Parental Experiences Survey (PES)

Detailed information for 2024

The Parental Experiences Survey collects information from parents on their experiences, knowledge and behaviours regarding pregnancy, childbirth, and access to and use of health care services. The survey aims to paint a more comprehensive picture of new parents and their infants. The results from the survey will help inform national recommendations for maternal and newborn care as well as efforts to improve the mental health and well-being of parents and families across Canada.

Data release - February 5, 2026

• Questionnaire(s) and reporting guide(s)
• Description
• Data sources and methodology
• Data accuracy

Description

The Parental Experiences Survey aims to paint a more comprehensive picture of new parents and their infants. The survey asks questions about pregnancy intention and history, health before, during and after pregnancy, experiences during labour and birth and with infant feeding, mental health, substance use, and experiences of violence or abuse.

The results from the survey will be used by healthcare professionals and researchers to inform national recommendations for maternal and newborn care, as well as to help improve the mental health and well-being of parents and families across Canada.

Reference period: 2024

Collection period: A 3-month collection period starting in the reference year.

Subjects

• Health
• Mental health and well-being
• Pregnancy and births

Data sources and methodology

Target population

Parents who gave birth to a liveborn infant who is at least 6 months old and less than 12 months old during the collection period, residing in one of the ten provinces and not institutionalized or living on reserve.

Married or common-law partners of birth parents, living in the same household as the birth parent and child.

Instrument design

The PES questionnaire content was developed in collaboration with the Public Health Agency of Canada, and in consultation with experts in the field of reproductive health. Modifications were applied to comply with standards and practices of Statistics Canada during survey development work. The PES is based on the Maternity Experiences Survey (MES) conducted in 2006.

Data collection for the PES was carried out using an EQ that could be filled out by respondents (rEQ) or through an interviewer (iEQ). The application was developed once the survey content was finalized in both official languages. The EQ was tested by the Questionnaire Design Resource Centre with selected participants to evaluate understanding and ease of use. The EQ then underwent user acceptance testing (UAT) to test the individual modules of the survey content and their text, skip patterns and logic flows. Accessibility testing was also conducted to make sure that the EQ was compliant with assisted devices such as screen readers.

Once all modules were verified as correct, they were integrated with the entry and exit components in the EQ application. The application included a functionality whereby, if a birthing parent's partner was also selected for the survey, questions requesting their partner's contact information appeared in the birthing parent questionnaire. When an email address was provided and the birthing parent questionnaire was submitted, a supplementary questionnaire for the partner was generated.

End-to-end testing was conducted when the fully integrated application was placed in a simulated collection environment. Test data were then collected, transmitted and extracted in real time, exactly as would be during the collection period.

Sampling

This is a sample survey with a cross-sectional design.

Sample design
The overall sample size for the PES is 31,500, consisting of 25,500 birth parents and 6,000 other parents.

Stratification method
The frame for the Parental Experience Survey was stratified by province and by the age group of the birth mother. A simple random sample was selected independently within each age group in each province.

Sampling and sub-sampling
Sufficient sample was allocated to each age group within each of the provinces so that the survey could produce estimates for each age group at the provincial and national level.

First, an initial sample of 25,500 targeted respondents was selected and sent to collection. The selection of a sample unit is done in one stage. A list of children aged 6 to 12 months old whose parent or guardian is a Canada Child Benefit (CCB) recipient is stratified by province and age group of the birth mother (as of October 29th, 2024). Within each stratum, the list is sorted by city, mailing address, and age group, and a systematic sample of children is drawn within each stratum. The CCB recipient corresponding to each selected child is added to the sample file.

Second, out of the 25,500 selected birth mothers, a subsample of 12,000 were identified so that their spouse would be selected for a supplement questionnaire. These spouses are selected at random using probability proportional to size (PPS) sampling within each stratum.

Data sources

Data collection for this reference period: 2024-10-29 to 2025-02-09

Responding to this survey is voluntary.

Data are collected directly from survey respondents.

Collection method:
EQ with CATI follow-up

View the Questionnaire(s) and reporting guide(s).

Error detection

Data collection for the PES was carried out using an electronic questionnaire (EQ) that could be filled out by respondents (rEQ) or through an interviewer (iEQ). The application included automated edits that prompted the respondent or interviewer to verify or correct answers, for example, when an answer was considered out of range, or when certain questions such as date of birth of the respondent or child were not answered.

Data processing after collection included steps to clean up duplicate records (for example, when an interviewer completed the survey with a respondent while an unsubmitted self-response remained) and to make manual corrections, such as clarifications found in interviewers' notes. Every blank variable was given one of two processing code values, '6' for a valid skip and '9' for a not stated status. For open-ended questions, the responses were reviewed to determine whether the response fit into an existing category or where a substantial number of the same response was given, new categories were created, with the remaining responses left as "Other." Questionnaire flow errors, where respondents answered questions that did not apply to them, were identified and superfluous data were removed by following the questionnaire flow implied by answers to previous questions. Three main categories of edits were then applied: validity, consistency and distribution. Validity edits verified the syntax of responses and included such things as checking that the data lay within an allowed range of values. For example, a range edit might be placed on the reported age of a respondent to ensure that it lies between 0 and 121 years.

