<%BANNER%>

Paramaters of Lighting Quality on Subjective Perceptions

Permanent Link: http://ufdc.ufl.edu/UFE0024659/00001

Material Information

Title: Paramaters of Lighting Quality on Subjective Perceptions
Physical Description: 1 online resource (120 p.)
Language: english
Creator: Gillar, Kimberly
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: flynn, lighting, parameters, perception
Architecture -- Dissertations, Academic -- UF
Genre: Architecture thesis, M.Arch.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PARAMETERS OF LIGHTING QUALITY ON SUBJECTIVE PERCEPTIONS Architecture cannot be viewed or appreciated without light and it dramatically affects one's impression of a room, as well as the efficiency of the way the space is used. Studies have illustrated light's influences on a space's perceptual, behavioral, and preferential features. Factors have been proposed and tested by creating types of spatial illumination by the location of light and dark, the articulation of light through surface, the intensity of light, and the color tone. In the 1970s, John Flynn published a study on psychological factors of lighting. This study attempted to reveal the psychological impression of factors involved with task lighting in a workspace with computer displays. All tests were done in a full-scale room, in which participants rated a series of lighting designs. Four subsequent factors were considered in the study. The first factor evaluated the effects for three different lighting conditions: fluorescent, halogen, and daylight. The second factor evaluated the effect of the subjects' eyesight, that is, whether they were wearing contacts, glasses, or neither. The third factor would be to evaluate the effect of the different light levels in the individual lighting conditions. The fourth factor would be incorporating an additional test to compare the results of a full-scale room with the results of viewing a three-dimensional rendering of the artificial lighting design. The purpose of this experiment was to evaluate if an image could achieve similar results of the actual experience in the full-scale lighting design. The design incorporated parameter needs for certain task lighting to be efficient and to be compared to Flynn's results. It also created a positive effect on one's perception of a space and made the occupant more comfortable in dealing with the best lighting situation for certain tasks.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kimberly Gillar.
Thesis: Thesis (M.Arch.)--University of Florida, 2009.
Local: Adviser: Gold, Martin A.
Local: Co-adviser: Kaye, Stanley.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024659:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024659/00001

Material Information

Title: Paramaters of Lighting Quality on Subjective Perceptions
Physical Description: 1 online resource (120 p.)
Language: english
Creator: Gillar, Kimberly
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: flynn, lighting, parameters, perception
Architecture -- Dissertations, Academic -- UF
Genre: Architecture thesis, M.Arch.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: PARAMETERS OF LIGHTING QUALITY ON SUBJECTIVE PERCEPTIONS Architecture cannot be viewed or appreciated without light and it dramatically affects one's impression of a room, as well as the efficiency of the way the space is used. Studies have illustrated light's influences on a space's perceptual, behavioral, and preferential features. Factors have been proposed and tested by creating types of spatial illumination by the location of light and dark, the articulation of light through surface, the intensity of light, and the color tone. In the 1970s, John Flynn published a study on psychological factors of lighting. This study attempted to reveal the psychological impression of factors involved with task lighting in a workspace with computer displays. All tests were done in a full-scale room, in which participants rated a series of lighting designs. Four subsequent factors were considered in the study. The first factor evaluated the effects for three different lighting conditions: fluorescent, halogen, and daylight. The second factor evaluated the effect of the subjects' eyesight, that is, whether they were wearing contacts, glasses, or neither. The third factor would be to evaluate the effect of the different light levels in the individual lighting conditions. The fourth factor would be incorporating an additional test to compare the results of a full-scale room with the results of viewing a three-dimensional rendering of the artificial lighting design. The purpose of this experiment was to evaluate if an image could achieve similar results of the actual experience in the full-scale lighting design. The design incorporated parameter needs for certain task lighting to be efficient and to be compared to Flynn's results. It also created a positive effect on one's perception of a space and made the occupant more comfortable in dealing with the best lighting situation for certain tasks.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Kimberly Gillar.
Thesis: Thesis (M.Arch.)--University of Florida, 2009.
Local: Adviser: Gold, Martin A.
Local: Co-adviser: Kaye, Stanley.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024659:00001


This item has the following downloads:


Full Text

PAGE 1

PARAMETERS OF LIGHTING QUALIT Y ON SUBJECTIVE PERCEPTIONS By KIMBERLY GILLAR A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARCHITECTURE UNIVERSITY OF FLORIDA 2009 1

PAGE 2

2009 Kimberly Gillar 2

PAGE 3

To my parents 3

PAGE 4

ACKNOWLEDGMENTS I thank my parents and grandparents for all the love and support they have given me throughout my years at the Universi ty of Florida. I thank my chairs, Martin Gold and Stanley Kaye, for their guidance and input in making my thesis a succe ss. I also thank Nam-Kyu Park and Margaret Portillo for their additiona l insight and interest in my research. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................8LIST OF FIGURES .......................................................................................................................10ABSTRACT ...................................................................................................................... .............14 CHAPTER 1 INTRODUCTION ................................................................................................................ ..16Background .................................................................................................................... .........17Objectives .................................................................................................................... ...........18Statement of Hypothesis .........................................................................................................19Overview ...................................................................................................................... ...........192 LITERATURE REVIEW .......................................................................................................21The Effects of Light on Huma n Judgment and Behavior .......................................................21Theoretical Context .........................................................................................................21Data and Analysis ............................................................................................................22Conference Room Study ..................................................................................................23Multiple Ro om Study ......................................................................................................23Auditorium Room Study .................................................................................................24Lighting Laboratory Study ..............................................................................................25Scaling of Slides Study ....................................................................................................27Results of Impressions .....................................................................................................27Visual clarity ............................................................................................................27Public vs. private space ............................................................................................28Relaxing vs. tense space ..................................................................................................28Spatial complexity ....................................................................................................29Spaciousness .............................................................................................................29Preference .................................................................................................................29Glare ......................................................................................................................... 30Psychological Processes Infl uencing Lighting Quality ..........................................................30Influence of Lighting Quality on Presence Task Performance in Virtual Environments .......35Cognitive and Perceptual Factors in Lighted Architectural Environments ............................36Methods ...........................................................................................................................37Rating .......................................................................................................................3 7Renderings ................................................................................................................38Results ......................................................................................................................39Relating to Flynns Work ................................................................................................40 5

PAGE 6

3 METHODOLOGY ................................................................................................................. 41Overview ...................................................................................................................... ...........41Research Design .....................................................................................................................42Independent Variables .....................................................................................................42Architectural context ................................................................................................42Lighting situations ....................................................................................................43Rendered situations ..................................................................................................44Dependent Variables .......................................................................................................46Rating scales .............................................................................................................46Descriptive information ............................................................................................46Subjects ...................................................................................................................... .............46Procedure ..................................................................................................................... ...........47Data Collection .......................................................................................................................47Limitations ................................................................................................................... ...........47Data Analysis Method ............................................................................................................484 RESULTS ..................................................................................................................... ..........50Factor One: Analyzing Data of the Three Lighting Conditions .............................................50Perceptual Clarity of Task Analysis ................................................................................50Evaluative Analysis .........................................................................................................51Spaciousness and Complexity Analysis ..........................................................................53Factor Two: Analyzing Data of the Corrective Lenses vs. None ...........................................55Fluorescent ................................................................................................................... ...56Perceptual clarity of task analysis ............................................................................56Evaluative analysis ...................................................................................................57Spaciousness and complexity analysis .....................................................................59Halogen ....................................................................................................................... .....61Perceptual clarity of task analysis ............................................................................61Evaluative analysis ...................................................................................................63Spaciousness and complexity analysis .....................................................................64Daylight ...........................................................................................................................66Perceptual clarity of task analysis ............................................................................66Evaluative analysis ...................................................................................................68Spaciousness and complexity analysis .....................................................................70Factor Three: Analyzing Data of the Different Light Levels .................................................72Halogen ....................................................................................................................... .....72Perceptual clarity of task analysis ............................................................................72Evaluative analysis ...................................................................................................74Spaciousness and complexity analysis .....................................................................75Daylight ...........................................................................................................................77Perceptual clarity of task analysis ............................................................................77Evaluative analysis ...................................................................................................79Spaciousness and complexity analysis .....................................................................80Factor Four: Analyzing the Data of Real vs. Rendered Environment ....................................82Fluorescent ................................................................................................................... ...82 6

PAGE 7

Perceptual clarity of task analysis ............................................................................82Evaluative analysis ...................................................................................................84Spaciousness and complexity analysis .....................................................................86Halogen ....................................................................................................................... .....88Perceptual clarity of task analysis ............................................................................88Evaluative analysis ...................................................................................................90Spaciousness and complexity analysis .....................................................................925 SUMMARY AND CONCLUSIONS .....................................................................................95Conclusions Compared to Flynns Impressions .....................................................................95Three Lighting Conditions ..................................................................................................... .97Corrective Lenses vs. None ....................................................................................................97Different Light Levels ............................................................................................................98Real vs. Rendered Environments ............................................................................................98Recommendations ............................................................................................................... ....99 APPENDIX A INSTRUCTIONS TO SUBJECTS .......................................................................................101B EXPERIMENT PROTOCOL ...............................................................................................104C SAMPLE OF RATING SCALE SURVEY ..........................................................................106D PARTICIPANT CONSENT FORM .....................................................................................108E STATISTICAL DATA .........................................................................................................110LIST OF REFERENCES .............................................................................................................119BIOGRAPHICAL SKETCH .......................................................................................................120 7

PAGE 8

LIST OF TABLES Table page 3-1 List of footcandle levels for each seat during the study. ....................................................434-1 Standard deviations of the three lighting conditions for perceptual clarity of task. ..........514-2 Standard deviations of the three lighting conditions for evaluative. .................................524-3 Standard deviations of the three lighting conditions for spaciousness and complexity. ...544-4 Standard deviations for fluorescent ligh ting, whether subjects are wearing corrective lenses or not for perceptual clarity of task. ........................................................................564-5 Standard deviations for fluorescent li ghting, whether the subjects were wearing corrective lenses or not for evaluative. ..............................................................................584-6 Standard deviations for fluorescent li ghting, whether the subjects were wearing corrective lenses or not for spaciousness and complexity. ................................................604-7 Standard deviations for halogen lighti ng, whether subjects ar e wearing corrective lenses or not for perceptual clarity of task. ........................................................................624-8 Standard deviations for halogen ligh ting, whether the subjects were wearing corrective lenses or not for evaluative. ..............................................................................634-9 Standard deviations for halogen lighti ng, whether subjects ar e wearing corrective lenses or not for spaciousness and complexity. .................................................................654-10 Standard deviations for daylighting, wh ether subjects are wear ing corrective lenses or not for perceptual clarity of task. ...................................................................................674-11 Standard deviations for daylighting, wh ether subjects are wear ing corrective lenses or not for evaluative. ..........................................................................................................694-12 Standard deviations for daylighting, wh ether subjects are wear ing corrective lenses or not for spaciousness and complexity. ............................................................................714-13 Standard deviations for diffe rent light levels of halogen lighting for perceptual clarity of task. ................................................................................................................................734-14 Standard deviations for di fferent light levels of hal ogen lighting for evaluative. .............744-15 Standard deviations for different light le vels of halogen lighting for spaciousness and complexity..................................................................................................................... .....764-16 Standard deviations for diffe rent light levels of dayligh ting for perceptual clarity of task. ......................................................................................................................... ...........78 8

PAGE 9

4-17 Standard deviations for different light levels of daylighting for evaluative. .....................794-18 Standard deviations for different light levels of dayligh ting for spaciousness and complexity..................................................................................................................... .....814-19 Standard deviations of the fluorescent real and rendered conditions for perceptual clarity of task......................................................................................................................834-20 Standard deviations of the fluorescent real and rendered conditi ons for evaluative. ........854-21 Standard deviations of the fluorescent real and rendered conditions for spaciousness and complexity. ..................................................................................................................874-22 Standard deviations of the halogen real and rendered c onditions for perceptual clarity of task. ................................................................................................................................894-23 Standard deviations of the halogen real and rendered conditi ons for evaluative. .............914-24 Standard deviations of the halogen real and rendered conditi ons for spaciousness and complexity..................................................................................................................... .....935-1 Comparison of Flynns impressions. .................................................................................955-2 Comparison of results to Flynn results .............................................................................96 9

PAGE 10

LIST OF FIGURES Figure page 3-1 Plan of the study room with seat assignments. ..................................................................423-2 Plan of room with locat ion of fluorescent lighting. ...........................................................443-3 Plan of room with locat ion of halogen lighting. ................................................................443-4 Plan of room with location of source of daylight. .............................................................443-5 Three-dimensional simulation of fluorescent light setting. ...............................................453-6 Three-dimensional simulation of halogen light setting. ....................................................454-1 Perceptual clarity comparison means of fl uorescent, halogen, and daylight lighting. ......504-2 Perceptual clarity comparison confiden ce intervals of fluorescent, halogen, and daylight lighting. ................................................................................................................514-3 Evaluative comparison means of fluorescent, halogen, and daylight lighting. .................524-4 Evaluative comparison confidence interval s of fluorescent, halogen, and daylight. .........534-5 Spaciousness and complexity compar ison means of fluorescent, halogen, and daylight lighting. ................................................................................................................544-6 Spaciousness and complexity comparis on confidence interval s of fluorescent, halogen, and daylight lighting............................................................................................554-7 Perceptual clarity comparison means for fluorescent lighting, wh ether the subjects were wearing correctiv e lenses or not. ...............................................................................564-8 Perceptual clarity comparison confidence intervals for fluorescent lighting, whether the subjects were wearing co rrective lenses or not. ...........................................................574-9 Evaluative comparison means for fluores cent lighting, whether the subjects were wearing corrective lenses or not. .......................................................................................584-10 Evaluative comparison confidence inte rvals for fluorescent lighting, whether the subjects were wearing co rrective lenses or not. .................................................................594-11 Spaciousness and complexity comparison means for fluorescent lighting, whether the subjects were wearing co rrective lenses or not. .................................................................604-12 Spaciousness and complexity compar ison confidence intervals for fluorescent lighting, whether the subjects were w earing corrective lenses or not. ...............................61 10

PAGE 11

4-13 Perceptual clarity comparison means for halogen lighting, whether the subjects were wearing corrective lenses or not. .......................................................................................624-14 Perceptual clarity comparison confidence intervals for halogen lighting, whether the subjects were wearing co rrective lenses or not. .................................................................624-15 Evaluative comparison means for hal ogen lighting, whether the subjects were wearing corrective lenses or not. .......................................................................................634-16 Evaluative comparison confidence inte rvals for halogen lighting, whether the subjects were wearing co rrective lenses or not. .................................................................644-17 Spaciousness and complexity comparison means for halogen lighting, whether the subjects were wearing co rrective lenses or not. .................................................................654-18 Spaciousness and complexity comparison confidence intervals for halogen lighting, whether the subjects were wear ing corrective lenses or not. .............................................664-19 Perceptual clarity comparison means fo r daylighting, whether the subjects were wearing corrective lenses or not. .......................................................................................674-20 Perceptual clarity comparison confidence intervals for daylighting, whether the subjects were wearing co rrective lenses or not. .................................................................684-21 Evaluative comparison means for daylig hting, whether the subj ects were wearing corrective lenses or not. .....................................................................................................684-22 Evaluative comparison confidence interv als for daylighting, wh ether the subjects were wearing correctiv e lenses or not. ...............................................................................694-23 Spaciousness and complexity comparison means for daylighting, whether the subjects were wearing co rrective lenses or not. .................................................................704-24 Spaciousness and complexity comparison confidence interval s for daylighting, whether the subjects were wear ing corrective lenses or not. .............................................714-25 Perceptual clarity comparison means for di fferent light levels of halogen lighting. .........724-26 Perceptual clarity comparison confidence intervals for different light levels of halogen lighting. ................................................................................................................734-27 Evaluative comparison means for differe nt light levels of halogen lighting. ....................744-28 Evaluative comparison confidence interval s for different light levels of halogen lighting. ..................................................................................................................... .........754-29 Spaciousness and complexity comparison means for different light levels of halogen lighting. ..................................................................................................................... .........76 11

PAGE 12

4-30 Spaciousness and complexity comparison confidence intervals for different light levels of halogen lighting. ..................................................................................................774-31 Perceptual clarity comparison means for different light levels of daylighting. .................784-32 Perceptual clarity comparison confidence intervals for different light levels of daylighitng. .................................................................................................................. ......784-33 Evaluative comparison m eans for different light le vels of daylighting. ............................794-34 Evaluative comparison confidence intervals fo r different light levels of daylighting. ......804-35 Spaciousness and complexity comparison means for different light levels of daylighting. .................................................................................................................. ......814-36 Spaciousness and complexity comparison confidence intervals for different light levels of daylighting. ..........................................................................................................824-37 Perceptual clarity of task comparis on means of fluorescent real and rendered environment ................................................................................................................... ....834-38 Perceptual clarity of task comparison c onfidence intervals of fluorescent real and rendered environment ........................................................................................................844-39 Evaluative comparison means of fluo rescent real and re ndered environment ..................854-40 Evaluative comparison confidence inte rvals of fluorescent real and rendered environment ................................................................................................................... ....864-41 Spaciousness and complexity comparison means of fluorescent real and rendered environment. .................................................................................................................. ....874-42 Spaciousness and complexity comparison confidence intervals of fluorescent real and rendered environment ..................................................................................................884-43 Perceptual clarity of task comparison means of halogen real and rendered environment ................................................................................................................... ....894-44 Perceptual clarity of task comparison confidence intervals of halogen real and rendered environment ........................................................................................................894-45 Evaluative comparison means of hal ogen real and rendered environment. .......................904-46 Evaluative comparison confidence in tervals of halogen real and rendered environment. .................................................................................................................. ....914-47 Spaciousness and complexity comparison means of halogen real and rendered environment. .................................................................................................................. ....92 12

PAGE 13

4-48 Spaciousness and complexity comparison confidence intervals of halogen real and rendered environment. .......................................................................................................93 13

PAGE 14

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Architecture PARAMETERS OF LIGHTING QUALIT Y ON SUBJECTIVE PERCEPTIONS By Kimberly Gillar August 2009 Chair: Martin Gold Cochair: Stanley Kaye Major: Architecture Architecture cannot be viewed or appreciated without light and it dramatically affects ones impression of a room, as well as the efficien cy of the way the space is used. Studies have illustrated lights influences on a spaces perceptual, behavioral, and preferential features. Factors have been proposed and tested by creating types of spatial illumi nation by the location of light and dark, the articulation of light through surf ace, the intensity of light, and the color tone. In the 1970s, John Flynn published a study on psyc hological factors of lighting. This study attempted to reveal the psychological impression of factors involved with task lighting in a workspace with computer displays. All tests were done in a full-scale room, in which participants rated a series of lighting designs. Four subsequent factors were c onsidered in the study. The first factor evaluated the effects for three differe nt lighting conditions: fl uorescent, halogen, and daylight. The second factor evalua ted the effect of the subjects eyesight, that is, whether they were wearing contacts, glasses, or neither. The third factor would be to evaluate the effect of the different light levels in th e individual lighting conditions. The fourth factor would be incorporating an additional test to compare the re sults of a full-scale room with the results of viewing a three-dimensional rend ering of the artificial lighting design. The purpose of this 14

PAGE 15

15 experiment was to evaluate if an image could achieve similar results of the actual experience in the full-scale lighting design. The design incorporated parameter needs for certa in task lighting to be efficient and to be compared to Flynns results. It also created a positive effect on ones perception of a space and made the occupant more comfortable in dealing w ith the best lighting situ ation for certain tasks.

PAGE 16

CHAPTER 1 INTRODUCTION To observe architecture, one needs illumination to see the forms. To see the way architecture is meant to be perceived, lighting needs to be addressed in a particular manner to reinforce the subjective response from the space. When designing for light, some may look at the required light levels needed in a programmed space. One could have the correct measurement of light, but some may perceive the space as too d im. The light illuminating the space needs to be designed. The psychological aspect of lighting with the structure of human impressions is still a mystery. But slowly the understanding of li ght, with the perceptual and psychological experience, as well as the knowledge of lights behavior and interaction with spaces, will aid in our understanding, to design competently for the occupants.1 Flynn profoundly stated that it was not just about the amount of light one needs. He conducted studies to create certai n parameters for lighting to create a definable effect on the occupants subjective impression of the space.2 To understand of how light can affect a space, knowledge on human perception and psychology is required. Understandi ng the functions of the eye w ith adaptation and the process of viewing will aid lighting design. Adapting to certain lighting conditions can be extreme if going into a relatively bright space, such as going from outdoors, to a relatively dark space, for example a movie theater. This transition of adapta tion needs to be taken into account. The positioning of light sources needs to be taken into consideration. High task lighting, for instance, can produce a dark surrounding to the human per ception, and it can create an uneasy feeling for the space because our peripheral vision is more se nsitive to movement than our line of sight.3 16

PAGE 17

Background In lighting design a set of parameters occurs on light levels pertaining to the program of the space, but concern has increased on the psychological aspects of the built environment. This deals with the body of work done by John Flynn and others who expanded on his work on the subjective responses in lighting design. Flynn realized that lig hting can aid in the selective process of viewing, and lighting can change the information of the visual field and studied lightings different features, such as the infl uence of behavior, mood, and task performance. With the conclusion of his studies, Flynn ma de a number of lighti ng recommendations, which became lighting guidelines of the Illuminating E ngineering Society of North America (IESNA). His parameters are basic to the patterns of light and shadow. Different variances occur in the human visual system, as well as there are differences among indivi duals, so there cannot be just one standardized measure of the visual effect.4 This study deals with Flynns theories, which encompass subjective responses of architectural lighting systems in the lighted environment, expanding on similar conclusions and the evolvi ng study of lighting design in today technology. Flynns research dealt with semantic diffe rential scaling (SDS) and multidimensional scaling (MDS), using a bi-polar rating scale wi th words relating to th e perception of lighting design. Semantic differential scaling is used in psychology as a way to characterize the experiences of certain individuals.5 This thesis deals with semantic differential scaling to document participants perceptions of lighted space. The first studies conducted by Flynn used two di fferent locations with different lighting conditions, but the physical parameters of the ro om stayed the same. He used a variety of variables: distribution of light, lo cation of luminaires, in tensity of light, and th e color tone of the light. An additional study by Flynn (1975) dealt w ith the comparisons of judgment of lighting through a two-dimensional image and judgment of lighting in three-dimensional spaces. This 17

PAGE 18

study was not researched thoroughly enough, but it demonstrated that different conclusions would be discovered if the study was solely done by ratings of li ghting through images. The Bartlett School of Architecture conti nued with the Flynns research, conducting studies in laboratories and in the field. Researchers used a wider range of luminous environments and had more subjects who participated in the project and further reinforced Flynns results. Davis (2006) evaluated participants impressions with lighting conditions that were solely digital renderings. Since the studies were c onducted using digital renderings, more lighting situations could be tested in a variety of different contexts. The renderings were quite minimal in room design: they had accurate lighting and refl ectance off the surfaces, but everything was in grayscale for proper judgment of just the lighting situation and not the room. Davis would have liked to do an extensive comparison study, as wi th Flynn, but with digita l renderings and real lighting environments. Objectives This thesis intends to create more studies relating to the methods of Flynns studies to further reinforce his results using todays tec hnologies in lighting desi gn. One goal deals with lighting cues producing specific subjective impr essions. All impressions are rated on task lighting as well. Another goal is to further test real lighting e nvironments to digital renderings. The subjective impressions produced by light ing cues are based on surveys conducted by Flynn: visual clarity, spaciousne ss, and preference. The lighting systems are different sources, fluorescent, halogen, and daylight, but in compar ing the results with hi s lighting modes, I am expecting similar conclusions. During the testing of the real e nvironment, participants took pa rt in an interactive task on the computer. Part of the survey dealt with cl arity based on Flynns ( 1975) study and to see if 18

PAGE 19

results varied now because subjects have will not just be judging on the clarity of the light in the room but judging on the clarity of th e task in the li ghting situation. In the past, Flynns studies dealt strictly wi th the impressions of lighting in a real environment and later studies were conducted digitally by other researcher s. Flynn stated that experiencing actual lighting conditions would obtain different results from those that are seen two-dimensionally, but his study to obtain that conclusion dealt with images. With todays technology in computer rendering programs, three-dimensional spaces can illustrate accurate lighting. Subjective tests have b een done solely with computer renderings, but no study has done to make a comparison with the real and rendered environments. If impressions of lighting design can be judged by computer renderings, it would be easier to conduct more studies in less time and not require a specialized room. Accurate li ght renderings can be pr esented in different contexts to further understand and enhance arch itectural lighting design. The results from these studies will hopefully further evolve from Flynns studies to aff ect todays lighting environments. Statement of Hypothesis Even with the evolution of lighting design, peop les perceptions of lighting design are still supported by Flynns results in the 1970s. Overview Chapter 2 provides a literature review of Flynns studies. The review includes Flynns overall research, his studies, and his methodol ogy with other researchers preceding studies based on Flynns work. Chapter 3 presents the me thodology used to conduct this research, data collection, and data analysis. Ch apter 4 discusses the analysis of the data collected from the surveys. Illustrations of the results with tabl es and charts are include d, as well as the findings 19

PAGE 20

and interpretations. Chapter 5 offers the conclusi ons of the research and the results, and makes recommendations for future research on subjective impressions in lighting design. 20

PAGE 21

CHAPTER 2 LITERATURE REVIEW The Effects of Light on Human Judgment and Behavior One of the bases of this study in dealing w ith the psychological aspects of lighting design came from John Flynn of the Architectural Engineering Department at Pennsylvania State University. He conducted two major studies in the 1970s; he studied the impressions that subjects receive from certain lit situations, and he studied how light can create a visual cue to imply the use of the space. He used two main research techniques: semantic differential scaling (SDS) and multidimensional scaling (MDS). Theoretical Context At the time of his study (1975), light was seen just to illuminate a task or a space for the user. Not many saw any other reasoning for certain lighting techniques. Flynn thought there was more to lighting than the visual performance it enhanced; he believed light could convey information. Light could begin to create visual cues in a space and affect the users sense of place. Flynn suggested that when the designer ch anges the lighting of the room, he changes the composition and relative strength of visual signa ls and cues, which will result in altering the impressions of meaning to the occupant.6 Psychological aspects of lighting suggest that when the designer changes lighting modes, he changes the composition of visual signals and cues, and this could alter the users impressions of the meaning for a room.6 The lighting design can affect the users orientation, comprehension, impression of activity setting, or mood. For example, corner-lighting affects the impression on the room size and shape, establishing visual limits to the room. The designer can use lighting to reinforce a public space, or use it to produce an in timate space to create a greater 21

PAGE 22

sense of privacy. Light cannot be associated onl y with the utility, task, and physiological needs of a space. Light should also deal with the user s sensitivities and socio-psychological needs. The function of lighting in a design needs to prov ide not only the light for the function of the space, but also to include aspects of the users attitude, well-being, and motivation. Flynns goal throughout his studies was to explore the ideas of lighting as a system of visual cues. He used a lighting cue theory w ith scaling his studies of several types of lighting.6 The first study was conducted at the General Elec tric Lighting Institute in Cleveland. Additional studies were performed at the I llumination Laboratory at Pennsylvania State University. These facilities provided a space that could create diffe rent spaces with lighting and not change any of the other physical parameters. His basic lighting variables included the dist ribution of light from luminaires, the location of luminaires in the rooms, the intensity of light on the horizontal activity surface, and the color tone of the light (being either warm or cool).6 To analyze his data, he used two methods: sema ntic differential scali ng to provide insight on the effect of various light settings with how the subject is feeling about the room; and multidimensional scaling, which deals with the su bjects perceptual judgm ents of the space. Using both methods, he was better able to meas ure the subjective impressions on the lighting design. Data and Analysis The initial tests were conducted in a demons tration room at the General Electric Lighting Institute in Cleveland. The advantages of th is space include a number of installed lighting arrangements that can be tested without changing the physical pa rameters of the room, and the space has access to subjects with a wide variety of backgrounds. 22

PAGE 23

Conference Room Study When the subjects entered the room, they ra ted the space not knowing that light effects were the focus of the experiment. They then ju dged the space and gave a value to all of its characteristics. After the initia l rating process, they were give n a vague explanation on the study and briefly shown the six different light settings They then rated each space after a one-minute period of adaptation of each condition. For each group, this light study was conducted in a different order because when looking at lighti ng conditions, the subject tends to compare the lighting condition shown before with the one that is currently being tested. Flynn dealt with five main factors of the analysis: evaluative, perceptu al clarity, spatial complexity, spaciousness, and formality. In reviewing the data, spatial complexity and formality were perceived as unimportant because throughout the spaces were minute change s in the ratings of the different lighting conditions. Overall, this study was conducted to illu strate the qualitative aspects of lighting that would affect the overall perceived quantity of light. Multiple Room Study The data of the initial study posed the ques tion: Would the results differ if conducting the same lighting condition but differing in the overa ll room shape? Another study was conducted at the institute testing three rooms simultaneously. The first room was the same room as the initial study, and the other two rooms varied in size an d shape. The function was set up as conference spaces. The first room was a moderate size rect angular conference room. The second room was a moderate size non-rectangular conference room The third room was a large rectangular conference room. Each of the lighting conditi ons, which Flynn called lighti ng modes, were controlled individually, usually by a dimmer sw itch so that the intensities a nd distribution were consistent. The five lighting conditions were as follows: 1) overhead fluorescent lighting using cool white 23

PAGE 24

lamps producing 40 footcandles on the horizontal wo rk space; 2) a system with all four walls illuminated and no central overhead lighting; 3) another system that combined the overhead fluorescents with the four walls illuminated; 4) overhead in candescent downlighting producing eight footcandles on the table; and 5) overhead incandescen t downlighting with one wall luminated. The six factors in these subjective ratings included visual clarity, spaciousness, pleasantness, spatial complexity, personal prom inence or anonymity, and relaxation or tension.7 Participants included a wider geographic and occu pational mix, such as, co ntractors, architects, engineers, educators, inte rior designs, and utilit y service representatives. The findings demonstrated a consistent pattern on relative response betw een the settings of the three rooms. The room th at was irregular was rated as being a tense and unpleasant space when it was being rated for the downlighting because th e light would not impose on the walls as much as the other spaces. Flynns study results reinforced his belief that the subjects impressions of the space were not on the appearan ce of the lighting system but on the pattern of light that it would create around them. Auditorium Room Study This study was conducted in a large square ro om with a higher ceiling, and participants were seated in an auditorium-style arrangeme nt facing a stage and a podium at Kent State University. The demographics of the subjects had changed to primarily architectural students from Kent State University. Th is study was to test the hypoth eses from the initial study by changing the room shape and finish, the activity setting, and having a diffe rent mix of subjects. Two groups of 30 participants rated the ligh ting conditions. The first group rated seven different lighting conditions in the spring of 1973, and the rating fo rm consisted of 13 scales. The second group rated 13 different lighting conditions in the winter of 1974, and the rating form consisted of 15 scales.6 24

PAGE 25

When analyzing the data, some ratings were close between the tw o unrelated groups and other ratings varied slightly, but, overall, they were generally cons istent in the direction of the ratings. Also, the rating patterns were consistent with the multidimensional scaling and semantic differential plots from the ini tial study done between 1971 and 1973. In further analysis, evident parallels occurred in the visual experience by different groups, even when they were in a significantly different room and activity context. Three additional components to the study were added: the glare/non-glare scale, uniform and non-uniform wall lighting, and the effects on th e lighting color tone. In dealing with the glare effect, no noticeable effect occurred on the ra tings of privacy, spaciousness, or complexity, but an effect appeared on clar ity and pleasantness appeared. In the initial study, wall lighting illustrated shifts of the impression on the space. During this study, in analyzing the information when wall lighting was added, the ratings shifte d toward more pleasant, clear, and spacious spaces. Since this study was conducted in spaces with various activities and various room shapes, the results were fairly consistent. It can be concluded that th e relating factors were influenced by the patterns of lighting more than the other physical aspect s of the rooms. With regard to the study of color effects, the results were irregular and seemed to be influenced not only by the warm/cool toned lights, but also from the distribution of the light. Diffused lighting was rated somewhat cooler than directional downlighting. Li ghting color provided significant ratings when dealing with pleasantness and sp aciousness, but contradictory for clarity and privacy because the orientation of the light affected this rating as well. Lighting Laboratory Study Flynns studies ended at the Architectural Engineering I llumination Laboratory at Pennsylvania State University between 1974 and 1975. Eight groups of 10 subjects were asked to evaluate eight different light settings. Semantic differential scaling (SDS) was used to rate the 25

PAGE 26

spaces initially and comparatively. Seven objectives were set for th is study: 1) to test the data from the Lighting Institute; 2) to study the effect of significant va riations in brightness intensities; 3) to test color tone variations; 4) to see if there is a pattern on s eat selection and ratings for the selected seat; 5) to see if ther e is a difference if one was wearing glasses, no glasses, or contacts; 6) to study when limited ener gy is available for non-uni form lighting; and 7) to further define the relationship between the ra tings and physical measurements of light in the space.6 This study contained three zones and some zones had different lighting conditions. During this study, each area with a seated subject was m easured to see if any co rrelation existed between the light measurement and the rating by the subjec t, but more of a correlation occurred between the ratings and patterns of light than the actual light measurement itself. One element that was examined was to provi de data in dealing with energy-budgeting situations. Flynn (1975) stated th at some non-quantitative aspects of the lighting can affect the overall perception of quantity of light. The study included tw o light conditions: one generated more electricity than the other. The one with greater energy usage had better ratings for visual clarity and spaciousness, but the one with lesser energy usage had be tter ratings in the evaluative field. The data showed that the lighting condi tion with lesser ener gy usage could still be efficient if used in a non-task situation. The result of his study showed no evidence th at would contradict the findings at the Lighting Institute. The data compiled from the st udy show that a visual pattern of light affects the subjects rating of the space, and even more so in this study because they differentiate the quality of light created on the different zones. 26

PAGE 27

Scaling of Slides Study Although the previous studies are quite useful to help designers choose appropriate lighting conditions, Flynn knew that it was not always easy to conduc t studies with a variety of spaces to judge. One would need a facility with a degree of fl exibility for precise and quick changing visual settings. Flynn thought a possible solution would be to crea te slide projections of the light settings. The study was conducted to see if a light setting from a slide proj ection could get similar ratings from a real setting. For this purpose, the slide pr ojections were ones from original studies of the lighting conditions. High quality slides were crea ted from continuously photographing the actual light settings and then re-creating the lighting condition accurately on the slide. When comparing the ratings of the slides to the real spaces the results at first seemed reasonably consistent, but the slid es were only approximately the sa me rating as the real room. In a further study, the ratings of the slides did not accurately di fferentiate between the lighting systems. If the study involved only slides, differe nt conclusions would have been formulated. Flynn concluded that during the time of the study the method used for the slides might not have been reliable and maybe certain aspects and rating s would have changed in rating a slide than the real space. Results of Impressions Visual clarity Visual clarity refers to the cl earness and distinctness of the visual environment. Clarity is reinforced by shadows, a higher luminance in the center of the room, and an emphasis on the horizontal surfaces, such as the work plane and ceiling.8 Two aspects of lighting discovered in this impression are as follows: 27

PAGE 28

Wall-lighting makes a significant contribution to impressions of clarity in a non-uniform setting along with overhead cool white fluorescents. Uniform overhead systems seem to increas e the impressions of clarity with central overhead cool white fluorescents. Public vs. private space When lighting a space, basic parameters were discovered on how to create a more public or private space. In dealing with these findings, one could properly light the space according to the type of space required. Four as pects of lighting dealing with this impression are as follows: Central overhead systems, even when they ar e non-uniform, are percei ved as public space. Non-uniform peripheral lighting systems, including overhead and wall-lighting, are perceived as more private. Cool color tones are seen as more public. If subjects are sitting in darker areas of the room, th ey perceive the space as more private. Relaxing vs. tense space The overall feeling of a space, wh ether it is relaxing or tense, is based on the patterns of light in a room, as well as the color of the light source. The patterns of light with relaxing and tense spaces can be combined with visual clar ity to create effective and comfortable working environments8. Four aspects of lighting dealing w ith this impression are as follows: Central overhead systems, both uniform and non-uniform, are perceived as more tense. Peripheral systems that are non-unifor m are perceived as more relaxing. Cool color tones of light are seen as more tense. Subjects sitting in darker zones perceive th e spaces in a great sense of relaxation. 28

PAGE 29

Spatial complexity The complexity of a space deals with the subj ects perception of a space as being either simple or complex. Two aspects of lighting de aling with this impression are as follows: Peripheral emphases in non-uniform settings seem to shift impressions toward greater complexity. Warmer tones of light seem to shift impressions toward greater complexity. Spaciousness The perception of the size of the space can be altered by the position and intensity of the lighting. Four aspects of lighting dealing with this impression are as follows: Uniform systems seem to shift impr essions toward more spaciousness. Non-uniform modes seem to be somewhat less spacious. To some extent, wall-ligh ting seems to give an eff ect of more spaciousness. Central overhead systems seem to be perceived as more spacious than non-uniform peripheral systems. Preference The effects on preference in a certain lighting situation deal with th e color tones of light, position of lights, and lighting zones. A co rrelation can occur betw een the preference and relaxing vs. tense scales. Three aspects of lighting dealing with this impression are as follows: Warm color tones in lighting a ffect preference more positively. Wall-lighting positively affects preference. Subjects sitting in darker zones seem to rate the preference of lighti ng situations higher. 29

PAGE 30

Glare Flynn tested for glare even though no apparent glare conditions were present in the light settings. Nevertheless, subjects were still able to respond to the scale.8 Two aspects of lighting dealing with this impression are as follows: Non-uniform cool white fluorescents seem to present less glare issues. More of an issue occurs with glare when light is positioned only in the center overhead of the room. Psychological Processes Influencing Lighting Quality Soon after Flynns studies, a group of researcher s at the Bartlett School of Architecture in London conducted similar studies. This paper presents a discussion concerning mediating psychological processes: perceived control, attention, environmenta l appraisal, and effect. Veitch (2001) not only discussed and analyzed the methods of John Flynn, but he also analyzed the long history of the quality of the luminous environmen t influencing task performance, comfort, and well-being. The overall focus of Veitchs study is on the scien tific evidence concerning the psychological aspects in the response to the overall luminous conditions. One lighting behavior element discussed rela ting to Flynns studies was the luminance distribution in rooms. Flynns work was seen as innovative and crea tive, as well as being the first to apply psychological techniques and multivariate st atistics to lighting research. He was the first to conduct these types of studies, but his data ha d many flaws. His sample size was too small for this type of analysis, as seen in the foundati ons of behavioral research. For every semantic differential, there should be 10 research subjects. Fl ynns study had 34 semantic differentials so he would need 340 participants, but he had only 96 reported participants.9 It was also unclear how the set of semantic differentials were deve loped, or even if they were valid descriptive words understood by the participants. During the study, the participants were never informed 30

PAGE 31

about the basis of the study so it was unclear to them if the evaluation concerned the room or the lighting. Also, Flynns studies have never been replicat ed by anyone independently. Referenced in many textbooks and writings, his studies concer ned particular lighting modes and subjective impressions, but this research is based only on one study. The research group at the Bartlett School of Architecture at Univer sity College London conducted a similar series of experiments pa rallel to Flynns work and used a similar methodology. The first study was conducted with 18 lighting configurations for a windowless two-person office. The research group was unable, however, to obtain an interpretable result in its MDS analysis.6 Later on, the team worked on lighti ng arrangements of office workspace in both laboratory mockups and field study. In anal yzing the way the studies were conducted and what lighting situations existed, the group needed to be able to identify the area of the visual field that influenced the assessment of the sp aces. During the research, the group discovered a horizontal band of 40 degrees wide, and it was centered at the hei ght of the viewer. The Bartlett group obtained similar results in several studies by using a wider range of luminous environment conditions and settings than Flynns group used. The Bartlett group further reinforced the brightness and uniformity di mensions of lighting appraisal. However, the groups sample size was too small fo r factor analysis to apply. In an attempt to verify the data recorded in the study before, a similar study was conducted. This time, 292 people were asked to rate th e appearance of the open-plan office using 27 semantic differential scales. After conducting th e study, the results were not the same as Flynns results or the previous test by the Bartlett group. The three components of th e study were visual attraction, complexity, and brightne ss. The participants were asked to judge the appearance of 31

PAGE 32

the room, not the lighting, to prevent a bias with the knowledge th at lighting is the focus of the experiment. Some authors actually disagreed with this method because it contributed to unwanted variance, and they would rather gi ve instructions to judge the lighting. Before the study of a non-uniform lighting dist ribution, with just the knowledge of visual adaptation, the Bartlett group thought that the overall perception would be murky. After conducting a study using an adjective checklist to describe each light ed environment, the Bartlett group found that the common experiences of gloom related to lighting were conditions of low adaptation luminances, mainly when vertical lu minances were low. Going against the groups hypothesis, not all non-uniform distri butions produced feelings of gloom, but the ones that were seen as gloom were associated with a dim or obscure periphery.9 The Bartlett group concluded its study by the wa y that people judge the lighting situation, depending on the appearance of a space. The main dimensions of these judgments appear to be brightness and interest. In the future, to be ab le to understand the aesthetic appeal of light, a systematic application needs to be in place to understand these judgments. Another psychological aspect in lighting beha vior relationships was the effect of the aesthetic judgments concerning the appearance of the space. The preference included the emotional component of how the space makes the participant feel. Robe rt Baron, a social and environmental psychologist, proposed a theory (as cited in Veitch, 2001) that environmental conditions create a state of positive effect that would lead to bette r performance, greater effort, and more willingness to help others.9 People prefer daylight or electric light, but the Bartlett group study was to see what light setting would be preferred to help with task performance. A preference of illuminance levels is 32

PAGE 33

normally higher than the recommended levels, but it might vary between i ndividuals, settings, and task. Several studies were conducted by many indi viduals to see if different preferences occurred, depending on sex and age. Leslie and Ha rtleb (as cited in Veitch, 2001) even suggested possible sex differences in illuminance preference (female participants preferred much lower levels than male participants), but they did not have many female participants. Butler and Biner (as cited in Veitch, 2001) found that females preferred different light levels from males in only 2 out of 43 behaviors; that preference related to washing dishes and parking garages.9 Boyce (1981) did a study to see if different preferences depended on age. The aging of the human eye affects ones vision, like yellowing of the eye and a worse visual sense. Older workers therefore need higher illuminance levels to see, but this is not a change in preference, just a change in illuminance need. Studies were conducted by Veitch (2001) on preference of illuminance levels as to the type of tasks being performed. When subjects were doing tasks on the computer they preferred lower illuminance levels than those doing paper tasks and visually demanding tasks, such as studying and reading. They preferred lower illumiance levels for more relaxing and intimate tasks, such as listening to music or talking with friends. In de aling with the distribut ion of light on certain visual tasks, Boyce (1981) evaluated various light settings and concluded that an acceptable task lighting method would be to provide a uniform distri bution over a large area of the desk surface. However, Slater and Boyce (as cited in Veitch, 2001) concluded th at illuminance ratios as low as 0.5 would be acceptable for most people, and bright surfaces without excessive glare were most highly rated.9 Veitch and Newsham (2001) discove red that participants preferred 33

PAGE 34

task/ambient lighting installations that are not uniform across the desk surface and that had lower ratings for glare. One key factor of rating a room is the brightne ss of vertical surfaces. Collins (as cited in Veitch, 2001) found that low ratings were give n by office occupants with a combination of indirect fluorescent lighting and furniture-mounted task lamps. The light ing created a relative high task illuminance but a low peripheral brightness around the wo rkstation. Collins (as cited in Veitch, 2001) concluded that people prefer brighter walls than darker ones, as long as they did not provide a source of glar e. After this preference was evident in studies, the next study to take place was to find how high the vertic al surface luminance should be. The tests to be considered are what tasks are taking place and whether subjects are in a reading, conference, or computer-based environment. Miller (as cited in Veitch, 2001) conducted an informal experiment of five different office scenes and determined that the most preferred was an equal amount of lighting on th e walls and the work space. The Bartlett group concluded that a non-uniform distribution of light for tasks on the computer was more comfortable. For tasks on the computer, the Bartlett research group found that non-uniform distributions of task and ambient lighting are desc ribed as more comfortable. When the overall illuminance level is high there is a desire for more non-uniform lighting but when the illuminance level is low there is a desire for more uniform lighting. 9 A secondary lighting element observed was glare. It was reasonable to assume that glare sources are not preferred because they can create visual disc omfort. Tests conducted by Veitch (2001) proved that overall light quality ratings with computers were affected greatly by the brightness of the screens. The brighter the imag e reflected from the scre en, the lower the rating of the light quality. 34

PAGE 35

All these studies tried to crea te a better method for predicting light quality from luminous conditions by organizing theore tical psychological processes dealing with lighting behavior relationships. These studies demonstrated that a need still exists for more lighting research because the past practices used a lot of energy, and we do not have that much energy supply anymore. Influence of Lighting Quality on Presence Task Performance in Virtual Environments Zimmons (2004) conducted a study to bring a re ality of the lighting environment in the virtual environment. He studied the importance of light in the physical environment and then related it to findings of importance in the virtual environment. He noted that light is critical to the visual experience because it exposes information about the environment and reveals the objects within it.10 Light falling on objects can create spatial cues pr oviding information about the objects relative positi ons and orientations. Flynn found that light can influe nce a users seating orientat ion, comprehension of room size and shape, task performance, and it can genera te certain behaviors. The way that light falls on an object can alter the impre ssion of that objects importance and use. Depending on the use of the space, whether it is public or private, a room can be lit differently to illustrate the intentions of the use and reinfo rce different set of behaviors. Zimmons (2004) investigated five hypotheses about lightin g quality in the virtual environment. These five hypotheses dealt with re sponses on lighting in low-stress environments, proving that subjects will be more accurate and have quicker response times in these types of environments. He tested lighting in the virtual environment on th e basis of lighting influences studied by Flynn. In task performance, the best visual percep tion occurs when the brightness ratio of the central task and background is between 1:1 a nd 4:1, where the tasks are brighter than the 35

PAGE 36

background.10 Zimmons study took this ratio into acc ount with three experiments giving a set of tasks that the participants were to complete In first experiment, called the pit experiment, participants performed a ball-dr opping task to determine if rendering quality influenced their ability to hit the target. The second experiment called the gallery experiment, tested the method of the number of times a participan t viewed a particular object in the space. This experiment, depending on the contrast of light ing objects, affected the viewing behavior of the participants where during high contrast tests, participants vi ews were attracted to highlighted objects. The third experiment, called the knot experiment, dealt with object selection, speed, and accuracy. Participants looked for objects on a table in the virtual environment and found a certain object. Zimmons (2004) study is about the use of virtua l environments with tools and information that affect the actual lighted e nvironment. The study did not add information to Flynns studies; it just used his findings to create virtual environments. Cognitive and Perceptual Factors in Lighted Architectural Environments Davis (2006) built on the work of Flynn. Th e purpose of his study was to link broader results of Flynns studies to the interaction of the envi ronment psychologically, not architecturally. In the research phase, Davis (2006) studied environm ent perception done by other researchers, such as Kaplan and Kaplan, Nasar, and Blasdel (as cited in Davis, 2006). They all conducted studies using multidimensional scaling (MDS). These studies did not involve real spaces, as in Flynns studies, but used render ed images. Subjects looked at the images and arranged them in certain orde rs of environment, making it easy to construct comparisons. Davis (2006) also looked at Fl ynns work, the studies conducted at the Bartlett School of Architecture (as cited in Veitc h, 2001), and the research done by Veitch (2001). Only Flynn had useful results using MDS techniques similar to the studies conducted for environmental psychology. Flynn also used factor an alysis to put subjective respons es into sets of impressions, 36

PAGE 37

but he did not create a strong relationship with environmental perceptio ns, since those studies were analyzed by using MDS methods. Davis study uses MDS so information will begin to bridge the gaps between the scope of light ing conditions and environmental perceptions. Methods Davis study used MDS for rating scales a nd preference judgments, which was one of the methods that were used in Flynns first study. Th ere were 24 different illumination situations for four different architectural c ontexts: an empty room, a caf, a lounge, and an office. The study had a total of 72 participants. Subjects comple ted distance judgments and preference selections for pair wise comparisons of stimuli, and comple ted a set of 22 ratings scales for each stimulus.11 The lighting systems were based on Flynns situati ons of overhead/periphe ral, bright/dim, and uniform/non-uniform. Davis conducted two studies ( 2006), the first of which was dealing with eight different lighting combinations: 1) central bright uniform (CBU), 2) central bright non-uniform (CBN), 3) central dim uniform (CDU), 4) central dim nonuniform (CDN), 5) perimeter bright uniform (PBU), 6) perimeter bright non-uniform (PBN ), 7) perimeter dim uniform (PDU), and 8) perimeter dim non-uniform (PDN). The sec ond study was conducted with 16 complex lighting situations, which combined central and perime ter lighting: CBU PBU, CBU PBN, CBU PDU, CBU PDN, CBN PBU, CBN PBN, CBN PDU, CBN PDN, CDU PBU, CDU PBN, CDU PDU, CDU PDN, CDN PBU, CDN PB N, CDN PDU, and CDN PDN.11 Rating When spaces were being rated 22 scales were used. The scales were selected from the scales used in studies by Flynn, Mehrabian and Russell, and Kaplan and Kaplan (as cited in Davis, 2006). 37

