Correlations and statistical analyses in evaluating user preferences in participatory design of healthcare facilities

Piotr Springer, Agata Gawlak

doi:10.37190/arc250416

Abstract

This article presents examples of using statistical methods in researching declarations of healthcare facility users’ willingness to participate in designing these facilities, as well as the relationship between user expectations and evaluation of existing hospital built space.

Selected methods allowing for assessment of correlations between significant sociodemographic factors and user preferences concerning hospital space organization are discussed. Practical application of statistical analysis in the participatory design process is also presented.

Full article view is only available on bigger screens.

2025

4(84)

Piotr Springer*, Agata Gawlak**

Correlations and statistical analyses

of user preference assessment

in participatory design of healthcare facilities

DOI: 10.37190/arc250416

Published in open access. CC BY NC ND license

Abstract

The article presents examples of the use of statistical methods in the study of users’ declarations of willingness to participate in the design of healthcare

facilities and the relationship of users’ expectations to the evaluation of the existing built space of hospitals. Selected methods for assessing the correlation

between relevant socio-demographic factors and the spectrum of users’ functional-spatial preferences, such as Mann–Whitney U test, Spearman rank

correlation, Kruskal–Wallis ANOVA test and stepwise regression models, are discussed. The study involved medical sta of selected hospital facilities in

Poznan and caregivers of patients in the emergency room of one of Poznan’s paediatric hospitals. The surveys were conducted in 2015 and 2021. The same

research methods were used in both cases. The methods used made it possible to assess preferences towards the designed space, as well as the willingness

to participate in participatory research (in the case of sta). Consequently, it was shown that the selected statistical methods can be an eective tool in

assessing the spatial needs of users at the stage of investment programming and design. Thus, they can be used by architects and designers.

Key words: statistical methods, correlations, health care architecture, participatory design

Introduction

American architect and representative of postmodernism

Michael Graves pointed out: I am qualied to speak about

the physical properties of hospital space not because I am

an architect, but because I have spent the last three years

as a patient there (Graves 2011). These words emphatical-

ly highlight the importance of the participation of medical

sta and patients in the design of healthcare facilities and

the relevance of the choice of research methods that catalyse

participatory processes.

The design of healthcare facilities is a speciality that is

particularly determined by innovative and rapidly develop-

ing medical technologies and, at the same time, governed

by a very large number of legal regulations. The eciency

of healthcare facilities must be ensured not only when the

facility is commissioned, but during its use, with incredi

bly dynamic changes in treatment processes and forms of

patient care. Hence, adaptation to changing functional and

spatial needs is permanently embedded in the programming

and design of these investments. Changing medical technol-

ogies generate the need to modernise and adapt the building

structure more often than in other facilities.

It should also not be forgotten that in the case of hospitals,

we are dealing with dierent groups of users of the same

space, demonstrating radically dierent spatial and func-

tional needs, which is important when including the users

in the design processes. Hence, selected methods are used

which take into account the specic characteristics of the us-

ers concerned, enabling the participation of all groups in the

programming and design process. Statistical methods used

in architecture and urban planning are often inferior to the

entire spectrum of qualitative methods. Nevertheless, they

ensure that user needs are evaluated in a systematic way,

on a much larger scale, with objective weights and criteria.

At the same time, they make it possible to analyse correla-

tions, especially when including a large group of users with

* ORCID:

0000-0001-8874-7229. Faculty of Architecture, Poznan

University of Technology, Poland, e-mail: piotr.springer@put.poznan.pl

** ORCID: 0000-0002-6234-7953. Faculty of Architecture, Poz-

nan University of Technology, Poland.

168

Piotr Springer, Agata Gawlak

dierent characteristics and needs in the evaluation process

(here: patients, sta, visitors).

The aim of the research work presented in this paper was

to demonstrate the feasibility of using statistical methods as

tools to assess the spatial needs of users of healthcare facil-

ities and their willingness to participate in the investment

programming and design process.

State of research

The role of the building user in the process of dening

the spatial structure of the building is closely related to the

design model adopted for the development project. When

trying to establish the relationship between the designer and

the other stakeholders in the design process, it is worth start-

ing with an understanding of the process. Design process

methodologies are well described in the scientic literature

(Asanowicz 2010; Barełkowski 2009; Prokopska 2012). Re -

in architectural theory denes the design process

based on the knowledge, experience and talent of the desig-

ner as historically primordial (Niezabitowska et al. 2021,

178). As civilisation developed and new social expectations

of architecture emerged, the process transformed. This ob-

servation, outlined by, among others, a Swiss research team

(Hegger et al. 2008, 187), was graphically presented by

Bea ta Majerska-Pałubicka (2014, 174) (Fig. 1).

