UEQ Data Analysis Tool

Author: Dr. Martin Schrepp

The goal of this tool is to make the analysis of short version of the UEQ (User Experience Questionnaire) data as easy as
possible for you. You simply enter the data obtained from your respondents into the Data Worksheed. The tool then
automatically calculates all statistics necessary to interpret the result. In addition several graphs are shown that you may
want to use in your result presentations. Simply copy them to your word processor of presentation tool.
The instructions in this tool are in English. To make the interpretation of the results easier and to have the charts in your
language you can switch the language in the following dropdown.

Choose your Language: English

Please Note: Changing the language will show the UEQ items and the UEQ dimensions in the selected language. All other
texts will remain in English.

The tool contains the following work sheets.

Data: Here you can enter the data of your study.
DT: Rescales the data to the range -3 to 3 and calculates the scale values for pragmatic and hedonic quality per person.
Results: Main results of the questionnaire. Here the scale means and the mean and standard deviation per item are
calculated.
Confidence_Intervals: Confidence intervals for the scale means and for the mean of each item are calculated.
Scale_Consistency: Here the Cronbach-Alpha Coefficient per scale is calculated. Check here if the scales have a sufficiently
high Alpha-Value. Scales with a small value (<0.6) should be treated carefully in the interpretation.
Benchmark: The benchmark shows you in an easy way how good the evaluated product is compared to the products in th
benchmark data set. There is also a graphic that shows for each scale how the results are related to the products in the
benchmark data set. Until there are enough data for the short version available the benchmark is based on the values
obtained from the full UEQ, so please interpret this with care.
Inconsistencies: On this work sheet the data are analysed for inconsistent answers, i.e. answers from participants that
answered at least a part of the items randomly.
Items: Contains the item and dimension names in the different languages. Used for switching the language.
 Analysis Tool
. Martin Schrepp

of the UEQ (User Experience Questionnaire) data as easy as

our respondents into the Data Worksheed. The tool then
the result. In addition several graphs are shown that you may
to your word processor of presentation tool.
rpretation of the results easier and to have the charts in your
down.

ms and the UEQ dimensions in the selected language. All other

e scale values for pragmatic and hedonic quality per person.

means and the mean and standard deviation per item are

eans and for the mean of each item are calculated.

er scale is calculated. Check here if the scales have a sufficiently
e treated carefully in the interpretation.
good the evaluated product is compared to the products in the
each scale how the results are related to the products in the
ort version available the benchmark is based on the values
re.
or inconsistent answers, i.e. answers from participants that

ent languages. Used for switching the language.

Please enter the data here!
Use the item numbers in the printed questionnaire and the categories 1 (if the
alternative on the extreme left is marked) to 7 (if the alternative on the extreme right
is marked).
Leave the cell empty if the person has not answered the item. Please do not enter a
special caracter in such cases, since this would cause errors in the calculations.
You can enter data for a maximum of 1000 participants. If you need more, you have to
adjust the fomulas in the Excel.

Items
1 2 3 4 5 6 7 8
1 6 4 5 3 4 3 5
5 5 4 5 2 3 4 5
6 7 2 5 5 6 5 4
5 6 6 4 4 6 6 4
4 4 4 5 5 5 4 5
6 4 4 5 6 6 5 5
5 6 6 6 3 4 4 3
7 6 4 5 5 3 5 4
7 7 6 5 4 6 5 5
6 6 5 5 6 5 6 6
Transformed Data
Here you can find the transformed values per item and the scale values per
participant. The +3 represent the most positive and the -3 the most negative value.

