Working out the survey questions: the temperature

August 17, 2021
survey

Working out the survey questions: the temperature

Why do we want to ask people how hot they feel?

People rarely get asked whether they feel hot. Experts assume that they simply know, after all, if the temperature is above 30ºC then probably everyone is feeling hot, right? There are even special models, used by architects to design comfortable buildings, that predict how many people would feel too hot or too cold in a given situation. For example, the classical Fanger model uses air temperature, mean radiant temperature, air speed, relative humidity, people’s activity and clothing. However, multiple studies (summarized by Rupp et al. 2014) showed that people experience heat (and cold) differently depending not only on their age, gender etc. but also depending on the region of living, season (people tolerate higher temperatures in the summer than in the winter), expectations and the level of control over the situation. Therefore, models cannot easily predict how people feel, so we should actually ask them.

How to ask about the experienced temperature?

Our first idea was simply to ask people “How do you feel?” and offer a set of options, preferably color coded:

Cold

Cool

Slightly cool

Neutral

Slightly warm

Warm

Hot

But as scholars we decided to conduct a literature review first. Out of all the texts we have read, we found the article by Marcel Schweiker and others (2016) to be the most useful. Schweiker and others showed that people’s experiences are multidimensional, for example one might feel hot and comfortable at the same time. And we might miss that if we only ask whether someone feels hot. Thus, we decided to expand the number of questions about temperature, and follow the ISO-10551 standard of asking such questions.

The ISO-10551 standard is, well, a standard of how to ask questions, but it still leaves some wiggle room. However, the gist of it is that we should first ask about perception, then about evaluation and finally about preference. Furthermore, the standard also includes questions on acceptability and tolerance. So, instead of asking one question the ISO advises asking five… which is a lot.

The perception question

The ISO standard questionnaires are typically aimed at office workers, or areas with limited heat/cold exposure, so the hottest they get is “hot” and the coldest is “cold”. However, our study will take place over the summer, when heatwaves are possible, thus (in line with ISO suggestion for “more intense” environments), we will include “very hot” and for symmetry “very cold”. We could also have “extremely hot” but that would make the scale span 11 points, which might be too much even for tolerant study participants.

1. How are you feeling now?

Are you feeling …

Very cold

Cold

Cool

Slightly cool

Neutral

Slightly warm

Warm

Hot

Very hot

The evaluation question

Respondents who find the temperature hot might still find it comfortable, or even enjoyable. Imagine sitting at a sunny beach with a drink in your right hand and an umbrella shading your head. It might be hot, but still really nice. Thus asking about comfort makes sense, but we also need to take note of the situation in which we are asking.

2. Do you find this…?

Comfortable

Slightly uncomfortable

Uncomfortable

Very uncomfortable

Extremely uncomfortable

The preference question

However, I found the third question suggested by ISO to be rather unintuitive, “How would you prefer to be now?”. I mean, if someone is very hot and uncomfortable, then you don’t need a detective to predict that this person would prefer to be (much) cooler. And we are going to study people over the summer, so they are going to feel hot. Or not… If they have an AC going inside, they might feel neutral or only slightly warm. If someone felt slightly warm, and found it uncomfortable, it is possible that they would prefer it to be a little warmer instead of cooler. Thus, we include the third question, despite the risk that the respondent might get frustrated about too many questions.

3. How would you prefer to be now?

Much cooler

Cooler

Slightly cooler

Neither warmer nor cooler (no change)

Slightly warmer

Warmer

Much warmer

The acceptance and tolerance questions

The ISO standard suggests including further two questions in the set: first, about acceptance of the environment; second, about tolerability of the environment. The acceptance question makes sense if it is asked by an employer or an architect responsible for the environment. For example, if workers feel hot and uncomfortable, they might not accept it if they are working as clerks at an office, but accept it if they work as glass blowers at a glass factory. In our survey, it is the respondents who (somewhat) control their environment, thus we think we can skip that question. We are, though, very much interested in the last question: asking respondents whether they find the temperature tolerable or bearable. We chose the word “bearable” because it has a better sounding translation into Polish, which might make it easier for our respondents to answer.

4. For you, is this temperature…?

Perfectly bearable

Slightly difficult to bear

Fairly difficult to bear

Very difficult to bear

Unbearable

The question translations

Schweiker published another paper with multiple scholars (2020), where they analysed results from questionnaires in 21 countries. Unfortunately, their results suggest that language matters and for example the word “hot” might be interpreted differently in different languages. Luckily, in their analysis translations into Polish and Spanish did not stand out as the most exceptional, thus the differences should be minor. Also, the paper provides examples of questionnaire translations to both Polish and Spanish, so we will be able to use them as the basis for our translations.

Fun fact!

Apparently, people expect spaces with air conditioning to provide ideal and stable temperature, thus they become less tolerant of a temperature that is not perfect. At the same time, those same people sitting in naturally ventilated buildings tolerate wider ranges of temperature and are even ready to interpret them as comfortable. Furthermore, people are far more content with the temperatures if they can somehow control them, either through opening windows or using fans (Rupp et al. 2014). Thus, large buildings with centrally controlled air conditioning seem like a recipe for harsh temperature dissatisfaction, despite the controllers’ best efforts.

Bibliography

Technical Committee : ISO/TC 159/SC 5. ‘ISO 10551:2019 Ergonomics of the Physical Environment — Subjective Judgement Scales for Assessing Physical Environments’. International Organization for Standardization, 2019. https://www.iso.org/standard/67186.html.

Rupp, Ricardo Forgiarini, Natalia Giraldo Vásquez, and Roberto Lamberts. 2015. ‘A Review of Human Thermal Comfort in the Built Environment’. Energy and Buildings 105 (October): 178–205. https://doi.org/10.1016/j.enbuild.2015.07.047.

Schweiker, Marcel, Maíra André, Farah Al-Atrash, Hanan Al-Khatri, Rea Risky Alprianti, Hayder Alsaad, Rucha Amin, et al. 2020. ‘Evaluating Assumptions of Scales for Subjective Assessment of Thermal Environments – Do Laypersons Perceive Them the Way We Researchers Believe?’ Energy and Buildings 211 (March): 109761. https://doi.org/10.1016/j.enbuild.2020.109761.

Schweiker, Marcel, Xaver Fuchs, Susanne Becker, Masanori Shukuya, Mateja Dovjak, Maren Hawighorst, and Jakub Kolarik. 2017. ‘Challenging the Assumptions for Thermal Sensation Scales’. Building Research & Information 45 (5): 572–89. https://doi.org/10.1080/09613218.2016.1183185.