Methods

Taking this into account, and noting the specicity of hos-

pital facilities, where a huge number of sensitive medi cal

processes are carried out, it is important to select such meth-

ods of evaluating the space of a built healthcare facility that

are as non-conicting as possible with the day-to-day ope-

ration of the facility under analysis, non-invasive and at the

same time enabling the collection of valuable and re liable

data in the shortest possible time. This specicity of carrying

out research in the buildings of healthcare facilities informed

the choice of statistical methods in the two case studies de-

scribed later in this thesis: a study of the socio-demograph-

ic factors inuencing the cooperation of hospital sta with

designers potentially undertaking the transformation of the

workplaces of the employees concerned, and a study which

involved the participation of patients of a paediatric hospi-

tal in the assessment of functional and spatial needs in the

emergency room. The described studies indicate that it is

worthwhile to integrate statistical methods into scientic

work in

the discipline of architecture and urban planning.

They oer, through data analysis and correlations, a chance

to uncover hidden connections in the perception of hospital

space, characterised by objectivity and detachment appro-

priate for the purposes of research, which were emphasised

by Jean S.K. Lee (1992, 89) in her comparative analysis of

quantitative and qualitative methods (Table 1).

Selected statistical methods were indicated for the analy-

ses: the Mann–Whitney U test, Spearman’s rank correlation,

Kruskal–Wallis ANOVA test and stepwise regression models.

The Mann-Whitney U test is a non-parametric test that

demonstrates whether sample values from two independent

populations show a magnitude correlation. The basic condi-

tion for the use of the U test is that the dependent variable is

represented on an ordinal or quantitative scale. This prereq-

uisite is fullled in the representation of the answers to ques-

tions on willingness to participate (the dependent variable of

the question) as a ve-point Likert scale (ordinal scale from 1

to 5) (Mann, Whitney 1947, 50–60). In the study described

here, this test was used, among other things, to compare the

declared level of willingness to participate for two separate

groups, e.g.: Who shows a greater willingness to participate:

women or men? Is the dierence statistically signicant?

Spearman’s rank correlation is a non-parametric mea-

sure of the monotonicity of the statistical relationship be-

tween variables. It describes the strength of the correlation

of two characteristics when both are qualitative, allowing

them to be ordered by the strength of the characteristic. For

the study in question, the values of age and length of service

of the sta were analysed (sortable by their strength – in

this case a numerical value) in relation to the declaration

of willingness to participate (also sortable by magnitude,

due to the structuring of the responses on a ve-point scale)

(Spearman 1904, 72–101).

The Kruskal–Wallis one-way ANOVA – one-way

analysis of variance for rank ANOVA – is a non-parametric

extension of the Mann–Whitney U test to more than two

populations. In the present study, three primary occupation-

al groups were dened, which qualies the use of the test in

question (Kruskal, Wallis 1952, 583–621). In the procedure

described, it was used, among other things, to compare the

declared level of willingness to participate for several sepa-

rate groups, e.g.: Who shows a greater willingness to par ti-

ci pate: doctors, nurses or other employees? Is the dierence

statistically signicant?

Stepwise regression models – a statistical method that

allows the degree of inuence of several explanatory vari-

ables on a given explanatory variable to be determined. In

a certain generalisation, regression models allow further

prediction of data (explanatory variables) on the basis of

separate known characteristics (explanatory variables). For

the present study, a stepwise model was adopted, in which

further explanatory variables would be added in an order of

signicance. This model was used to indicate whether hos-

pital sta’s willingness to participate in architectural design

is more inuenced by sociodemographic factors or by the

spatial characteristics of their current workplace.

Fig. 1. Design processes: A) linear, B) iterative; C) integrated

(source: Majerska-Pałubicka 2014, 174)

Il. 1. Procesy projektowania: A) linearny, B) iteracyjny;

C) zintegrowany (źródło: Majerska-Pałubicka 2014, 174)

Correlations and statistical analyses of user preference assessment in participatory design of healthcare facilities

169

A case study of the use of statistical methods

in research on hospital staff participation

in the design process

Six hospitals in Poznań were selected to carry out a sur vey

(expert questionnaire) among sta of hospital bed wards

The research work in question

, referred, among other things,

to the past experience of medical sta with par ti ci patory de-

sign, hence an important criterion for the se lec tion of the

territorial scope (in this case the selection of hos pitals) was

to select units that had undergone signicant spa tial trans-

formations in the last ten years.