Items
1 2 3 4 5 6 7 8
-3 2 0 1 -1 0 -1 1
1 1 0 1 -2 -1 0 1
2 3 -2 1 1 2 1 0
1 2 2 0 0 2 2 0
0 0 0 1 1 1 0 1
2 0 0 1 2 2 1 1
1 2 2 2 -1 0 0 -1
3 2 0 1 1 -1 1 0
3 3 2 1 0 2 1 1
2 2 1 1 2 1 2 2
 Skale means per person
Pragmatic Quality Hedonic Quality Overall
0.00 -0.25 -0.13
0.75 -0.50 0.13
1.00 1.00 1.00
1.25 1.00 1.13
0.25 0.75 0.50
0.75 1.50 1.13
1.75 -0.50 0.63
1.50 0.25 0.88
2.25 1.00 1.63
1.50 1.75 1.63
Results
You can interpret the means of the scales pragmatic quality and hedonic quality.
Values between -0.8and 0.8 represent a neural evaluation of the corresponding scale, values > 0,8 represent a positive ev
The range of the scales is between -3 (horribly bad) and +3 (extremely good). But in real applications in general only values
a range of different persons with different opinions and answer tendencies, for example the avoidance of extreme answer
Thus, even a quite good value of +1.5 for a scale looks from the purely visual standpoint on a scale range of -3 to +3 not as

Item Mean Variance Std. Dev. No. Negative Positive

1 1.2 3.1 1.8 10 obstructive supportive
2 1.7 1.1 1.1 10 complicated easy
3 0.5 1.6 1.3 10 inefficient efficient
4 1.0 0.2 0.5 10 confusing clear
5 0.3 1.8 1.3 10 boring exciting
6 0.8 1.5 1.2 10 not interesting interesting
7 0.7 0.9 0.9 10 conventional inventive
8 0.6 0.7 0.8 10 usual leading edge

Mean value per Item

-3 -2 -1 0 1 2

8
es > 0,8 represent a positive evaluation and values < -0,8 represent a negative evaluation.
plications in general only values in a restricted range will be observed. It is due to the calculation of means over
e avoidance of extreme answer categories, extremely unlikely to observe values above +2 or below -2.
a scale range of -3 to +3 not as positive as it really is.

Scale Short UEQ Scales

Pragmatic Quality Pragmatic Quality 1.100
Pragmatic Quality Hedonic Quality 0.600
Pragmatic Quality Overall 0.850
Pragmatic Quality
Hedonic Quality
Hedonic Quality
Hedonic Quality
Hedonic Quality

2 3 1

-1

-2
ty ty all
uali uali v er
c Q c Q O
ati oni
gm d
P ra He
 all
v er
O
Confidence intervals for items and scales
Here the 5% confidence intervals for the scale means and the means of the single items are shown. The confidence interval is
The smaller the confidence interval the higher is the precision of the estimation and the more you can trust your results. The w
available data and on how consistenly the persons judged the evaluated product. The more consistent their opinion is the sma

Confidence interval (p=0.05) per item

Item Mean Std. Dev. N Confidence Confidence interval
1 1.200 1.751 10 1.085 0.115 2.285
2 1.700 1.059 10 0.657 1.043 2.357
3 0.500 1.269 10 0.787 -0.287 1.287
4 1.000 0.471 10 0.292 0.708 1.292
5 0.300 1.337 10 0.829 -0.529 1.129
6 0.800 1.229 10 0.762 0.038 1.562
7 0.700 0.949 10 0.588 0.112 1.288
8 0.600 0.843 10 0.523 0.077 1.123
tems are shown. The confidence interval is a measure for the precision of the estimation of the mean.
d the more you can trust your results. The width of the confidence interval depends on the number of
he more consistent their opinion is the smaller is the confidence interval.