Consistency edits verified that relationships between questions are respected. These edits can be based on logical, legal, accounting or structural relationships between questions or parts of a question. One example is the relationship between date of birth and marital status: a person younger than 15 years cannot have a marital status other than "never married and not living common law."

Distribution edits, sometimes called statistical edits or outlier detection, were performed by looking at data across questionnaires. These edits attempted to identify records that were outliers with respect to the distribution of the data.

Imputation

Imputation is a process used to determine and assign replacement values to resolve problems of missing, invalid or inconsistent data. This is done by changing some of the missing values and the responses on the record being edited to ensure that a plausible, internally consistent record is created.

For the 2024 PES, three variables from the questionnaire needed imputation: birth parent's age, child's date of birth, and a question from the childhood maltreatment module. Furthermore, the gender variables of the birth parent and the second parent were recoded to their associated new 2-category gender variables.

The birth parent derived age variable is equal to the age provided on the questionnaire if available, otherwise it's calculated using the date of birth provided on the questionnaire. For respondents that did not answer any of these questions, their age is calculated using the date of birth from the sample file.

On the questionnaire, respondents were asked to provide the date of birth of their child that is targeted for the survey. Missing dates of birth are imputed using the child's date of birth from the sample file.

The childhood maltreatment question EVA_35B was incorrectly translated in the French questionnaire, both for the French self-respondents and the French telephone respondents. EVA_35B was imputed for this group of respondents using nearest-neighbor hot deck imputation.

Incoherences were observed within the birth parent's gender variable. For a subsample of the respondents, the gender of the child was likely reported instead of their own gender. For these cases, responses are set to missing.

For both the birth parent and the second parent, given that the non-binary population is small, data aggregation to a two-category gender variable is necessary for PES to protect the confidentiality of responses provided. In these cases, individuals in the category "non-binary persons" are distributed into the other two gender categories and are denoted by the "+" symbol.

To do so, the new 2-category variables for the birth parent's and the second parent's gender is created in two stages. First, the missing data is randomly imputed into (Men/Women/Non-binary persons) as to preserve the original distribution of the 3-category gender variable. Once imputation of the missing data is done, the random recoding to Men+1 and Women+2 of the records associated with non-binary individuals is performed.

Note that the PES is not designed to use gender variables for same-gender couples analysis.

Estimation

The Parental Experiences Survey includes three sets of weights: i) the weight of the birth parent, ii) the weight of the spouse of the birth parent, and iii) the weight of the couple.
The estimation of population characteristics from a sample survey is based on the premise that each person in the sample represents a certain number of other persons in addition to themselves. This number is called the 'survey weight'. The process of computing survey weights for each survey respondent involves several steps.

1) Each selected birth parent respondent is given an initial weight equal to the inverse of its selection probability from the sampling frame (CCB). Respondents identified as out-of-scope during collection are dropped from the sample.

2) The birth parent respondents' weights are then adjusted to account for non-response based on selected variables from the frame.

3) The birth parent respondents' weights are trimmed so that extreme weights - compared with other units of the same domain of interest - don't have a large impact on the variance.

4) Each selected second parent respondent is given an initial weight equal to the inverse of its selection probability from the sampling frame multiplied by the weight of the associated birth parent. Respondents identified as out-of-scope during collection are dropped from the sample.

5) The second parent respondents' weights are then adjusted to account for non-response based on selected variables from the frame.

6) Each respondent couple (a respondent birth parent with their respondent spouse/second parent) is assigned the final second parent weight.

7) The couples' weights are then adjusted via post-stratification, using the birth parent's province and age group, to ensure that, within each province and age group, the sum of the couple weights match with the sum of weights of the in-scope responding birth parents whose response on the survey indicates they have a spouse.

Variance estimation is based on a re-sampling method called bootstrap estimation.

The Generalized Estimation System from Statistics Canada (G-Est) was used to generate the survey weights and bootstrap weights.

Quality evaluation

While rigorous quality assurance mechanisms are applied at all stages of the statistical process, the validation and detailed review of data by statisticians is the ultimate verification of quality prior to release. Many validation measures were implemented, they include:

a. Verification of estimates through cross-tabulations

b. Consultation with stakeholders internal to Statistics Canada

c. Consultation with external stakeholders.

Disclosure control

Statistics Canada is prohibited by law from releasing any information it collects that could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.

Revisions and seasonal adjustment

This methodology type does not apply to this survey.

Data accuracy

Survey errors come from a variety of different sources. One dimension of survey error is sampling error. Sampling error is defined as the error that arises because an estimate is based on a sample rather than the entire population. Sampling error can be expressed through a confidence interval (CI) or coefficient of variation (CV).

The following are approximate sampling error estimates for Canada level estimates. These are based on average results; these are not results for a specific variable.

- Approximate length of 95% confidence intervals for a proportion of 50% (Canada level): 2.7%

- Approximate length of 95% confidence intervals for a proportion of 10% (Canada level): 1.6%

- Approximate coefficients of variation a proportion of 10% (Canada level): 4.0%