[ METHODOLOGY



Relative poverty determined by % of students eligible for free/reduced-price lunch:

  • Low Poverty
    (0–25%) n=690 teachers and 197 principals,
  • Mid-low Poverty (26–50%) n=923 teachers and 257 principals,
  • Mid-high Poverty (51–75%) n=1062 teachers and 288 principals, and
  • High Poverty (76%+) n=1019 teachers and 285 principals.

A national survey of 4,721 public school educators was conducted by YouGov between July 22, 2016 and August 26, 2016, via an email-to-online survey method. Lists of teachers and principals were sourced from Market Data Retrieval’s (MDR) database of public school Pre-K–12 teachers and principals. A total of 3,694 teachers (including 76 school librarians) and 1,027 principals (including 146 vice principals) completed the survey. When data is presented among teachers, librarians are included unless otherwise specified. When the term “educators” is used to describe charts, tables and data findings, we are referencing teachers, librarians and principals combined.

The sponsor of the research was not revealed to the respondents. Participation was incented with a gift certificate to an online education store, which was revealed at the end of the survey to be the Scholastic Teacher Store. Principals were additionally given the option to access a webinar of the results upon project completion.

In order to pull lists that were reflective of the distribution of public school teachers (Pre-K–12) across its national education database, MDR created over 3,000 audience segments using unique combinations of states, school urbanicity, percentage of students receiving free/reduced-price lunch, years of experience and school grade level. A proportionate random sample was then pulled from each audience segment to ensure a representative sample across these criteria.

The resulting data was weighted using a two-step process, separately for teachers and principals. Each state was first weighted to the appropriate proportion within the appropriate census region, and then the teachers and principals within each region were weighted on specific characteristics, based on available National Center for Education Statistics (NCES) and MDR information. Teachers were weighted on gender, years of teaching experience, school urbanicity, school grade range, district enrollment and percentage of students receiving free/reduced-price lunch. Principals were weighted on gender, metro status, school grade range, district enrollment and percentage of students receiving free/reduced-price lunch.

Study Preparation

Prior to questionnaire design, online focus groups were conducted in the spring of 2016 to obtain input on potential questions. Four focus groups were held with 2016 State Teachers of the Year, and two focus groups were conducted with principals. In addition, an in-person working session was conducted with the 2016 State Teachers of the Year to gain further feedback on the relevance and language of drafted questions. The survey was pre-tested in early July of 2016, including live interviews via telephone and web conference with eight teachers and five principals.

The quotes that appear throughout this series were captured in one of several ways:

  • Educators who participated in the online survey had the opportunity to answer one of six open-ended questions regarding educational issues, and were also able to share any additional thoughts they had.
  • Verbatims were collected from educators who participated in the focus groups.

Data Presentation

In some cases, percentages may not sum to 100% due to rounding. Additionally, when two or more scale points are combined, for example, to show the percentage who say they agree or agree strongly, percentages may round up or down by one percentage point.

Data are presented throughout this series in charts and tables. Due to the robust nature of the sample, it is safe to assume that any difference of 7 points or more between teacher subgroups, any difference of 10 points or more between principal subgroups and any difference of 6 points or more when teachers and principals are combined is statistically significant at the 90% or 95% confidence level.

In this series you will see two main subgroupings analyzed – poverty level and grade level. Poverty levels are determined by NCES definitions of high- and low-poverty schools, based on the percentage of students eligible for free/reduced-price lunch. The four breaks shown in the repGrade levels for teachers are based on self-reported grades taught and defined as follows: Elementary (Pre-K–5) n=1928, Middle School (6–8) n=1011 and High School (9–12, AP) n=1056. Some teachers may teach grades that span multiple categories, in which case they were counted in the appropriate multiple categories. Principal grade-level subgroups are based on pre-coded classifications of Elementary n=625, Middle School n= 169 or High School n=189.