�鶹������ý

The 2026 National Parent Survey is a nationally representative survey of parents and caregivers of children ages five and younger, conducted by the New Practice Lab in partnership with . The survey was fielded between January 16 and February 2, 2026, and reflects the views of 5,472 parents across all 50 states and the District of Columbia.

The included 37 questions, the majority of which were closed-ended (e.g., multiple-choice, rating scales, select-all-that-apply). Five open-ended questions invited parents to respond in their own words. Responses were not required for all questions; respondents could skip or decline to answer individual items. The survey was designed to cover topics including respondent demographics, current and ideal work arrangements, current and ideal child care arrangements, leave taken and ideal leave duration, barriers to ideal time use, challenges, and hopes for the future. The survey was administered in both English and Spanish, primarily online, with a small number of interviews completed by telephone.

The sample draws from probability-based and nonprobability panels. The first is , a probability-based panel designed to be representative of the U.S. population, from which 1,104 completed interviews were obtained. The second is a nonprobability panel, from which an additional 4,368 interviews were collected using demographic quotas to reflect the national population of parents. Participants received a small monetary incentive for survey completion. To improve the precision of estimates for lower-income families, parents in households with incomes below 200 percent of the federal poverty level were sampled at a higher rate. To ensure the sample is nationally representative, data were weighted using benchmarks from the 2025 Current Population Survey, blending the probability and non-probability samples and adjusting for differences in age, sex, race/ethnicity, education, income, and census region. The overall margin of error for the combined sample is ±2.0 percentage points at the 95% confidence level. Margins of error for subgroup analyses—such as those by gender, income, race and ethnicity, or geography—are larger and vary depending on subgroup sample size. Subgroup findings should be interpreted accordingly.

Additional questions or requests for the full methodology report can be submitted to [email protected].

The survey relies on self-reported data, meaning responses reflect parents’ own perceptions and experiences, not independently verified information. As readers review this report, conduct their own analysis of the public use data file, and seek to interpret results, we hope to emphasize that this effort was not designed to explore the “why” behind individual responses, or how people formulated their ideal scenario. We caution readers to consider how the following biases could impact responses:

• Social Desirability Bias: Parents may overindex toward perceived “ideal” behaviors (e.g., perceptions and attitudes about familial roles, acceptability of leave in their current work environment) and underreport stigmatized ones (e.g., acceptability of certain care scenarios) to appear more favorable.
• Non-Response Bias: Parents who choose not to participate may differ systematically from those who do (e.g., busier, more stressed, or less engaged parents may opt out), skewing results toward more available or motivated respondents. Although the data are weighted to match national population benchmarks, survey-based estimates are subject to both sampling and nonsampling error.
• Recall Bias: Parents asked to report past behaviors or events (e.g., “Thinking back to when your youngest child was born, how much time did you take off from work?”) may misremember frequency, duration, or timing, especially for emotionally charged experiences.
• Language and Literacy Bias: A survey offered only in English and Spanish, and at this particular reading level, will underrepresent non-English-speaking families and those with lower literacy.
• Volunteer Bias: Parents who self-select into surveys tend to be more educated, opinionated, or invested in the survey’s topic, making them unrepresentative of the broader parent population.
• Mode Bias/Online/Digital Access Bias: Responses could be skewed by those who can complete an online survey as opposed to a small sample on the telephone; sensitive questions in particular yield different answers across modes.
• Demographic Underrepresentation: Certain groups such as LGBTQ+ parents, young parents, incarcerated parents, non-custodial parents, or parents of children with disabilities may be undersampled, limiting generalizability.
• Temporal Bias: The survey captures a single snapshot in time (mid-January to early February 2026). Findings may be influenced by recent events (e.g., state politics, news about day care centers in the area, or policy changes) that don’t represent typical conditions. This is not a longitudinal study; it describes what parents are experiencing and what their current preferences are, but cannot establish whether conditions are improving or worsening.
• Definition/Construct Bias: Question wording and terms used like “child care center,” “family day care or home-based child care program,” or even “family” may be interpreted differently by respondents, introducing inconsistency. This is especially true across diverse family structures. The experiences of parents who identify their gender outside male or female are also not separately represented in the findings. 

We hold our team to the highest possible standard for rigor and analysis, but we acknowledge that, as humans, we are prone to error. Accompanying this report is a folder of . To reduce the risk of reidentification, the public-use data files do not include the full text of the open-ended question responses. Any mistakes in this report are the authors’ alone and can be submitted to [email protected].

Editorial disclosure: The findings represented in this report represent the authors’ analysis and do not necessarily reflect the positions of �鶹������ý, its staff, fellows, funders, or board of directors.