Confidence intervals (p=0.05) per scale

Scale Mean Std. Dev. N Confidence Confidence interval
Pragmatic Quality 1.100 0.689 10 0.427 0.673 1.527
Hedonic Quality 0.600 0.810 10 0.502 0.098 1.102
Overall 0.850 0.580 10 0.359 0.491 1.209
Correlations of the items per scale and Cronbachs Alpha-Coefficient
Items that belong to the same scale should show in general a high correaltion. The Alpha-Coefficient (Cronbach, 1951) is a
accepted rule how big the value of the coefficient should be. Many authors assuem that a scale should show an alpha valu
an methodological standpoint such a use of a cut-off criterium is not really well-founded (see for example Schmitt, N., 199
Coefficient should be interpreted carefully.
If the value of the Alpha-Coefficient for a scale shows a massive deviation from a reasonable target value, for example 0.7,
context interpreted by several participants in an unexpected way. In such cases the corresponding scale should be interpre

Pragmatic Quality Hedonic Quality

Items Correlation Items Correlation
1.2 0.22 5.6 0.51
1.3 0.10 5.7 0.60
1.4 0.00 5.8 0.32
2.3 0.12 6.7 0.51
2.4 0.00 6.8 0.13
3.4 0.00 7.8 0.11
Average 0.07 Average 0.36
Alpha 0.24 Alpha 0.70
fficient
efficient (Cronbach, 1951) is a measure for the consistence of a scale. There is no generally
ale should show an alpha value > 0.7 to be considered as sufficiently consistent. However, from
e for example Schmitt, N., 1996). Especially if you have only a small sample the value of the Alpha-

target value, for example 0.7, this can be a hint that some items of the scale are in the given
onding scale should be interpreted very carefully.
Benchmark
The measured scale means are set in relation to existing values from a benchmark data set. This data set contains data fro
software, web pages, web shops, social networks). Currently these benchmark data are based on the full UEQ, since we do
published in 2017, so it is pretty new in the moment). The scale values of the short version are a reasonable good approxim
approximation possible to use the data from the full UEQ benchmark also as a benchmark for the short UEQ. However, the
only a first rough estimation. When enough data with the short UEQ are available we will replace this benchmark with a sp
The comparison of the results for the evaluated product with the data in the benchmark allows conclusions about the rela
Two versions of the benchmark chart are shown. The first one shows only the scale scores in relation to the benchmark ca
scores.

Please help to increase the data basis for the benchmark! If you use the short UEQ to evaluate products it would be quite
number of participants in your study and the scale means. We will of course handle these information absolutely confiden

Scale Mean Comparisson to benchmark

Pragmatic Quality 1.1 Below average
Hedonic Quality 0.6 Below Average
Overall 0.85 Below Average

2.50

2.00

1.50
Good
1.00 Above Average
Below Average
0.50 Bad
Lower Border
0.00 Mean

-0.50

-1.00
Pragmatic Quality Hedonic Quality Overall

Table to create the benchmark graph (purely technical, please ignore)

Scale Lower Border Bad
Pragmatic Quality -1.00 0.72
Hedonic Quality -1.00 0.35
Overall -1.00 0.59

Benchmark borders (purely technical, please ignore)

Scale 25% 50%
Pragmatic Q. 0.72 1.17
Hedonic Q. 0.35 0.85
Overall 0.59 0.98
 m a benchmark data set. This data set contains data from 21175 persons from 468 studies concerning different products (business
benchmark data are based on the full UEQ, since we do not have enough data for the short version available (the short version was
ues of the short version are a reasonable good approximation of the corresponding values of the full version, thus it is as a rough
ark also as a benchmark for the short UEQ. However, these are still to different questionnaires, so interpret these data with care. They
Q are available we will replace this benchmark with a special benchmark for the short version.
ata in the benchmark allows conclusions about the relative quality of the evaluated product compared to other products.
ws only the scale scores in relation to the benchmark categories, the second one shows in addition the confidence intervals of the scale

se the short UEQ to evaluate products it would be quite helpful for us if you will share information concerning the type of product, the
of course handle these information absolutely confidential and will use it solely to improve the benchmark.

