involvement_survey_str2s: Parent involvement survey: stratified, two-stage sample

In nrba: Methods for Conducting Nonresponse Bias Analysis (NRBA)

Parent involvement survey: stratified, two-stage sample

Description

An example dataset describing a stratified, multistage sample of 1,000 parents of students with disabilities, from a population of 20,000. The parent involvement survey measures a single key outcome: whether "parents perceive that schools facilitate parent involvement as a means of improving services and results for children with disabilities."

The sample was selected by sampling 5 schools from each of 20 districts, and then sampling parents of 10 children in each sampled school. The variable BASE_WEIGHT provides the base sampling weight. The variable SCHOOL_DISTRICT was used for stratification, and the variables SCHOOL_ID and UNIQUE_ID uniquely identify the first and second stage sampling units (schools and parents). The variables N_SCHOOLS_IN_DISTRICT and N_STUDENTS_IN_SCHOOL can be used to provide finite population corrections.

Usage

involvement_survey_str2s

Format

A data frame with 5,000 rows and 18 variables

Fields

UNIQUE_ID

A unique identifier for students

RESPONSE_STATUS

Survey response/eligibility status: 'Respondent', 'Nonrespondent', 'Ineligble', 'Unknown'

WHETHER_PARENT_AGREES

Parent agreement ('AGREE' or 'DISAGREE') for whether they perceive that schools facilitate parent involvement

SCHOOL_DISTRICT

A unique identifier for school districts

SCHOOL_ID

A unique identifier for schools, nested within districts

STUDENT_GRADE

Student's grade level: 'PK', 'K', 1-12

STUDENT_AGE

Student's age, measured in years

STUDENT_DISABILITY_CODE

Code for student's disability category (e.g. 'VI' for 'Visual Impairments')

STUDENT_DISABILITY_CATEGORY

Student's disability category (e.g. 'Visual Impairments')

STUDENT_SEX

'Female' or 'Male'

STUDENT_RACE

Seven-level code with descriptive label (e.g. 'AS7 (Asian)')

PARENT_HAS_EMAIL

Whether parent has an e-mail address ('Has Email' vs 'No Email')

PARENT_HAS_EMAIL_BENCHMARK

Population benchmark for category of PARENT_HAS_EMAIL

STUDENT_RACE_BENCHMARK

Population benchmark for category of STUDENT_RACE

N_SCHOOLS_IN_DISTRICT

Total number of schools in each district

N_STUDENTS_IN_SCHOOL

Total number of students in each school

BASE_WEIGHT

Sampling weight to use for weighted estimates

CONTACT_ATTEMPTS

The number of contact attempts made for each case (ranges between 1 and 6)

Examples

# Load the data
involvement_survey_str2s

# Prepare the data for analysis with the 'survey' package

  library(survey)

  involvement_survey <- svydesign(
    data = involvement_survey_str2s,
    weights = ~ BASE_WEIGHT,
    strata =  ~ SCHOOL_DISTRICT,
    ids =     ~ SCHOOL_ID             + UNIQUE_ID,
    fpc =     ~ N_SCHOOLS_IN_DISTRICT + N_STUDENTS_IN_SCHOOL
  )

nrba documentation built on Nov. 21, 2023, 9:07 a.m.