A total of 154 respondents took part in the survey. The

survey was conducted in 2021 only among those who were

employed and actively performing their duties in a given

medical facility. The vast majority of people surveyed were

women (109 respondents, 71% of the total survey group).

The survey methodology was developed by the author on

the basis of his own experience in the programming and de-

sign of healthcare facilities and the current state of research.

In formulating the structure of the survey tool, reference

was made primarily to the work of Scandinavian research-

ers at the forefront of considering participatory design of

hospitals. One of the primary ones was the methodology of

the HospiCaseY project

. In this study, the current hospital

space was evaluated by sta, in separate functional catego-

ries, using a Likert scale. The content of the survey sheet

was veried for understanding by potential respondents

The following hospitals in Poznań were selected to conduct an ex-

pert survey among the sta of inpatient wards Gynecology and Obstet-

rics Clinical Hospital of the Karol Marcinkowski Medical Universi ty in

Poznań, ul. Polna 33, 60-535 Poznań, Independent Public Healthcare Cen-

ter of the Ministry of Internal Aairs and Administration in Poznań named

after Prof. Ludwik Bierkowski, ul. Dojazd 34, 60-631 Poznań, He lio dor

Święcicki Clinical Hospital of the Karol Marcinkowski Medi cal University

in Poznań, ul. Przybyszewskiego 49, 60-355 Poznań, Ka rol Jonscher Clin-

ical Hospital of the Karol Marcinkowski Medical Uni versity in Poznań,

ul. Szpitalna 27/33, 60-572 Poznań, Transgura tion Clinical Hospital of

the Karol Marcinkowski Medical University in Poznań, ul. Augustyna Sza -

marzewskiego 84, 60-569 Poznań, Józef Struś Multispecialist Municipal

Hos pital with ZOL SP ZOZ in Poznań, ul. Szwajcarska 3, 61-285 Poznań.

Chapter based on dissertation (Springer 2023).

Overcoming the language barrier of publishing the report in Finn-

ish was helped by the report of the subsequent EVICURES research proj-

ect, which refers in detail to the output of the preceding HospiCaseY study

(Nykänen et al. 2016).

during a pilot study combined with interviews, conducted

in two medical facilities: Face&Skin Dermatology and Der-

matosurgery and Poznań Askodent Implant Centre. During

the pilot, particular attention was paid to the content of the

questionnaire, relating to participatory theory and the level

of medical sta understanding of the nomenclature, identi-

cal to that of the architecture survey.

The content of the questionnaire was contained in two

pages of A4 sheet and consisted of ve main parts:

– information about the respondents,

– questions on the declaration of willingness to partici-

pate in the design of the renovation/expansion,

– questions on the evaluation of the quality of the respon-

dents’ current workplace,

– questions on previous experience with participatory

design,

– open-ended questions about the hospital space.

The primary intention was to construct the spreadsheet in

a way that allowed it to be received unambiguously and as

intended. The resulting tool enabled the collection of data

and analysis of their internal correlation.

An important part of planning a reliable statistical survey

is determining the minimum sample. For this purpose, the

formula (Kuszewski, Podgórski 2008, 26) was used:

min

= [N

(α

f (1 – f ))]/[N

+ α

f (1 – f )],

where:

min

– minimum sample size,

–

size of the population from which the sample is taken,

α – condence level for the results,

f – size of the fraction,

e – assumed maximum error, expressed as a fractional

number.

The study assumed:

– population size – approximate, total number of doctors

and doctors and nurses: personnel actively practising their

profession in Poznań hospitals: 3990

Due to the lack of current and unambiguous data on the number of

doctors and nurses strictly for the city of Poznań (only general data from

the Central Statistical Oce and the Supreme Chamber of Physicians for

the Wielkopolskie Voivodeship are available), the size of Poznań’s med-

ical sta was calculated by assuming a nationwide ratio of 2.4

doctors

per 1,000 inhabitants and 5.1 nurses per 1,000 inhabitants (European

Quantitative Qualitative

Ontological assumption objectivity subjectivity

Epistemological assumption positivism phenomenology

Aims of inquiry universality particularity

Role of researcher outsider insider

Researcher-respondent relationship detachment involvement

Research methods statistics description

Table 1. Statistical methods in the context of quantitative and qualitative research (elaborated by P. Springer, based on Lee 1992, 89)

Tabela 1. Metody statystyczne w kontekście badań ilościowych i jakościowych (oprac. przez P. Springer, na podstawie Lee 1992, 89)

170

Piotr Springer, Agata Gawlak

– condence level – non-random, strictly dened popu-

lation assumed: 95%

– fraction size – it was assumed that the sheets would be

completed correctly by at least 90% of the population: 0.9

– maximum error – the study group was assumed to be

representative of the issue in question: 5%.