Interpretation
50% of results better, 25% of results worse
50% of results better, 25% of results worse
50% of results better, 25% of results worse
2.50

2.00

1.50

Good
1.00
Above Average
Below Average
Bad 0.50
Lower Border
Mean
0.00

-0.50
Overall
-1.00
Pragmatic Quality Hedonic Quality Overall
ore)
Below Average Above Average Good Excellent
0.45 0.38 0.19 0.76
0.5 0.35 0.39 0.91
0.39 0.33 0.27 0.92

75% 90%
1.55 1.74
1.2 1.59
1.31 1.58
nt products (business
(the short version was
thus it is as a rough
ese data with care. They are

er products.
ence intervals of the scale

g the type of product, the

Good
Above Average
Below Average
Bad
Lower Border
Mean

Overall

Mean
1.1
0.6
0.85
Detect Suspicious Data
Especially if the short version of the UEQ is applied as an Online-Questionnaire not all
participants will answer all items seriously. To detect such more or less random or not
serious answers a simple heuristic is used. All items in a scale should measure a similar
quality aspect. The idea to detect random or not serious answers is to check how
much the best and worst evaluation of an item in a scale differ. If there is a big
difference (>3) this is seen as an indicator for a problematic data pattern. Of course
such situations can also result from random response errors or a misunderstanding of
an item. Thus, it makes no sense to consider a response as problematic if this occurs
just for a single scale. But if this is true for both scales this is of course a clear hint that
the response is somehow suspicious.
The table below shows the responses coded from -3 to 3. The table next to it shows
for each scale if there is a difference >3 between the best and worst item. The Critical?
column shows for how many scales this is the case.

Items
1 2 3 4 5 6 7 8
-3 2 0 1 -1 0 -1 1
1 1 0 1 -2 -1 0 1
2 3 -2 1 1 2 1 0
1 2 2 0 0 2 2 0
0 0 0 1 1 1 0 1
2 0 0 1 2 2 1 1
1 2 2 2 -1 0 0 -1
3 2 0 1 1 -1 1 0
3 3 2 1 0 2 1 1
2 2 1 1 2 1 2 2
 Scales with inconsistent answers
Pragmatic Quality Hedonic Quality Critical?
1 1
0
1 1
0
0
0
0
0
0
0
Language L6 R6 L7 R7
German langweilig spannend uninteressant interessant
English boring exciting not interesting interesting
Spanish aburrido emocionante no interesante interesante
Portuguese Aborrecido Excitante Desinteressante Interessante
Turkish sıkıcı heyecan verici ilginç olmayan ilginç
French Ennuyeux Captivant Inintéressant Intéressant
Italian noioso appassionante non interessante interessante
Russian скучный захватывающий неинтересный интересный
Chinese 乏味的 带劲的 无趣的 有趣的
Japanese 退屈だ エキサイティングおもしろくない おもしろい
Indonesian membosankan mengasyikkan tidak menarik menarik
Dutch vervelend spannend oninteressant interessant
Estonian igav põnev ebahuvitav huvitav
Slovene dolgočasen napet nezanimiv zanimiv
Swedish Tråkig Spännande Ointressant Intressant
Greek βαρετό συναρπαστικό αδιάφορο ενδιαφέρον
Polish nudny ekscytujący nieinteresujący interesujący
Hindi उबाऊ दिलचस्प दिलचस्प नहीं दिलचस्प
Bulgarian скучен вълнуващ безинтересен интересен
Czech nudný vzrušující nezajímavý zajímavý
Malay membosankan menyeronokkan tidak menarik minamenarik minat
Persian ‫خسته کننده‬ ‫جذاب‬ ‫عدم جذابیت‬ ‫جذاب‬
Thai น่าเบื่อ น่าตื่นเตน้ ไมน่ ่ าสนใจ น่าสนใจ
Danish kedelige spændende uinteressante interessante
Norwegian kjedelig spennende uinteressant interessant
Finnish pitkäveteistä jännittävää yksitoikkoista kiinnostavaa
Hungarian unalmas izgalmas nem érdekes érdekes
Croatian dosadan uzbudljiv nezanimljiv zanimljiv
Bosnian dosadni uzbudljivi nezanimljivi zanimljivi
Hebrew ‫משעממים‬ ‫מלהיבים‬ ‫לא מעניינים‬ ‫מעניינים‬
Arabic ‫ممل‬ ‫مثير‬ ‫غير مثير لالهتمام‬ ‫مثير لالهتمام‬
Bengali একঘেয়ে উত্তেজনাপূর্ণ আকর্ষণীয়নয় মজাদার
Kannada ನೀರಸವಾದ ನೀರಸವಾದ ಕುತೂಹಲವಲ್ಲದ ಕುತೂಹಲವಲ್ಲದ
Marathi कं टाळवाणा बाहेर पडत आहे मनोरंजक नाही मनोरंजक
Tamil சலிப்பான ஆர்வமான சுவாரஸ்யம் சுவாரஸ்யம்
Slovak nudný vzrušujúci nezaujímavý zaujímavý
L10 R10 L11 R11 L13
originell konventionell behindernd unterstützend kompliziert
inventive conventional obstructive supportive complicated
original convencional obstructivo impulsor de apoyo complicado
Original Convencional Obstrutivo Condutor Complicado
özgün geleneksel engelleyici destekleyici karmaşık
Original Conventionnel Rigide Facilitant Compliqué
originale convenzionale ostruttiva di supporto complicato
изобретательный консервативный мешающий помогающий сложный
独创的 常规的 碍手碍脚的 能提供辅助的 复杂的
独特だ 従来どおり 妨げになる 助けられる 複雑
berdaya cipta konvensional menghalangi mendukung rumit
origineel conventioneel belemmerend ondersteunend complex
leidlik konservatiivne takistav abistav keeruline
nov star se ne da upravljati se z lahkoto upravljkompliciran
Uppfinningsrik Fantasilös Hindrande Stödjande Komplicerad
εφευρετικό συμβατικό παρελκυστικό υποστηρικτικό περίπλοκο
pomysłowy konwencjonalny przeszkadzający wspierający skomplikowany
आविष्कारशील परम्परागत प्रतिरोधी सहायक जटिल
идеен конвенционален възпрепятстващ помагащ сложен
vynalézavý obvyklý bránící podporující složitý
terdapat rekaan ba berdasarkan kebiasmenghalang menyokong menyulitkan
‫مبتکر‬/‫نوآوری‬ ‫سنتی‬/‫مرسوم‬ ‫مختل کننده‬ ‫پشتیبان‬ ‫پیچیده‬
ชา่ งประดิษฐ์ ตามแบบแผน ขัดขวาง สนับสนุ น ซั บซ้อน
opfindsom traditionsbunden hindrende understøttende indviklede
