Dataset from the survey of ventilation provision and use in homes in Great Britain
Description
This dataset contains self-reported responses from an online questionnaire survey conducted on the 17th of June 2022. The survey was completed by 2,039 adults in England, Wales, and Scotland. The survey was designed to assess the provision and use of ventilation in homes, including mechanical ventilation systems, trickle vents, and window-opening behaviours. The data was collected by YouGov, using a structured online questionnaire that included questions across four main categories: • Socioeconomic and demographic characteristics of respondents (e.g., age, gender, employment status, household composition). • Dwelling features, including construction, heating system type, and ventilation provisions. • Ventilation behaviour, including the frequency of window and ventilation use, perceived barriers, and other behavioral factors. • Contextual factors, including perceptions of indoor air quality, presence of damp or mould, and knowledge of ventilation. The dataset is provided as a comma-separated value (.csv) file containing 396 variables. The responses are formatted as binary, continuous, discrete, categorical, and free-text data. Supporting files include: • Questionnaire (.docx): a complete copy of the survey, including question routing and response types. Note that the original survey was issued using YouGov's online platform. • Variable description file (.xlsx): contains a description of all variables, valid ranges, missing data codes, and frequency distribution of variables. This dataset can be utilized by researchers, public health professionals, and policymakers to analyse self-reported ventilation behaviours and their relationship with household and socio-demographic characteristics. Considerations for Interpretation: • The data is self-reported - responses reflect participants’ perceptions and behaviours rather than direct measurements. • The survey was conducted during the summer, which may influence responses (time of observation bias and recall bias). • The sample is large and broadly representative but is not strictly representative of the British population or housing stock.
Files
Steps to reproduce
Data Processing and Cleaning The collected data were processed and analysed using Python. The key steps included: 1. Data Cleaning: • Binary responses (e.g., Yes/No) were converted into Boolean values (True/False). • Responses completed in less than 5 minutes were excluded. • Mutually exclusive responses, such as contradictory reports on ventilation types within the same household, were identified and removed. 2. Data Classification: • Homes were categorized based on construction age (e.g., pre-1980, pre-1990) to analyse ventilation trends over time and in relation to the building regulations. • Mechanical ventilation in kitchens and bathrooms were identified and grouped based on responses about different ventilation systems. • Window opening behaviour was categorised for daytime and nighttime use in both the summer and the winter to analyse seasonal ventilation trends. 3. Data Analysis: • Frequency distributions and summary statistics were generated to examine variable counts and response patterns. • Crosstab analysis was conducted to compare ventilation practices by dwelling type, tenure, and household characteristics. • Chi-square tests were used to determine the statistical significance of relationships between factors, such as ventilation provision and dwelling age. • Data was structured to allow further policy-relevant analysis.