exercise 24
Introduction to Probability and Statistics
Beaver/Mendenhall
Expert Verified
hi hi hi hi w w w w     However, for some child sampling strata, the three-median rule caused too many weights to be truncated. The percentage of truncated weights was kept to less than 3 percent, so, for some child sampling strata, a three-and-a-half-median or four-median rule was used. For the children who had their full sample weight truncated, all the replicate weights were reduced by the same percentage. Post-stratification of Enrolled Child Weight The nonresponse adjusted children’s weight was further adjusted by a post-stratification procedure. The control totals for post-stratification contained the number of special education children enrolled by December 2003, by age, for each of the 50 states and the District of Columbia. Post-stratification was necessary because several states did not have any children sampled, either because, by chance, no LEAs in those states were selected, or none of the selected LEAs in a state responded. It should be noted that the control totals are snapshot figures, while the PEELS population includes children enrolled during a certain time period. The control totals also include children from the very small (size 5) school districts, which were not covered (but were adjusted for) by the PEELS sample. The post-strata were formed by crossing the three age groups and nine subregions formed by combining states within the same region by their geographical proximity. The size of states in terms of number of children was also taken into consideration in order to obtain similar-sized post-strata. After the post-stratification was applied, the final enrolled children’s base weight was created. This weight is called the children’s base weight, although it resulted from various adjustments, because it will be the base for further nonresponse adjustments for different data collection instruments. These are discussed in the following section. Parent Interview Weights The parent interview was attempted for all enrolled children, but some parents did not respond. The weights for the parent interview data were created by adjusting the enrolled children’s base weights for parent nonresponse. The nonresponse adjustment cells were the same as the ones formed for the nonresponse adjustment to obtain the enrolled children’s base weight. This worked well because the response rate for the parent interview was very high. In Waves 1 and 2 at the completion of imputation, parent interview data and corresponding weights were available for 96 percent and 93 percent, respectively, of the children in the augmented sample. In Wave 3, parent interview data and corresponding weights were available for 88 percent of the children in the augmented sample. Parent interview data and corresponding weights were available for 83 percent of the children in the augmented sample in all three waves.
We have textbook solutions for you!
The document you are viewing contains questions related to this textbook.
Introduction to Probability and Statistics
The document you are viewing contains questions related to this textbook.
Chapter 3 / Exercise 24
Introduction to Probability and Statistics
Beaver/Mendenhall
Expert Verified
B-8 Child Assessment Weights The child assessment was done in two ways. Most of the children were assessed directly, but for children who could not complete the direct assessment, an alternate assessment was conducted. Together,

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture