Abstract
Throughout the 2020–2021 school year, families’ access to—and desire to participate in—in-person educational experiences was highly unequal. Concerns about “school hesitancy” in light of COVID-19 have continued into the 2021–2022 school year, driven both by concerns about well-being and concerns about safety. Using a nationally representative sample of families, we tested a messaging intervention aimed at reducing school hesitancy. We found that targeted messaging to address well-being and safety concerns substantially improved parent reports of their likelihood of sending their child back for in-person learning for parents who were previously unsure. The findings suggest the importance of careful COVID-related communication from schools.
As of fall 2021, the Centers for Disease Control and Prevention’s (2021) Guidance for COVID-19 Prevention in K–12 Schools places a high priority on schools being fully open—with adequate safety protocols—for the 2021–2022 school year. The priority on reopening is attributable, at least in part, to evidence that in-person learning is superior to online or hybrid learning. For instance, parents expressed greater satisfaction and less concern with in-person than online learning (Rapaport et al, 2020) and evidence from before and during the pandemic suggests students’ learning was greater in in-person settings (e.g., Chetty et al., 2020; Domingue et al., 2021; Kaufman & Diliberti, 2021).
Yet opening schools will not actually result in all students retuning to in-person learning. A nontrivial proportion of families—18% according to our own analysis of nationally representative Understanding America Study (UAS) data collected in August, 2021—are “school hesitant,” either planning to keep their child online in the fall or are unsure about online versus in-person learning. Furthermore, school hesitant parents are disproportionately from already marginalized groups (Black and Hispanic families were 23% and 20% hesitant in August, 2021, respectively).
Parents have a variety of reasons for preferring to keep their children home. Researchers and journalists have posed several narratives about why parents keep children home including fears about COVID-19, lack of trust in schools that had underserved their children prepandemic, the inadequacy of school physical infrastructure, and school closures (Camp & Zamarro, 2021; Dee et al., 2021; “More Non-White Than . . .,” 2021; Kogan, 2021). There were also large gaps in preferences and enrollment decisions based on demographics such as race/ethnicity and children’s age (Dee et al., 2021; Musaddiq et al., 2021) Recent UAS data suggest that the primary reasons parents reported for keeping their children at home included those related to the importance of child well-being (which we previously labeled “fit,” e.g., child was happier and/or more academically successful at home) and safety (Saavedra et al., 2021). In that study, we found that among parents who did not send their child/ren to school in April/May 2021, 27% selected reasons related to child well-being, 28% to safety, and 33% to both well-being and safety (the other 12% selected a variety of other reasons).
Since policy makers and practitioners want to encourage school-hesitant parents to send their children back to in-person school in the 2021–2022 school year, we wondered whether well-targeted messages might address parents’ concerns and increase the likelihood that parents would send their children back. We created a brief hypothetical messaging intervention and tested it using a nationally representative sample of K–12 parents. Our results indicate that messaging targeted at the primary concerns of school-hesitant parents—those related to child well-being and safety—could substantially increase parents’ likelihood to report they will send their children back in person in the fall. We found the messaging intervention worked well for parents who reported they were “unsure” at baseline, but did not work for parents who reported at baseline they were not planning to send their children back in person. If the COVID-19 pandemic continues, or in future public health crises, similar messages may be an important tool in the arsenal for schools and public health officials to boost school attendance.
Method
Since 2014, the nationally representative, longitudinal UAS panel study has been surveying U.S. households about a range of issues including economics, health, and politics. At the onset of the pandemic, we began collecting data about families’ K–12 education experiences related to COVID-19. Since spring 2020, we have collected over 20 waves of survey data on a range of education topics. For this analysis, we use data collected between June 9 and July 21, 2021 (UAS348). Details on the sample, including comparisons to national demographics and our data cleaning procedures, are available in the online UAS348 topline and crosstab documents available on the UAS website. The response rate for this wave of the survey was 63%.
For our messaging intervention, we used simple random assignment to randomize parents into four groups and delivered respondents one of four hypothetical messages in the context of the regular online survey that they take on computers or phones/tablets (the full messaging texts are available in the online Appendix available on the journal website; they were administered in English and Spanish):
1) A basic control message, simply welcoming parents and their children back to school and indicating the school would comply with Centers for Disease Control and Prevention guidelines.