Based on the above factors, a minimum sample size of

134 respondents was determined. The data collected in the

course of this survey research made it possible to determine

which socio-demographic factors may inuence the de-

clared degree of involvement of medical personnel in design

participation. Based on the indicated statistical methods, the

investigators drew conclusions, of which the following are

worth mentioning:

1. The majority of respondents declared that they would

like to be informed about plans to renovate/expand their

workplace (in relation to the statement: “I would like to be

informed about renovation/expansion plans”, 42.21% of

respondents answered “denitely yes”, while 30.52% an-

swered “rather yes”, which adds up to 72.73% interest in

par ticipation in the form of information).

2. The majority of respondents said they would like to be

consulted on their workplace renovation/expansion project

Observatory on Health Systems and Policies 2021), for a population of

Poznań in 2020 of 532,048 inhabitants (Bieniek et al. 2021, 76).

The survey sheets were handed over to the medical sta in the so-

cial areas of the hospital wards accessible only to them, which guaranteed

a high level of condence and at the same time a low maximum error.

A level of fractionation conducive to the reliability of the study

was ensured by the implementation of the pilot study described above.

(in relation to the statement: “I would like to be consulted

on renovation/expansion plans”, 29.87% of the percentage

of respondents answered “denitely yes”, while 27.27% an-

swered “rather yes”, which adds up to 57.14% interest in

participation in the form of consultation).

In order to analyse the latent correlations based on sta-

tistical tests, it was found that, among other things, dier-

ences between the answers given by representatives of sep-

arate professional groups were noticeable and statistically

signicant for consulting and declarations of devoting free

time. In all categories of participation, doctors expressed

the greatest willingness to participate in the project process

(Table 2). On the basis of the results obtained, it can be con-

cluded that the use of statistical methods may prove helpful

in identifying specic groups of end users for whom there is

the greatest justication for engaging in participatory design

processes, due to their declaration of willingness to cooper-

ate with designers.

A case study of the use of statistical methods

in surveys to assess patient satisfaction

and spatial expectations

A study conducted in 2015 examined the participation of

paediatric hospital patients in the assessment of function-

al-spatial needs in the emergency room

(Gawlak 2019). It

used the same statistical methods. The author’s survey was

B. Krysiewicz Specialist Complex for Maternal and Child Health

Care in Poznań (hereinafter: SZOZ).

Question

Profession

Average*

Median

Min.

Max.

Lower quartile

Upper quartile

Standard

deviation

Standard error

1a) I would like to be informed

about renovation/expansion plans

doctor 62 4.24 4 1 5 4 5 0.88 0.11

nurse 67 3.90 4 1 5 3 5 1.18 0.14

other 25 3.68 4 1 5 3 5 1.28 0.26

1b) I would like to be consulted

on renovation/expansion plans

doctor 62 3.95 4 1 5 3 5 1.05 0.13

nurse 67 3.34 3 1 5 2 4 1.25 0.15

other 25 3.40 3 1 5 2 5 1.38 0.28

1c) I would like to decide on the scope

of the renovation/expansion.

doctor 62 3.56 4 1 5 3 5 1.15 0.15

nurse 67 3.19 3 1 5 2 4 1.20 0.15

other 25 3.20 3 1 5 2 4 1.26 0.25

1d) I am willing to devote my free time

to participate in the design

of the renovation/expansion of my workplace

doctor 62 3.21 4 1 5 2 4 1.29 0.16

nurse 67 2.66 2 1 5 2 4 1.25 0.15

other 25 2.96 3 1 5 2 4 1.21 0.24

Table 2. Comparison of responses to questions concerning declarations of willingness to participate, depending on occupation

(elaborated by P. Springer)

Tabela 2. Porównanie odpowiedzi na pytania dotyczące deklaracji woli partycypacji w zależności od przynależności do wykonywanego zawodu

(oprac. P. Springer)

*The darker shade of green in the table corresponds to values that were found to be statistically signicant in the Kruskal–Wallis ANOVA test.

Correlations and statistical analyses of user preference assessment in participatory design of healthcare facilities

171

conducted among 212 emergency room patients (“satis-

faction assessment”, survey 01) and 203 potential hospital

patients (“expectations assessment”, survey 02).

Satisfaction assessment

The rst part of the study (satisfaction assessment, sur-

vey 01) was conducted among the parents (or guardians) of

patients currently in the emergency room. The evaluation

of their overall impression of the emergency room and the

evaluation of the solutions integrally related to the quality

of their stay in the hospital space were studied, but in this

paper, due to the stated aim of the study, selected results

are discussed, referring to the relation of the evaluation of

the space in relation to: sociodemographic characteristics of

the respondents, time of arrival at the emergency room and

waiting time for registration.

In selecting the size of the study sample, reference was

made to the average number of paediatric patients admitted

by the emergency room at the HES in the year preceding the

survey, which was approximately 20,000 (the record num-

ber of patients admitted was 30,000 in 2004). The resulting

annual average was then converted into a monthly average.

Among the respondents, women were in the vast majori-

ty, 86%, and men 14%. The results of the comparative anal-

ysis show that statistically signicant dierences (p < 0.05)

in the opinion of women and men on the evaluation of the

quality of solutions concerned functional-spatial solutions

(colour scheme, access to drinks, art and design, other fur-

niture, play area). Women rated most of these solutions sig-

nicantly higher. Men only rated the immediate surround-

ings of the hospital, the lighting and the waiting area higher.

However, this dierence was not statistically signicant

(p ≥ 0.05). The result shows that there is a correlation be-

tween the assessment of the emergency room space and the

gender of the respondent.

Further comparative analyses were carried out to demon-

strate the relationship between the assessment of the quality

of the space and functional arrangements of the hospital and

the hour of arrival in the room of the subject. The Kruskal–

Wallis rank-sum ANOVA test was used to compare the as-

sessment related to the quality of stay and the hour of arrival

at the hospital (10.00–12.00, 12.00–15.00, 15.00–20.00). It

was assessed that those who arrived in the morning hours,

i.e., between 10.00 and 12.00, rated the functional-spa-

tial solutions in the emergency room the lowest, while the

highest average ratings were given by the respondents who

arrived in the afternoon and evening hours, i.e., between

15.00 and 20.00.

Another comparative analysis was carried out between

the subjects’ evaluation of the space and their waiting time

for registration. For this purpose, the Kruskal-Wallis rank

ANOVA test was again used. The evaluation of spatial-func-

tional solutions for issues such as lighting, art and design,

play area, waiting area, colour scheme, sanitary facilities

diered statistically signicantly (p < 0.05), depending on

the waiting time for registration. The shorter the waiting

time for registration, the higher the ratings were given to the

in dividual solutions. Those waiting more than 30 min for

re gistration were mostly less satised with the functional

and spatial solutions of the chamber compared to the other

waiting patients.

A comparative analysis was then carried out between the

waiting time for a medical examination and the qualitative

assessment of the functional and spatial arrangements of

the emergency room. For this purpose, the Kruskal–Wallis

rank-sum ANOVA test was used.

On the basis of the research carried out, one might be

tempted to conclude that sociodemographic factors and the

situational context (waiting time for a medical procedure

and the time of arrival at the hospital) should also be taken

into account when shaping tools for evaluating a hospital

building. The results (opinions) obtained through the use

of statistical methods can become an extremely important

guide line for designers of healthcare facilities who will

want to evaluate the way in which a facility that will be

subject to spatial transformations functions.

Assessment of expectations

The sample size for the expectations survey (survey 02)

was selected adequately to the sample size dened for sur-

vey 01. The survey was conducted on a randomly selected

group of 203 people (66% women, 34% men) who were not

emergency room patients at the time of the survey. Expec-

tations towards the quality of the hospital space were ex-

amined in relation to several sociodemographic factors, but

the article exemplarily discusses the results relating to the

inuence of gender and age on these expectations.

In order to demonstrate the relationship between the as-

sessment of the expected quality of hospital space and func-

tional solutions and the gender of the respondent, compara-

tive analyses were carried out. An analysis was made of the

answers given by women and by men. For this purpose, the

Mann-Whitney U test was used. The results of the test indi-

cate that the evaluations of the functional-spatial solutions

of the emergency room, such as intimacy, accessibility, were

rated higher by women, and this dierence was statistically

signicant (p < 0.05). For the other solutions, the vast major-

ity (art, waiting room, sanitary facilities, colours, furniture

and lighting) were also rated higher by women, but the dif-

ference did not show statistical signicance (p ≥ 0.05). Men

only rated access to refreshments, the hospital building envi-

ronment and the changing room higher than women, this dif-

ference was also not statistically signicant (p ≥ 0.05). The

above proportions are analogous to the results obtained in

survey 01, where women also rated most of the chamber’s

functional-spatial solutions higher than men.

The results of the test indicate that the gender of the user

of a space has little inuence on the evaluation of the func-

tional-spatial solutions if they are not a user of that space at

the time of the survey (here: potential hospital patient). The

most convergent responses of the respondents were indicat-

ing: tv, a separate room or so-called other form of improve-

ment within the emergency room. The biggest dierence

in the answers given concerned: so-called other solutions

(22% women vs. 10% men), wi- access (14% women vs.

20% men).

Next, the relationship between the assessment of the qua li -

ty of the space and the age of the respondents was compared.

172

Piotr Springer, Agata Gawlak

To test the correlation of data related to quality of stay and

age of carers (in terms of the age groups adopted in the

stu dy: 18–25, 26–35, 36–45 and over 45 years), the Krus-

kal– Wallis rank-sum ANOVA test was used. Respondents

responded most concurrently by indicating: tv, a separate

room or so-called “other” form of improvement within the

emergency room.

Relationship of satisfaction ratings to expectations

The results of the study were then collated using, among

other things, statistical analysis, including comparative anal-

yses of the two research models (“satisfaction assessment”

vs. “expectation assessment”). In both models, the inuence

of factors such as the respondent’s age, gender, time of day,

waiting time for registration or medical examination on the

perception of hospital space was considered. On this basis,

as well as in the light of the literature studies carried out, it

was proven that the functional-spatial solutions of the emer-

gency room have a signicant impact on the well-being,

safety and satisfaction of patients. The analysis carried out

was intended to conrm that emergency room patient satis-

faction is shaped on many levels related to variables such as

gender or age, while for qualitative studies of the facilities,

the status of the users surveyed is also of great importance,

i.e., whether they make their assessments while “in situ” or

based only on their expectations of the assessed (surveyed)

space. The evaluations of specic functional-spatial solu-

tions of the reception room from both survey studies (“satis-

fac tion evaluation” vs. “expectations evaluation”) were

com pared with each other. The Mann-Whitney U-test was

used to compare the results of the two surveys. Thus: in sur-

vey 01, the following were rated highest: lighting, chang-

ing rooms, access to drinks and art, while in survey 02 the

following were indicated as most important: lighting of the

chamber, accessibility, waiting room (overall impression),

sanitary facilities and the hospital environment.

Importantly, respondents in the emergency room gener-

ally made proportionately lower assessments of spatial ar-

rangements than prospective patients (Fig. 2).

The second comparison (“satisfaction rating” vs. “expec-

tation rating”) concerned the respondents’ suggested possi-

ble improvements within the hospital emergency room.

Similarly, the responses of those who stay in the emer-

gency room were compared with those of respondents who

can imagine this stay. The so-called test of dierences be-

tween the two percentage structure indicators was again ap-

plied. The results show statistically signicant dierences

in the assessment of the impact of individual facilities on

the comfort, satisfaction and sense of security of emergency

room patients (Fig. 3).

By demonstrating signicant dierences in the percep-

tion of hospital space and qualitative assessment between

those surveyed and present in the emergency room at the

same time during the survey and those surveyed who are

potential hospital patients, the expectations of potential hos-

pital patients were veried in relation to their actual sen-

sations and actual perception of hospital space and spatial

needs.

The conclusions of the described research work (Gawlak

2019) point to the possibilities of using the results of the

conducted research and, above all, to the necessity of con-

tinuing them in a broader context, i.e., taking into account

other, including “non-architectural” factors inuencing

patients’ satisfaction with their hospital stay. Comprehen-

sive qualitative studies of architectural space, taking into

account, for example, the acoustic quality of the space, the

reputation of the hospital in question or the reputation of the

medical sta, will allow an objective diagnosis of this space

to be made by assessing the satisfaction of its users and,

consequently, to properly guide design work when mod-

ernising, rebuilding, renovating or building a new hospital.

In view of the above, it can be stated with certainty that

the qualitative assessment of the architectural space should

be carried out taking into account the patient’s expectations.

This is conrmed by the analysis presented, according to

which the potential patient’s expectations of the hospital ar-

chitecture are relatively higher than the satisfaction rating of

the patient staying in the hospital (Fig. 2).

In essence, the indicated conclusions of the studies sup-

port the thesis that hospital architecture shapes the com-

The Y-axis indicates the number of points allocated by respondents on a five-

point scale indicating how important each spatial aspect is to them, where:

1 – don’t know, 2 – not important, 3 – not very important, 4 – quite important,

5 – very important.

The Y-axis indicates the number of people, who have pointed out a solution

listed on the X-axis.

Fig. 2. Assessment of the quality of emergency room solutions.

Comparison of satisfaction with expectations (elaborated by A. Gawlak)

Il. 2. Ocena jakości rozwiązań w obrębie izby przyjęć.

