Research and Development in the Higher Education Sector (RDHES)
Detailed information for 2014-2015
This survey collects information related to research and development (R&D) in post secondary institutions in Canada, in particular information related to faculty teaching, research, administration and service. The data from the survey is an important component in estimating higher education research and development expenditures (HERD).
Data release - July 29, 2015
The objective of the survey is to derive coefficients of time spent by faculty members on R&D activities by academic field of science and higher education institution size. These coefficients will be used in a Statistic Canada's model to obtain total expenditure of the Higher education sector on R&D which is also known as HERD (see record number 5109).
- Human resources in science and technology
- Research and development
- Science and technology
Data sources and methodology
The target population for this survey consists of the faculty members that were recorded as working in Canada's higher education sector.
The research and development in the higher education sector (RHES) survey used an electronic version of the survey questionnaire to collect information from respondents.
The questionnaire was designed and developed using the OECD guidelines as outlined in the Frascati Manual: guidelines for collecting and reporting data on research and experimental development (2015). The questionnaire was designed to account for the different types of activities performed by the faculty members in a given academic year.
Questionnaire content was tested in both official languages. The questionnaire was tested in December 2014. In depth one on one interviews were conducted with various members of the higher education sector in Montreal, Ottawa and Toronto to ensure the questions, concepts and terminology were appropriate. Recommendations provided by Questionnaire Design Resource Centre were reviewed and changes were incorporated in the questionnaire.
This is a sample survey with a cross-sectional design.
The HESA frame was cross-referenced with the HERD model to ensure all universities were sampled for the survey. A sample of approximately 40,000 faculty members' e-mail addresses was selected. The number of e-mail addresses selected within each sampled university depended on the size of the higher education institution. The size criteria were obtained from the HERD model and consisted of institutional information on the amount of research and development conducted, operating expenditure and the number of doctorate programs. The survey sample provides comparable representation between the natural and social sciences.
Data collection for this reference period: 2015-02-09 to 2015-03-23
Responding to this survey is mandatory.
Data are collected directly from survey respondents.
Data were collected through an electronic questionnaire. Initial contact was made in both official languages by e-mail. Follow-up is done with the respondent when necessary and in cases of non-response.
Data capture was made electronically with the support of Statistics Canada's Collection Planning and Research Division (CPRD). Average time to complete was 40 minutes.
View the Questionnaire(s) and reporting guide(s) .
Consistency edits were integrated into the electronic questionnaire. These edits helped assure the sum of parts equalled the total reported; but were not mandatory.
For key variables, the consistency edits were re-applied during processing to trigger imputation.
Outlier detection edits were also applied for key variables during data processing, first at a macro level to find outliers, and then at a micro level to trigger imputation.
For key variables describing the time spent on different professional activities, totals were adjusted to equal the sum of parts.
"Other, please specify" responses to key classification variable on the field of science were manually imputed.
Responses that were outliers for key variables on the time spent on different professional activities had both the key variable's values reduced, and their weight reduced so that they were only self-representing.
A model was applied to combine time spent on different professional activities per week when classes were in session and not in session.
Estimates were produced using the Horvitz-Thompson estimator, together with a weight adjustment for total non-response and an adjustment to re-distribute the weight of outlier responses.
Data were verified to ensure internal consistency.
The time-use coefficients derived from the survey were compared with coefficients previously obtained from Estimation of Time Spent on Research in the Higher Education Sector survey, 2001 and applied to Higher Education in Research and Development in Canada (HERD) model. Internal comparisons of the results were done between both outputs in the HERD model to determine the quality of the results. Investigation was done on large outliers by comparing results with other sectors conducting research and development in Canada.
Due to the uniqueness of this survey, national and international comparison of time-use coefficients were also verified based on published sources from various countries.
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.
In order to prevent any data disclosure, confidentiality analysis is done using the Statistics Canada Generalized Disclosure Control System (G-Confid). G-Confid is used for primary suppression (direct disclosure) as well as for secondary suppression (residual disclosure). Direct disclosure occurs when the value in a tabulation cell is composed of or dominated by few enterprises while residual disclosure occurs when confidential information can be derived indirectly by piecing together information from different sources or data series.
Revisions and seasonal adjustment
This methodology does not apply to this survey.
As the frame was created approximately two years before data collection, the sampling and estimation methodologies were designed to control for potential coverage errors and reduce their potential effects on the estimates. However, the survey results remain subject to both sampling and non-sampling error. Sampling error results when observations are made only on a sample and not on the entire population. All other errors arising from the various phases of a survey are referred to as non-sampling errors. For example, these types of errors can occur when a respondent provides incorrect information or does not answer certain questions; when a unit in the target population is omitted or covered more than once; when an out of scope unit is included by mistake or when errors occur in data processing, such as coding or capture errors.