How do social scientists study energy and people?

Match six questions about a city’s heat-pump programme to six methods, and see what each can and cannot tell a modeller

X6 of Energy Systems, Step by Step: match questions to survey, interview, choice experiment, ethnography, policy analysis and participatory mapping; see what each method gives a model and cannot tell; mark which involve people; and test one survey result for selection.

Researcher · L4
12 min
After A13
Completed

Prairie City's council (fictional) weighs a heat-pump programme. Which of its questions can the energy model answer by itself?

How much the winter evening peak changes if 2,000 homes switch. Yes: given stated assumptions about the homes, an energy model computes the change in the peak.
Why renters rarely sign up. No: reasons come from people, and interviews are one way to ask them.
Which programme design residents would choose. No: preferences come from people, and a choice experiment is one way to measure them.

Six methods, what each asks, the data it needs and what it can tell a modeller

Method	Asks	Data	Gives a model	Cannot tell
Survey	How many, how much, how often, across a population.	A questionnaire answered by a sample chosen to represent a population.	Shares and rates by household group, such as who has a heat pump or who struggles with bills.	Why people act as they do, in depth; or whether a programme caused a change, without a comparison design.
Interview	Why and how, in people's own words.	Open conversations with a few chosen people, recorded and transcribed.	Reasons, barriers and assumptions worth testing, including ones a model leaves out.	How common each view is across a city.
Choice experiment	Which trade-offs people accept.	Repeated choices between offers whose features vary, made by a sample of people.	How much each feature weighs, such as bill saving against control, for adoption scenarios.	What people do once the offer is real: the choices are stated, not observed.
Ethnography	How people live with energy, day to day.	Long observation and conversation in homes or communities.	Routines a model's behaviour rules should reflect, such as how heating is really used.	Numbers that hold for a whole city.
Document-based policy analysis	How rules came about, and who shaped them.	Laws, rulings, official case records and the documents parties submitted, read and compared over time.	The rules and limits a model must respect, and which of them are open to change.	Documents alone do not show how people experience the rules.
Participatory mapping	Which places matter to people, and why.	Residents mark places on maps, in workshops or online.	Map layers used to compare possible locations, to exclude or weigh, as in lesson A19.	How the whole population feels, when few take part.

Six questions and the method that fits each best

What share of Prairie City's households had trouble paying an energy bill last year, by income group? Survey.
Why do landlords turn down a free efficiency upgrade for their rental homes? Interview.
How much bill saving would a household trade for letting the utility adjust its thermostat at the peak? Choice experiment.
How do families actually use their heating through a winter, and what does comfort mean to them? Ethnography.
How did the state's rules for rooftop solar bill credits change over ten years, and who argued for what? Document-based policy analysis.
Which places do residents want kept clear of a new solar farm, and why? Participatory mapping.

A survey finds that homes in the heat-pump programme use less energy than homes outside it. Does that show the programme cut their use?

Yes: the difference is the effect. Not by itself: homes that joined may have used less before the programme (selection).
Not by itself: the homes that joined may differ. Right: joiners may differ before the programme; lesson M14's methods build a fair comparison.
No: surveys cannot measure energy use. No: a survey can record use or bills; the trouble is the comparison, not the survey.

With JavaScript on, you match the questions to methods, open each method, mark which involve people and test one survey result.

Illustrative. Prairie City and its council are fictional. The method summaries are this course's own, at awareness level; the survey figures are EIA's, dated below.

Need a refresher? A13 on who pays and who benefits, by household group.

What you just did, as diagrams

Different questions: Each method answers its own kind of question: how many, why, which trade-off, how people live, how rules came about, which places matter.
For a modeller: Findings become inputs, household groups, behaviour rules, constraints, map layers or scenarios.
Ethics board first: Before research with people, or with identifiable records about them, ask the institution's ethics board what review and consent apply; exemptions and waivers follow its process.
Not causal by itself: A difference between joiners and others does not, by itself, establish the programme's effect; a fair comparison is a design choice of its own.
Neutral: No method is ranked above another here: each is the right tool for some questions and the wrong one for others.

Where we stop

This side topic stops at choosing a method

Social-science methods enter the course as what each can tell a modeller and how to work with the people who use them; training in a method is beyond it.

This side topic covers

What six methods answer, the data each needs and what each cannot tell.
Why research with people starts with the ethics board, which decides what review and consent apply.
How findings become a model's inputs, groups, layers or scenarios.

Left to specialists

Training in any method: sampling, interview guides, experimental design, coding.
Designing questionnaires, choice sets or mapping workshops.
Fieldwork itself.

Check yourself

Two quick checks.

Myth or fact? “An online poll of 50 volunteers shows what share of a city's households struggle with their energy bills.”

Myth
Fact

Answer

Myth. Volunteers choose themselves; a share for the whole city needs a sample drawn to represent it, as large surveys such as EIA's do.

A modeller needs to know how many of a city's homes have a heat pump today, by income group. Which method fits best?

A survey of a representative sample
Ethnography in a few homes
Interviews with installers

Answer

A survey: a count by group across a population is what surveys do; ethnography and interviews explain how and why, in depth.

Numbers, sources and what’s next

Numbers to remember

Number	What it means	Source
27%	U.S. households reporting energy insecurity in 2020, about 34 million: difficulty paying energy bills, or a home kept at an unsafe temperature because of energy costs (EIA's Residential Energy Consumption Survey).	EIA, Today in Energy · 11 April 2022, read 7 November 2026
18,496	Households that answered the 2020 survey, by web or mail questionnaire, late 2020 to early 2021.	EIA, Today in Energy · 11 April 2022, read 7 November 2026
7	Method categories given codes of practice in a review of energy social science: experiments, literature reviews, data collection, data analysis, quantitative energy modeling, qualitative analysis and case studies.	Sovacool, Axsen and Sorrell (2018), Promoting novelty, rigor, and style in energy social science · Published 3 October 2018, read 7 November 2026

Keep going

Back to the course map

Sources and further reading

EIA, Today in Energy: In 2020, 27% of U.S. households had difficulty meeting their energy needs. Link. 11 April 2022, read 7 November 2026.
- Energy insecurity: difficulty paying energy bills, or a home kept at an unsafe temperature because of energy cost concerns.
- 34 million households, 27% of all U.S. households, in 2020 (Residential Energy Consumption Survey).
- 18,496 households answered by web or mail questionnaire in late 2020 and early 2021.
- Lower-income households, households with children and renters reported more energy insecurity.
Sovacool, Axsen and Sorrell (2018), Promoting novelty, rigor, and style in energy social science: Towards codes of practice for appropriate methods and research design, Energy Research & Social Science 45, 12-42. Link. Published 3 October 2018, read 7 November 2026.
- The abstract (open access, CC BY): codes of practice for seven method categories.
- Balancing methodological appropriateness rather than relying on evidence hierarchies.
U.S. Department of Energy, Order 443.1D, Protection of Human Research Subjects (approved 5 August 2026). Link. Read 3 November 2026 (lesson R9).
- The rules for research with people on DOE-supported projects, as used in lesson R9: review, exemption determinations and informed consent.

The six method summaries are this course's own, at awareness level; the codes of practice above give each method's standards.

Spotted an error, or stuck? Write to hlli@ksu.edu. Corrections are made in place and dated here.

Text licensed CC BY 4.0, code MIT: teachers and students may reuse and adapt it, with credit to Honglin Li.