Muutke küpsiste eelistusi

E-raamat: Doing Better Statistics in Human-Computer Interaction

(University of York)
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 07-Feb-2019
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781108665223
  • Formaat - EPUB+DRM
  • Hind: 38,27 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
  • Formaat: EPUB+DRM
  • Ilmumisaeg: 07-Feb-2019
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781108665223

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

Written for human-computer interaction (HCI) researchers - whether undergraduates, professors, or UX professionals who need to analyse quantitative data - this book helps to improve readers' knowledge of the modern best practice in statistics and their understanding of how to do statistical analysis on their own data.

Each chapter of this book covers specific topics in statistical analysis, such as robust alternatives to t-tests or how to develop a questionnaire. They also address particular questions on these topics, which are commonly asked by human-computer interaction (HCI) researchers when planning or completing the analysis of their data. The book presents the current best practice in statistics, drawing on the state-of-the-art literature that is rarely presented in HCI. This is achieved by providing strong arguments that support good statistical analysis without relying on mathematical explanations. It additionally offers some philosophical underpinnings for statistics, so that readers can see how statistics fit with experimental design and the fundamental goal of discovering new HCI knowledge.

Arvustused

'If you, and your experiments, have been bruised by statistical misfortune, then this is the book for you. Paul Cairns' wise and pragmatic advice talks us through the practical use of statistics in Human-Computer Interaction, showing his own bruises when necessary. This should become the standard reference that the field needs.' Alan Blackwell, University of Cambridge 'In Human-Computer Interaction, we gather data from experiment designs that are often more complex or messy than those presented as examples in a basic textbook on statistics. Cairns presents digestible information for an interdisciplinary audience with expertise and authority. I will be buying a copy of this book for my students, and also one for myself!' Regan Mandryk, University of Saskatchewan, Canada 'This is a must-read for novice or well-established researchers alike, who are worried about whether they are conducting the correct statistical analyses of their data. Paul Cairns makes learning about statistics seem both fun and interesting. I'm confident that this book will positively impact the quality of future Human-Computer Interaction research.' Anna L. Cox, University College London Interaction Centre