Porównanie satysfakcji z oczekiwaniami (oprac. A. Gawlak)

Fig. 3. Comparison of indicated possible improvements

between patients, residing in the emergency room and -responders

who are potential patients (elaborated by A. Gawlak)

Il. 3. Porównanie możliwych ulepszeń wskazywanych przez pacjentów

przebywających w izbie przyjęć i respondentów będącymi

potencjalnymi pacjentami (oprac. A. Gawlak)

Correlations and statistical analyses of user preference assessment in participatory design of healthcare facilities

173

fort and well-being of patients, depending on socio-de-

mographic and other conditions, such as waiting times for

registra

tion and medical examination, and, above all, pa-

tients’ ex pectations of the characteristics of the built space.

The con clusions of the described research demonstrate the

applica bility of statistical methods in the pre-design evalua-

tion process of the built space of healthcare facilities.

Summary

Both case studies presented above used convergent sta-

tistical methods. The aim was to create project recommen-

dations based on reliable data.

A limitation of the conducted studies, but also of simi-

lar ones, is usually the so-called institutional barriers, often

the fear of evaluation and possible detection of irregulari-

ties, sometimes simply the reluctance to introduce research

teams evaluating space and functional-spatial solutions into

the constructed healthcare facility.

The life cycle of a building consists of stages that last

from several months to several tens or hundreds of years

(Niezabitowska 2004). For the creation of space quality in

architecture, by far the most important is the planning and

programming phase of a building, and then the actual archi-

tectural design, because it is at the planning and program-

ming stage of an investment that decisions are made relating

to, for example, the choice of location for the building (in

the case of a new hospital) and the creation of a functional

programme begins, which is of key importance for the qual-

ity of the space designed later.

Consequently, it is important from the users’ point of

view to programme healthcare buildings properly and com-

prehensively, based on qualitative studies for buildings and

qualitative assessments for buildings with similar functional

programmes.

To this end, during the programming stage of an invest-

ment, whether it involves a new hospital or the expansion,

renovation or modernisation of an existing one, eorts should

be made to

include representatives of all groups of future

users of the building in the design process: sta and pa

tients, which is possible – this is conrmed by the case stud-

ies cited in this article. Thus, it should be concluded that the

aim of the research work as stated in the introduction (statis-

tical methods can be an eective tool in assessing the spatial

needs of users at the stage of programming and designing an

investment) has been achieved.

This is borne out by the Evidence-Based Design process

in hospital design, i.e., making design decisions related to

the functional-spatial programme based on the results of re-

liable scientic research.

In terms of practical recommendations arising from this

research work, it is worth highlighting:

The need to use nomenclature in the survey sheets that

is understandable to the recipient – depending on the re-

search: sta or patients. To this end, pilot surveys should be

carried out before the implementation of the targeted studies.

2. With regard to surveys involving hospital sta – when

formulating the survey, it is necessary to narrow its scope

to aspects relevant to the investment in question, so that the

survey does not involve long-term involvement of the sta,

who are likely to ll in the questionnaires during their work.

3. In relation to surveys with the participation of patients

– both the observations of this user group in relation to the

existing space and the expectations towards the space after

the implementation of the target design intervention (con-

struction, reconstruction or extension) should be considered

relevant.

4. When selecting the study group, particular attention

should be paid to ensure a minimum size, determined by the

formula indicated in chapter: “Case study of the use of sta-

tistical methods in research on designer participation with

hospital sta”.

At the same time, consideration should be given to lim-

iting access to the surveys to designated users only, which

will increase the condence level for the results.

Translated by

Agata Gawlak

References

Asanowicz, Aleksander. “Geneza metodologii projektowania.” Architec-

turae et Artibus 2, no. 4 (2010): 11–8.

Barełkowski, Robert. “Zapiski o procesie projektowym.” Przestrzeń

i Forma 12, (2009): 85–102. https://doi.org/10.13140/2.1.3219.4242.

Bieniek, Agnieszka, Emilia Bogacka, Ewa Filipczuk, Marta Kowalczyk,

Katarzyna Piętak, and Maciej Pilarski. Rocznik statystyczny woje-

wództwa wielkopolskiego. Urząd Statystyczny w Poznaniu, 2021.

European Observatory on Health Systems and Policies. “State of Health

in the EU, Polska, Prol systemu ochrony zdrowia 2021.” Pub-

lished December 13, 2021. Accessed January 20, 2022, at www.

oecd-ilibrary.org/social-issues-migration-health/poland-coun-

try-health-prole-2021_e836525a-en.

Gawlak, Agata. Architektura szpitala: izba przyjęć w szpitalu pe dia t

rycz nym. Wydawnictwo Politechniki Poznańskiej, 2019.