2) A specific “safety” message, that adds to (1) plans to describe specific safety protocols, vaccination rates, and COVID rates in the community.
3) A specific “well-being” message (in the past we have also called it “fit,” see Saavedra et al., 2021), that adds to (1) an acknowledgment of the academic and nonacademic effects of the pandemic on children and encouragement to address those effects by sending their children back in person.
4) A specific “combined” message, that adds to (1) both the safety and well-being messages.
Immediately after the message, we asked participants “On a scale from 1 to 10, how likely would you be to send [NAME] to school full-time, in-person after reading this message?”
To analyze our experimental data, we first assigned respondents to one of three “baseline” groups based on their response to a question on the survey that was administered before exposure to the experimental manipulation: “Are you planning to send [NAME] to school in-person at the beginning of the 2021–2022 school year?” (with response options “yes,” “no,” and “unsure”). We then regressed a binary indicator of whether the respondent planned to send their child back to school in-person or not (as measured postmessage) 1 on the two treatment indicators, 2 parents’ previous plans, and their interaction, with the control message as the reference category. 3 We also fit models that included additional covariates (e.g., parent race/ethnicity, parent educational attainment, household income, and urbanicity), different ways of binning the 1–10 scale, and two model types. We report results from the most parsimonious model here, documenting additional analyses and their results in the online Appendix (available on the journal website). The primary results from the most parsimonious model were substantively unchanged in each of our sensitivity analyses.
Results
The results, shown in Figure 1 (with corresponding Table A1 in the online Appendix available on the journal website), indicate the treatments had no effect on parents who said “yes” at baseline, negative but nonsignificant effects on parents who said “no” at baseline, and positive and statistically significant effects on parents who said “unsure” at baseline. The magnitude of the effects on “unsure” parents was substantial—reading a specialized message rather than the control message increased the probability that parents would report being likely to send their child in person in the fall by about 26 percentage points, from 50% to 76% for unsure parents. Analyzing the outcome in different ways—for instance, treating it as a continuous variable, dichotomizing it to >5 versus 5 or less—produced the same pattern of results, though the effect was insignificant at the .05 level if the outcome was treated continuously.
Predicted probability of parents reporting they would send their child to school in-person at the beginning of the 2021–2022 school year after exposure to messaging, by previous plans.
Discussion
Our results demonstrate that a brief messaging intervention targeted at hesitant (i.e., “unsure”) parents’ most serious concerns can affect their reported likelihood to send their children to in-person school. Though based on a hypothetical situation and not actual enrollment decisions, these findings imply that districts seeking to boost in-person enrollment should identify parents’ sources of concern and provide clear messaging to assuage those concerns. Done via email or text message with existing school district communication systems, this would be a relatively easy, low-cost intervention that could meaningfully boost student enrollment—only the cost of the time to discern the issues and write the text. However, recent research suggests substantial numbers of families are not well reachable through district communication systems (Musaddiq et al., 2020), so these approaches would not reach everyone. Policy makers and practitioners seeking to message in these ways should align their messages with best practices from the messaging literature (see, e.g., Musaddiq et al., 2020; Oreopoulos, 2020). Whether the effects we observe here would persist over time or would translate to actual behavior is also important to understand in a follow-up study. Future work should continue to probe ways to support communication with parents to drive better student opportunities during and after the COVID-19 pandemic.
Supplemental Material
sj-pdf-1-edr-10.3102_0013189X211070813 – Supplemental material for The Impact of a Messaging Intervention on Parents’ School Hesitancy During COVID-19
Supplemental material, sj-pdf-1-edr-10.3102_0013189X211070813 for The Impact of a Messaging Intervention on Parents’ School Hesitancy During COVID-19 by Morgan S. Polikoff, Daniel Silver, Marshall Garland, Anna Rosefsky Saavedra, Amie Rapaport and Michael Fienberg in Educational Researcher
Footnotes
Notes
Authors
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