oppfinnsom tradisjonell hindrende støttende komplisert
kekseliäs perinteinen vaikeuttavaa tukevaa monimutkaista
ötletes szokványos hátráltatók támogatók bonyolult
originalan neoriginalan ometajući podupirući kompliciran
domišljati uobičajeni ometajući od pomoći složeni
‫פורצי דרך‬ ‫שגרתיים‬ ‫מכשילות‬ ‫תומכות‬ ‫מסובכים‬
‫مستحدث‬
َ ‫تقليدي‬ ‫عائق‬ ‫داعم‬ ‫معقَّد‬
উদ্ভাবক গতানুগতিক প্রতিবন্ধক সহায়ক জটিল
ಹೊಸದಾಗಿ ಕಂಡುಹಿಡಿ ಹೊಸದಾಗಿ ಕಂಡುಹಿಡಿ ಪ್ರತಿರೋಧಿಸುವ ಪ್ರತಿರೋಧಿಸುವ ಕ್ಲಿಷ್ಟಕರ
शोधक पारंपारिक अडवणूक करणारा आधार देणारा बिकट
கண்டுபிடிப் வழக்கமான தடை செய்யும் ஆதரவு சிக்கலான
moderný tradičný obmedzujúci podporujúci zložitý
R13 L15 R15 L20 R20
einfach herkömmlich neuartig ineffizient effizient
easy usual leading edge inefficient efficient
facil convencional novedoso ineficiente eficiente
Fácil Comum Vanguardista Ineficiente Eficiente
sade alışıldık eşi görülmedik verimsiz verimli
Simple Habituel Avant-gardiste Inefficace Efficace
facile usuale moderno inefficiente efficiente
простой обыденный передовой неэффективный эффективный
简单的 传统的 新颖的 低效的 高效的
簡単 普通 斬新的 効率が悪い 効率が良い
sederhana lazim terdepan tidak efisien efisien
eenvoudig gebruikelijk nieuw inefficiënt efficiënt
lihtne tavaline uuenduslik ebaefektiivne efektiivne
enostaven zastarel modern ni učinkovit učinkovit
Enkel Bakåtsträvande I framkant Ineffektiv Effektiv
εύκολο συνηθισμένο πρωτοπόρο ανεπαρκές επαρκές
łatwy zwykły skrajny nieefektywny efektywny
आसान हमेशा की तरह अग्रणी धार अक्षम कु शल
лесен обикновен иновативен Мотлив Експедитивен
jednoduchý tradiční moderní neefektivní efektivní
mudah biasa mempunyai kelebihtidak berkesan berkesan
‫آسان‬ ‫معمولی‬ ‫پیشرو در فناوری‬ ‫عدم کفایت‬ ‫کافی‬
เรี ยบงา่ ย ดูทั่วไป ้
ดูชันนา ไร้ประสิทธิภาพ มีประสิทธิภาพ
nemme sædvanlig avanceret ineffektivt ydedygtigt
lett vanlig ledende ineffektivt effektivt
helppoa tavanomainen huippulaatuinen tehoton tehokas
könnyű átlagos élvonalban lévő nem hatékony hatékony
jednostavan uobičajen nov neučinkovit učinkovit
jednostavni svakodnevni vrhunski neefikasan efikasan
‫קלים‬ ‫בחזית החדשנות רגילים‬ ‫לא יעיל‬ ‫יעיל‬
‫بسيط‬ ‫اعتيادي‬ ‫جديد من نوعه‬ ‫فعال‬
َّ ‫غير‬ ‫فعال‬
َّ
সহজ সাধারণ নেতৃ স্থানীয় অকার্যকর কার্যকর
ಕ್ಲಿಷ್ಟಕರ ಸಾಮಾನ್ಯ ಸಾಮಾನ್ಯ ದಕ್ಷವಲ್ಲದ ದಕ್ಷವಲ್ಲದ
सोपे नेहमीच्या पुढचे टोक अकार्यक्षम कार्यक्षम
சுலபம் சாதாரணமான முன்னணி விளதிறனற்ற திறன்
jednoduchý bežný špičkový neefektívny efektívny
L21 R21 Pragmatic Quality Hedonic Quality
übersichtlich verwirrend Pragmatische Qualität Hedonische Qualität
clear confusing Pragmatic Quality Hedonic Quality
claro confuso Calidad Pragmática Calidad Hedónico
Evidente Confuso Qualidade Pragmática Qualidade Hedônica
açık kafa karıştırıcı Pragmatik Kalite Hedonik Kalite
Clair Déroutant Qualité Pragmatique Qualité Hédonique
chiaro confuso Qualità Pragmatico Qualità eEonica
наглядный запутанный Прагматические качесгедонистическоe качествo
一目了然的 令人眼花缭乱的 务实的品质 享乐质量
すっきりしているごちゃごちゃして実用的な品質 ヘドニック品質
jelas membingungkan Kualitas Pragmatis Kualitas Hedonis
overzichtelijk verwarrend Pragmatische KwaliteitHedonistische Kwaliteit
selge ebamäärane Pragmaatiline kvaliteetHedooniline kvaliteet (hinnang meeldivuse järgi)
pregleden ustvarja zmedo pragmatično kakovostihedonsko kakovosti
Tydlig Förvirrande Pragmatiska kvaliteter Hedoniska kvaliteter
σαφές μπερδεμένο ρεαλιστική ποιότητα ηδονική ποιότητας
przejrzysty zagmatwany Pragmatyczne jakości Hedoniczna jakości