Muu info

This book addresses common questions from HCI researchers when trying to do statistical analysis on their data.
List of Figures page
xi
List of Tables
xiv
Acknowledgements xv
Getting Started 1(8)
PART I WHY WE USE STATISTICS
9(60)
1 How Statistics Support Science
11(14)
1.1 The Problem of Induction
12(3)
1.2 Severe Testing
15(2)
1.3 Evidence in HCI
17(1)
1.4 New Experimentalism in HCI
18(3)
1.5 Big Data
21(1)
1.6 Conclusions
22(3)
2 Testing the Null
25(13)
2.1 The Basics of NHST
26(4)
2.2 Going beyond p-Values
30(3)
2.3 NHST and Severe Testing
33(2)
2.4 Honesty in Statistics
35(3)
3 Constraining Bayes
38(15)
3.1 Defining Probability
40(2)
3.2 Plausibility
42(2)
3.3 Unconstrained Bayes
44(3)
3.4 The Bayesian Critique of Frequentism
47(1)
3.5 Being Careful: A Response to the Critique
48(2)
3.6 So, Frequentist or Bayesian?
50(3)
4 Effects: What Tests Test
53(16)
4.1 Location
55(3)
4.2 Dominance
58(1)
4.3 Variation
59(1)
4.4 Estimation and Significance
60(2)
4.5 Big, Small and Zero Effects
62(2)
4.6 Choosing Tests to See Effects
64(5)
PART II HOW TO USE STATISTICS
69(162)
5 Planning Your Statistical Analysis
71(9)
5.1 Principle 1: Articulation
73(2)
5.2 Principle 2: Simplicity
75(2)
5.3 Principle 3: Honesty
77(2)
5.4 Conclusions
79(1)
6 A Cautionary Tail: Why You Should Not Do a One-Tailed Test
80(6)
6.1 A Tale of Two Tails
81(1)
6.2 One-Tail Bad, Two-Tails Better
82(4)
7 Is This Normal?
86(9)
7.1 What Makes Data Normal?
86(4)
7.2 The Problems of Non-normal Data
90(1)
7.3 Testing for Normality
91(2)
7.4 Implications
93(2)
8 Sorting Out Outliers
95(9)
8.1 Detecting Outliers
96(2)
8.2 Sources and Remedies for Outliers
98(4)
8.2.1 Errors in Data
99(1)
8.2.2 Mischievous Participants
99(1)
8.2.3 Faulty Study Design
100(1)
8.2.4 Natural Variation
101(1)
8.3 Conclusions
102(2)
9 Power and Two Types of Error
104(10)
9.1 Type I and Type II Errors
105(1)
9.2 Defining Power
106(2)
9.3 Power and Sample Sizes
108(2)
9.4 Power and the Quality of Tests
110(2)
9.5 Summary
112(2)
10 Using Non-Parametric Tests
114(11)
10.1 The Mechanics of Ranks
115(2)
10.2 Analysing Errors
117(3)
10.2.1 Type I Errors
117(2)
10.2.2 Type II Errors
119(1)
10.3 Practical Use
120(2)
10.4 Reporting Non-Parametric Tests
122(1)
10.5 Summary
123(2)
11 A Robust t-Test
125(14)
11.1 A Traditional t-Test
126(4)
11.2 Simple Solutions?
130(1)
11.3 Location, Location, Location
131(1)
11.4 Trimmed and Winsorized Means
132(2)
11.5 M-Estimators
134(1)
11.6 Back to t-Tests
135(1)
11.7 Overall Advice
136(3)
12 The ANOVA Family and Friends
139(16)
12.1 What ANOVA Does
140(5)
12.2 Is ANOVA Robust?
145(2)
12.3 Robust Alternatives to ANOVA
147(4)
12.3.1 Non-Parametric Alternatives
147(2)
12.3.2 Changes of Location
149(1)
12.3.3 Do Something Else
150(1)
12.4 Summary
151(4)
13 Exploring, Over-Testing and Fishing
155(12)
13.1 Exploring After a Severe Test
156(1)
13.2 Exploratory Studies
157(2)
13.3 Over-Testing
159(3)
13.3.1 ANOVA Can (Sometimes) Help
160(1)
13.3.2 Planned Comparisons
161(1)
13.3.3 The Bonferroni Correction
161(1)
13.3.4 Bayesian Methods Can Over-Test Too
162(1)
13.4 Fishing
162(2)
13.5 Some Rules of Exploration
164(3)
14 When Is a Correlation Not a Correlation?
167(8)
14.1 Defining Correlation
168(2)
14.2 Outlying Points
170(1)
14.3 Clusters
171(2)
14.4 Avoiding Problems
173(1)
14.5 A Final Warning
174(1)
15 What Makes a Good Likert Item?
175(11)
15.1 Some Important Context
176(2)
15.2 Should Items Have a Midpoint?
178(2)
15.3 How Many Options?
180(1)
15.4 Label All Options or Just End-Points?
181(1)
15.5 The Final Story?
182(4)
16 The Meaning of Factors
186(18)
16.1 From Concepts to Items
188(2)
16.2 From Items to Factors
190(8)
16.2.1 The Methods of Factor Analysis
192(2)
16.2.2 Finding Factors
194(4)
16.3 From Factors to Concepts?
198(2)
16.4 What Does It Mean?
200(4)
17 Unreliable Reliability: The Problem of Cronbach's Alpha
204(1)
17.1 Reliability and Validity
205(1)
17.2 A Simple Model
206(2)
17.3 When α Is Low
208(1)
17.4 When α Is Too High
209(2)
17.5 Beyond α
211(3)
18 Tests for Questionnaires
214(1)
18.1 Testing Likert Items
215(1)
18.1.1 Type I Analysis
216(2)
18.1.2 Power Analysis
218(4)
18.1.3 Which Test for Liken Items?
222(1)
18.2 Questionnaire Data
222(2)
18.2.1 Type I Analysis
224(1)
18.2.2 Power Analysis
225(4)
18.2.3 Which Test for Questionnaires?
229(1)
18.3 One Final Observation
229(2)
Index 231
Paul Cairns is a reader in Human-Computer Interaction at the University of York and Scholar-in-Residence for The AbleGamers Charity that helps people with disabilities combat social isolation by making videogames more accessible. He has taught statistics at all levels of education for nearly twenty years. His particular research interest is in players' experiences of digital games, and his expertise in experimental and statistical methods was developed through working in this area.