National Public Radio/Robert Wood Johnson Foundation/Harvard School of Public Health Poll: Sports and Health in America, United States, 2015 (ICPSR 38385)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Study Purpose View help for Study Purpose

The study was intended to estimate the share of American adults and children engaged in sports and exercise and the effect of sports on quality of life.

Study Design View help for Study Design

The study used phone numbers randomly generated from landline and cell phone samples, with an overlapping frame design, through Marketing Systems Group's (MSG) GENESYS sampling system. The sample consisted of four components: 1) Main RDD dual frame sample (N=1,400) included overlapping landline (n=670) and cell phone (n=730) frames of interviews with randomly selected adults, 2) Main RDD dual frame sport-participants oversample (N=575) included respondents from overlapping landline (n=205) and cell phone (n=370) frames who stated they participated in any sports during the previous year, 3) Sports-participant recruit oversample (N=198) included respondents who said they had participated in sports during a previous SSRS Omnibus survey and agreed to be called back, and 4) Middle/High-School parent oversample (N=333) included respondents who completed the SSRS omnibus survey in recent months and indicated they were parents (or legal guardians) of children living in their households, and that at least one child in the house was between the ages of 11 and 17.

The questionnaire was developed by HORP, NPR, and RWJ in consultation with the SSRS project team. The questionnaire was also translated into Spanish.

Sample View help for Sample

The sample includes 2,506 national adults, age 18 or older across the 50 states and the District of Columbia, including oversamples of adults who said they played or participated in any sport in the past year and parents of junior high school, middle school or high school children.

Universe View help for Universe

National adult, including oversamples of adults who said they played or participated in any sport in the past year, and parents of junior high school, middle school or high school children

Unit(s) of Observation View help for Unit(s) of Observation

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Response Rates View help for Response Rates

The combined response rate across all sample types was 17%.

Original Release Date View help for Original Release Date

2022-03-10

Weight View help for Weight

The study dataset contains weight factors that should be employed in any data analysis. Typically weights are used in an attempt to assure that the survey sample more accurately represents the population. The weight variables in this study are TOTLWGT, QGRP1WGT, and QGRP2WGT.

The survey data were weighted to adjust for the fact that not all survey respondents were selected with the same probabilities, account for the sample design and oversamples, and account for systematic nonresponse along known population parameters.

For the Main Landline and Cell Phone RDD sample, the following weighting was involved: 1) adjustment for the likelihood of selection (base-weight), 2) post-stratification weighting using Iterative Proportional Fitting (or "raking") to reflect the distribution of the adult U.S. population along known population parameters.

For all RDD components combined, the following weighting was involved: 1) the population incidence of sports participation was estimated from the weighted main-RDD sample to estimate the extent to which sports participants were oversampled.

For all samples combined, the following weighting was involved: 1) propensity scoring for the sports recruits and parent callbacks, 2) estimating the share of the parent population, 3) sport participants and non-participants were then raked separately, 4) weight truncation ("trimming") to reduce variance, 5) proportionately combining the sport-participant and non-participant weighted sample, and 6) unique weights were created for split-samples (asked only sports or only exercise-related questions).