स्पष्ट भ्रमित व्यावहारिक गुणवत्ता सुखात्मक गुणवत्ता
ясен объркващ Прагматично качествоХедонично качество
jasný matoucí Pragmatická kvalita Hédonická kvalita
jelas mengelirukan Kualiti yang berguna Kualiti Kesenangan
‫واضح‬ ‫مبهم‬/‫گیج کننده‬ ‫کیفیت استفاده‬ ‫کیفیت طراحی‬
ชั ดเจน สั บสน คุณภาพในทางปฏิบัติ คุณภาพความเพลิดเพลิน
tydelige forvirrende Pragmatic Quality Hedonic Quality
tydelig forvirrende Pragmatic Quality Hedonic Quality
selkeätä sekavaa Pragmatic Quality Hedonic Quality
világos nehezen értelmezhPragmatic Quality Hedonic Quality
jasan zbunjujuć Pragmatična kvaliteta Hedonička kvaliteta
jasni zbunjujući Pragmatic Quality Hedonic Quality
‫ברורים‬ ‫ מבלבלים‬Pragmatic Quality Hedonic Quality
‫واضح‬ ‫ ُمربك‬Pragmatic Quality Hedonic Quality
স্পষ্ট বিভ্রান্তিকর Pragmatic Quality Hedonic Quality
ಸ್ಪಷ್ಟತೆ ಸ್ಪಷ್ಟತೆ Pragmatic Quality Hedonic Quality
स्पष्ट गोंधळात टाकणारे Pragmatic Quality Hedonic Quality
தெளிவு குழப்பமான Pragmatic Quality Hedonic Quality
jasný mätúci Pragmatická kvalita Hedonická kvalita
eeldivuse järgi)
Sample size to reach a certain precision concerning the estimated scale means
If you have not many data points the error bars for the scale means in the Results Tab can be quite big. The error bars show
true value of the scale mean should be located with a 95% probability. Thus, big error bars stand for a not very precise me
A natural question is thus to ask, how many person you need (i.e. how big your sample should be) if you want to measure
tab tries to answer this question.
For a given precision E (deviation between true scale mean in the population and the estimated scale mean from the samp
standard deviation of the scale value in the population (the smaller this is the higher is the agreement between different p
ask) and the chance (error probability) P you still accept that the true scale mean is outside the area described by the erro
the confidence interval).
Of course the standard deviation of the scale value in the population is not known. For the data in this tab we estimate thi
already entered data. Thus, if you enter more data the estimation of this value and thus the values in the table below will c
The following table shows for typical values of E and P the required sample size. Please note that these are just rough estim
you need to achieve an accurate measurement!

Scale Standard Deviation

Pragmatic Quality 0.69
Hedonic Quality 0.81

Condition Pragmatic Quality Hedonic Quality

Precision=0.5, Err.Prob.=0.1 5 7
Precision=0.5, Err.Prob.=0.05 7 10
Precision=0.5, Err.Prob.=0.01 17 17
Precision=0.25, Err.Prob.=0.1 21 29
Precision=0.25, Err.Prob.=0.05 29 40
Precision=0.25, Err.Prob.=0.01 51 70
Precision=0.1, Err.Prob.=0.1 129 178
Precision=0.1, Err.Prob.=0.05 182 252
Precision=0.1, Err.Prob.=0.01 316 436
ated scale means
be quite big. The error bars show the area (confidence interval) in which the
stand for a not very precise measurement.
uld be) if you want to measure the scale means with a certain precision. This

ated scale mean from the sample) the required sample size depends on the
agreement between different persons and thus the less persons you need to
the area described by the error bars (i.e. it will be with a probability 1-P inside

data in this tab we estimate this value from the standard deviation of the
e values in the table below will change!!
e that these are just rough estimations to give you an idea on how many data