The 2018 Stress in America™ survey was conducted online within the United States by The Harris Poll on behalf of the American Psychological Association (APA) between July 27 and Aug. 28, 2018, among 3,458 adults age 18+ who reside in the U.S. In addition to the main sample, interviews among teens ages 15 to 17 (n=300) were collected as well as oversamples of black and Hispanic Gen Z (ages 15 to 21) (n=792). Interviews were conducted in English (n=4,278) and Spanish (n=272). Data were weighted to reflect their proportions in the population based on the 2017 Current Population Survey (CPS) by the Census. Weighting variables for the adults ages 18+ included education, age by gender, race/ethnicity, region, household income and time spent online. Hispanic adults were also weighted for acculturation taking into account respondents’ household language as well as their ability to read and speak in English and Spanish. Country of origin (U.S./non-U.S.) was also included for Hispanic and Asian subgroups. Propensity score weighting was used for the adults to adjust for respondents’ propensity to be online. A propensity score allows us to adjust for attitudinal and behavioral differences between those who are online versus those who are not, those who join online panels versus those who do not, and those who responded to this survey versus those who did not. Weighting variables for the teens ages 15 to 17 included parent’s education, age by gender, race/ethnicity, region, and size of household. Weighting variables for the Gen Z groups (ages 15 to 21) included education, age by gender, race/ethnicity, region, household income, size of household, and employment status. Country of origin (US/non-US) was also included for Gen Z Hispanic subgroups.
All sample surveys and polls, whether or not they use probability sampling, are subject to multiple sources of error, which are most often not possible to fully estimate and control, including sampling error, coverage error, error associated with nonresponse, error associated with question wording and response options, and post-survey weighting and adjustments. Therefore, The Harris Poll avoids the words “margin of error” as they are misleading. All that can be calculated are different possible sampling errors with different probabilities for unweighted, random samples with 100 percent response rates. These are only theoretical because no published polls – whether conducted by telephone or online – come close to the ideal probability sample given the sources of error previously noted. Because the sample is based on those who were invited and agreed to participate in an online research panel, no estimates of theoretical sampling error can be calculated.
Sample Size and Key Sample Definition
Sample Size
Throughout this report different segments of adults are discussed. Demographic sub-groups of adults include and are defined below:
2018 Survey
- Gender: Male (n=1,362); Female (n=2080); Transgender (n=15)
- Race/Ethnicity: White (n=1089); Hispanic (n=799); black (n=800); Asian (n=500); Native American or Alaskan native (n=201)
- Gen Z: White (n=187); Hispanic (n=505); black (n=537)
- Generation: Gen Z (15 to 21) n=531 (includes n=300 Gen Z “teens” and n=231 Gen Z “adults” 18-21); Millennials (22-39 years old) n=1,055; Xers (40-53 years old) n=735; Baby Boomers (54-72 years old) n=1,213; Matures or Older Adults (age 73 or older) n=224
- Region: East (n=620); Midwest (n=611); South (n=1,384); West (n=843)
- Parents: Parents (n=986); no children (n=2,472)
- Political Party Affiliation: Democrat (n=1,498); Republican (n=686); Independent (n=641)
Race/Ethnicity Definition
Harris Poll’s methods for asking about and categorizing race and ethnicity are in accordance with the approach taken by the U.S. Census Bureau1 which treats race and ethnicity as two separate and distinct categories, examined in two separate questions. Per the Census’ framework, ethnicity is categorized as either ‘Hispanic or Latino’ or ‘not Hispanic or Latino.’ Race is comprised of five categories: White, black/African-American, Asian, Native American/Alaskan Native, Hawaiian Native or other Pacific Islander. The Census further describes that the concept of race reflects self-identification by people according to the race or races with which they most closely identify and that people who report themselves as Hispanic can be of any race.
Using the framework described above and the questions below, if a respondent identified as Hispanic at Q1, for the purposes of this research they were included in the Hispanic sub-group regardless of their response to the subsequent question which examined race. All other groups were identified based solely on their answer to Q2. Very few indicated they were of mixed race or some other race. Those who did are included in the total sample but have been excluded from the racial/ethnic analysis as there were too few to analyze separately (~1 percent of the total sample).
BASE: U.S. RESPONDENTS
Q1: Are you of Hispanic, Latino, or Spanish origin?
- 1 - Yes
- 2 - No
- 9 - Prefer not to answer
BASE: U.S. RESPONDENTS
Q2: What is you race?
- White
- Black or African American
- Native American or Alaskan native
- South Asian
- Chinese
- Korean
- Japanese
- Filipino
- Arab/West Asian
- Pacific Islander
- Other Asian
- Mixed Race
- Some other race
- Prefer not to answer
1 https://www.census.gov/population/hispanic/about/faq.html