Survey data for the conference paper 'Mixed-Methods Evaluation of Critical Data Requirements for Heat Pump Product Configuration' presented at CIB W78 2026 – 43rd International CIB W78 Conference on IT in Construction.
This repository is an appendix to the associated research paper (see
CITATION.cff). It provides:
- the survey response data and its codebook,
- evaluation figures for each question block, most of which did not fit into the conference paper,
- descriptive and exploratory analysis tables for each question block.
This paper has a companion paper at ECPPM 2026, Critical Information Requirements for Building Service Product Selection: An Evaluation of Open Information Specification Formats, which references the survey published here. Its supplementary material — example IDS and LOIN specifications for heat pumps — is available at:
https://github.com/Design-Computation-RWTH/ecppm-2026-productSelection
Note on language. The dataset and codebook are in German, as the survey was conducted in German and no translation layer is published for the original survey data. All statistical analysis artifacts (figures and tables) in this repository appendix are however in English, matching the translated survey items as presented in the conference paper.Additionally, this README briefly describes every question block in English so the original survey data can be re-interpreted without reading German.
The survey was carried out as part of a research project funded through the Zukunft Bau research funding programme: https://www.zukunftbau.de/projekte/forschungsfoerderung/1008187-2424
ISO 16757 defines data structures for product data of technical building equipment. The survey investigates which heat pump properties practitioners need, when they need them over the course of a project, and how important they are — in order to inform the ongoing standardisation work.
The online survey was conducted with SoSci Survey, the data was collected between 13.04.2026 and 03.09.2026 at https://www.soscisurvey.de/rwth_aachen_iso16757/.
| Stage | n |
|---|---|
| Raw records with at least one page submitted | 48 |
| Records with at least one question answered (published dataset) | 42 |
Records with professional role (BE02) answered (paper analysis set) |
36 |
Questions were generally optional, so analyses should report the sample size of the relevant block. The six process-only records are excluded from the published dataset; all other records are retained to permit alternative inclusion criteria. The paper's analysis set can be reproduced with:
d <- read.csv("data/survey_responses.csv")
analysis <- d[!is.na(d$BE02) & d$BE02 > 0, ] # n = 36| Path | Description |
|---|---|
data/survey_responses.csv |
Survey responses (UTF-8, . as decimal separator) |
data/codebook.xlsx |
SoSci Survey codebook: variable labels, response codes and response labels |
analysis/figures/ |
Evaluation figures |
analysis/tables/ |
Descriptive, distribution, group-comparison and joint-analysis tables |
survey_responses.csv is a verbatim subset of the original SoSci Survey export:
only whole records and columns were removed. The codebook is published unchanged
and therefore also lists excluded variables.
Variable prefixes identify question blocks; _NN follows display order.
| Prefix | Content |
|---|---|
BE* |
Professional role and background, project and heat-pump experience |
ME02 |
First project phase in which each of 46 heat-pump properties is needed |
ME04 |
Importance of the same 46 properties on a six-point scale |
GE02–GE04 |
Importance of geometry detail, connections and product spaces |
RN* |
Experience with and importance of VDI 3805 and ISO 16757 |
SO02–SO06 |
Software, exchange platforms and formats, and product-data sources |
SD* |
Seriousness check and closing free-text comments |
ME02 stores card positions on a continuous 0–1000 timeline divided into
eight equal HOAI project phases:
| Value range | Phase |
|---|---|
0–124 |
LP1 |
125–249 |
LP2 |
250–374 |
LP3 |
375–499 |
LP4 |
500–624 |
LP5 |
625–749 |
LP6 |
750–874 |
LP7 |
875–1000 |
LP8 |
ME02 and ME04 use matching item numbers and can be joined item by item.
The remaining fields (CASE, timestamps, page durations, completion state and
similar variables) are SoSci Survey process metadata.
| Code | Meaning |
|---|---|
-9 |
Item left unanswered / skipped (for ME02: card left unplaced) |
-1 |
Explicit "unknown / no answer" option selected |
| empty | Page not reached (structural non-response) |
The GNU R export writes checkbox values as F/T, whereas the codebook uses
1/2 (not selected/selected). For FINISHED and Q_VIEWER, F/T
corresponds to 0/1.
Two scales have non-sequential codes and must be ordered by their labels:
BE08:1(none),2(low),5(some),3(considerable),4(much).RN04andRN08:1, 2, 3, 5, 6, 7; code4is unused.
QUESTNNR identifies the full (iso16757_umfrage) or shortened
(iso16757_umfrage_kurz) questionnaire. The shortened version omitted:
| Variables | Question | Asked until |
|---|---|---|
SO03, SO03_01–SO03_15, SO03_15a |
Data exchange platforms | 2026-06-12 |
RN07, RN08, RN09_01 |
ISO 16757 experience / importance / comment | 2026-07-10 |
These responses remain published but are not part of the paper analysis. Their reduced sample size must be considered in reuse.
| Variable | Reason |
|---|---|
BE05, BE06_01, SD02, SD03 |
Low informational value and re-identification risk in a small professional sample |
SERIAL, REF, MAILSENT |
Administrative metadata fields, empty in every record |
The unchanged codebook still lists these fields. Six unanswered records were
also removed. Original CASE numbers were retained and are therefore not
consecutive.
This project was funded by the Federal Institute for Research on Building, Urban Affairs and Spatial Development on behalf of the Federal Ministry for Housing, Urban Development and Construction, using funds from the Zukunft Bau research funding programme. We want to thank our project partners for their versatile support.
Data and figures are licensed under
CC-BY-4.0 — see LICENSE.
CC BY 4.0 is one-way compatible with the CC BY-SA 4.0 licence of the project report: this material may be reused in the report and in other ShareAlike works, provided attribution is given. The reverse does not hold.
See CITATION.cff, or use the "Cite this repository" button on GitHub.