Appendix A
Methodology
Survey Population and Response Rate
Varying Degrees was administered using NORC’s AmeriSpeak Panel for the sample source. Funded and operated by NORC, AmeriSpeak is a probability-based panel designed to be representative of the U.S. household population.1 A general population sample of adults 18 years of age and older was selected from NORC’s AmeriSpeak Panel for this study. Additional African Americans,2 Hispanics/Latinos,3 and Asian Americans were also selected from NORC’s AmeriSpeak Panel for three specific race-ethnicity oversamples. College students—those currently enrolled in an associate or bachelor's degree program—were selected from the AmeriSpeak Panel to create an oversample of current college students.
The screening and main interview stages of data collection were conducted during a single survey session. There was a screening stage to identify qualified and eligible panelists for the college student oversample; these are the respondents qualified to participate in the second stage, which is the main study interview.
Respondents who answered the screener, regardless of eligibility, were considered a screener complete. Respondents who were determined to be eligible for the study, based on the screener, then completed the survey were considered a survey completer. The screener completion rate was 29.51 percent. The incidence rate was 11.49 percent. Among those cases that qualified for the main study interview, the interview completion rate was 93.24 percent. The summary statistics on sample performance are shown in Table 1 below.4
Table 1. Sample Performance Summary—College Student Oversample
| Sampled / Invited Panelists | # Screening Interviews Completed | Screener Completion Rate | # Panelists Eligible for Interview | Incidence / Eligibility Rate | # Survey Interviews Completed | Interview Completion Rate |
|---|---|---|---|---|---|---|
| 14,393 | 4,247 | 29.51% | 488 | 11.49% | 455 | 93.24% |
Data Collection Procedures
To administer the phone survey, NORC dialed the sampled phone-mode panelists throughout the field period. In addition, starting on May 16, AmeriSpeak web-mode panelists for whom AmeriSpeak had a phone number also were called to encourage response. These web panelists were allowed to complete the survey via phone if convenient. Panelists were offered the cash equivalent of $3. In the middle of the field period, incentives for African Americans and college student oversamples were increased to $6. At the end of the field period, a final increase to $10 for the college students was implemented.
Margin of Error and Design Effect
The margin of error and design effect for the general population and each oversampled group are as follows:
| Samples | Sample Size | Margin of Error | Design Effect |
|---|---|---|---|
| General Population | 2,029 | 3.35% | 2.38 |
| African American | 414 | 7.30% | 2.30 |
| Asian American | 233 | 9.38% | 2.14 |
| Latino/Hispanic | 410 | 8.01% | 2.74 |
| College Student | 513 | 6.00% | 1.92 |
Citations
- For more detailed information on the AmeriSpeak panel recruitment and management methodology, see Appendix B.
- We refer to African Americans as Black Americans in this report.
- We refer to Hispanic/Latino Americans as Latinx Americans in this report.
- Please download NORC Card for the AAPOR response rate located here: website TK.