Graves, Michael. “Michael Graves at TEDMED 2011.” Video, 25 min.,

45 sec. Accessed June 6, 2024, at https://www.tedmed.com/talk/mi-

chael-graves-at-tedmed- 2011/.

Hegger, Manfred, Mathias Fuchs, Thomas Stark, and Martin Zeumer.

Energy Manual, Sustainable Architecture. Birkhauser, 2008.

Kruskal, William H., and Wilson A. Wallis. “Use of Ranks in One-Cri-

terion Variance Analysis.” Journal of the American Statistical Asso-

ciation 47, no. 260 (1952): 583–621. http://dx.doi.org/10.1080/016

21459.1952.10483441.

Kuszewski, Piotr, and Jarosław Podgórski. Statystyka, wzory i tablice.

Szkoła Główna Handlowa w Warszawie, 2008.

Lee, Jean S.K. “Quanitative Versus Qualitative Research Methods – Two

Approaches to Organization Studies.” Asia Pacic Journal of Ma na

ge ment 9, no. 1 (1992): 87–94, https://doi.org/10.1007/ BF

01732039.

Majerska-Pałubicka, Beata. Zintegrowane projektowanie architekto nicz

ne w kontekście zrównoważonego rozwoju. Doskonalenie procesu.

Wydawnictwo Politechniki Śląskiej, 2014.

Mann, Henry, and Donald Ransom Whitney. “On a Test of Whether one

of Two Random Variables is Stochastically Larger than the Other.”

Annals of Mathematical Statistics 18, no. 1 (1947): 50–60. http://

dx.doi.org/10.1214/aoms/1177730491.

Niezabitowska, Elżbieta, ed. Wybrane elementy Facility Management

w architekturze. Wydawnictwo Politechniki Śląskiej, 2004.

174

Piotr Springer, Agata Gawlak

Niezabitowska, Elżbieta, Anna Szewczenko, Michał Tomanek, and Mag -

dalena Jamrozik-Szatanek. Projektowanie obiektów szpitalnych, rola

badań naukowych w doskonaleniu jakości funkcjonowania szpi tali.

Wydawnictwo Naukowe Śląsk, 2021.

Nykänen, Esa, Tuomaala Pekka, Laarni Jari, et al. A UserOriented, Evidence-

based Besign Project of the First Finnish Single Room ICU. Results of

EVICURES Project, VTT Technical Research Centre of Finland, 2016.

Prokopska, Aleksandra. Projektowanie architektoniczne. Procesy wstęp-

ne. Ocyna Wydawnicza Politechniki Rzeszowskiej, 2012.

Spearman, Charles. “The Proof and Measurement of Association be-

tween Two Things.” The American Journal of Psychology 15, no. 1

(1904): 72–101, https://doi.org/10.2307/1412159.

Springer, Piotr. Projektowanie partycypacyjne obiektów szpitalnych.

PhD

diss., Politechnika Poznańska, 2023.

Streszczenie

Korelacje i analizy statystyczne w ocenie preferencji użytkowników w projektowaniu partycypacyjnym obiektów ochrony zdrowia

W artykule zaprezentowano przykłady wykorzystania metod statystycznych w badaniach deklaracji woli partycypacji użytkowników obiektów

ochrony zdrowia w projektowaniu tych obiektów oraz relację oczekiwań użytkowników do oceny zastanej przestrzeni zbudowanej szpitali. Omówione

zostały wybrane metody pozwalające na ocenę korelacji pomiędzy istotnymi czynnikami socjodemogracznymi a spektrum preferencji funkcjonalno

-przestrzennych użytkowników, takie jak: test U Manna–Whitneya, korelacja rang Spearmana, test ANOVA Kruskala–Wallisa oraz krokowo postępu-

jące modele regresyjne. W badaniu wzięli udział pracownicy medyczni wybranych obiektów szpitalnych Poznania oraz opiekunowie pacjentów izby

przyjęć jednego z poznańskich szpitali pediatrycznych. Badania przeprowadzono w 2015 i 2021 r. W obu przypadkach wykorzystano te same metody

badawcze. Zastosowane metody umożliwiły ocenę preferencji względem projektowanej przestrzeni, jak również wolę udziału w badaniach partycypa-

cyjnych (w przypadku personelu). W konsekwencji wykazano, iż wybrane metody statystyczne mogą stanowić efektywne narzędzie w ocenie potrzeb

przestrzennych użytkowników na etapie programowania i projektowania inwestycji. Tym samym mogą zostać wykorzystanie przez architektów oraz

projektantów.

Słowa kluczowe: metody statystyczne, korelacje, architektura ochrony zdrowia, projektowanie partycypacyjne