PAGE 38

Renderings The 24 lighting conditions were rendered in the program AGI 32 by Lighting Analysts Inc. The program AGI 32 is an available lighting system analysis and simulation program that has 3D rendering capabilities. The room was dimensioned at 40 feet by 40 feet with a 12-foot-high ceiling. The reflectance of the room was 80% fo r the ceiling and 50% for the walls, floor, and furnishings. For the renderings with centr al uniform lighting, a grid of 25 luminaire locations was defined. The grid was located 10 feet from each wall and the fixtures were spaced five feet apart. The luminaire used was an ERCO 2300 downlight, with a 50-watt halogen MR16 lamp with a beam angle of 55 degrees. For the central non-uniform lighting, ther e were five luminaire locations. Instead of a 50-wa tt halogen MR16 lamp, a 70-watt ha logen MR16 with a beam angle of 10 degrees was used. The computer renderings used an ERCO fl uorescent wall wash with a T5 High Output lamp for the perimeter uniform lighting. The lu minaires ran continuously and were three feet from the wall, aimed toward the wall. For the co mputer renderings with perimeter lighting in a non-uniform system, an ERCO 50-watt MR16 wall wa sh luminaire was located around the room at different points. This created the non-uniform pattern on the three walls being viewed, and the lights were positioned between two and five feet from the wall so that the patterns on the wall would differ.11 After all the rendered images were created, they were put into a presentation file and reviewed to make sure the intended variations we re present. Some images were modified to the intended differences because they were looking identical. Receiving accurate ratings of re ndering images can be very difficult due to the difficulty to accurately portray the qualities of a room and lighting in a room with a simulation. Several 38

PAGE 39

researchers validated the results of simulated envi ronments to the actual environments. Future researchers need to compare the results of simulated environments with those in real environments, similar to the study conducted by Flynn (1975) with the slide representations. Results The multidimensional scaling (MDS) analysis provided evidence of perceptual dimensions of lighting presented with specifica lly central or perimeter emphasis.11 Perimeter areas were seen as more important in non-workplace settings, such as a caf, while the central areas were seen as more important for the work setting. Four factors were seen as si gnificant: arousal, evaluative, sp aciousness, and uniformity. In summarizing the arousal factor results, uniform light is more arousing than non-uniform lighting. Bright lighting is more arousing than dim ligh ting, and central lighting is more arousing than perimeter lighting. In summarizing the evaluativ e factor results, uniform lighting had higher scores in an empty room and l ounge, but non-uniform lighting had hi gher scores in the caf. The evaluative scores did not seem to change in the office with the uniformity of the lighting. Bright lighting made a difference only in the evaluative sc ore for the lounge, receiv ing higher scores for bright lighting compared to dim lighting, but brightness did not seem to matter for other contexts. Central lighting produ ced higher evaluative scores than perimeter lighting for the caf and office, but the empty room had high eval uative scores for peri meter lighting. Little difference occurred in the scores in the lounge context. In the spaci ousness factor results, uniform lighting was seen as more spacious th an non-uniform lighting, brighter lighting was more spacious than dimmer lighting, and uniform lighting was seen as more spacious than nonuniform lighting. In the uniformity factor resu lts, the only significance was the uniform lighting and the non-uniform lighting, but the scores we re more significant for the empty room and 39

PAGE 40

lounge context compared to the caf and office. The brightness and area lighting did not affect the uniformity factor score.11 Relating to Flynns Work Daviss (2006) project was a more in-depth i nvestigation of the dimensions of perception of lighted environments because he dealt with computer simulated lighting conditions. He was able to produce more lighting situations and had particular control over the physical parameters. Flynns study with overhead/peripheral lighting dealt with the overall lighting of the space. Davis dealt with central/perimeter lighting with an emphasis for all contexts. Flynns lighting mode with uniform/non-uniform lighting dealt with the uniformity of lighting on surfaces, while Davis dealt with perimeter uniformity in all the c ontexts except for office. Flynn also dealt with uniform/non-uniform lighting with the objects of a room, which correlated with Daviss central uniformity for all contexts except the caf. The bright/dim mode for Flynn dealt with the intensity of light on the central horizontal plane, but for Davis, central brightness for the caf and other contexts seemed to reinforce the central /perimeter mode. As previously noted, the study confirmed results found by Flynn, but with some modifications. 40

PAGE 41

CHAPTER 3 METHODOLOGY Overview In earlier research, architectural lighting systems and the users perceptions were rated in a variety of ways. Flynn (1975) had a small number of lighting conditions, and, for part of his studies, he did not account each individuals per ception of lighting. He touched on the study of two-dimensional imaging and relating the ratings to real lighting situations. Two different methods were used to analyze the data: 1) semant ic differential, where he could provide insight on how people felt in a space; and 2) multidimen sional scaling, where he could analyze the perceptual judgments of the spac e. In the studies conducted by Veitch (2001), the locations of the studies were conducted in labor atories with lighting conditions, as well as in existing office spaces. The studies were to further the resear ch previously done by Flynn and also to conduct studies with more participants and more lighti ng situations. One of the studies dealt with preference and task performance using different lighting levels. Zimmon s (2004) was using only the information provided by Flynns studies to influence the virtua l studies he conducted. Davis (2006) conducted his studies using rendered images for participants judgments. He analyzed his data using MDS methods, just as Flynn did, for part of his studies to reinforce the results found by Flynn. With the knowledge of these other studies a nd wanting to continue to strengthen these studies results found by these studi es, the experiment needed to follow the same methods used from the previous studies. Two experiments were conducted: one in a real environment and the second in a judgment of the same space but in dig ital format. Three lighti ng situations were in the real environment, and two li ghting situations were in the re ndered environment. Thirty-two 41

PAGE 42

subjects participated in the study, 19 subjects in each study ranging from groups of two to eight people in each of the sessions. Research Design Independent Variables Architectural context Light studies need to be conducted in a space where the lighting parameters can be controlled, but the physical aspects of the room can st ay intact. In most of the prior studies, the types of room conditions were shown in conference rooms or classrooms. This study was in a computer classroom setting. The room provided space for 10 subjects to be studied, and each participant was provided a desktop computer for the task portion of the real environment light settings and rendering environmen t studies (see Figure 3-1). Four sessions were available for each experiment so participants could choose an av ailable time period. The range of participants in each session was between one and eight subjects. Figure 3-1. Plan of the study room with seat assignments. In each session, the subjects were told the in tention of the study while the first lighting condition was present. In the real environment light settings, the participants were given an interactive task on the computer. They were then to ld the basis of the experiment is dealing with 42

PAGE 43

three different lighting conditions, as well as the clarity of the task on the computer. During the sessions of the rendered lighting environments, pa rticipants were told how the basis of the experiment was going to be conducted by looking at two different lighting conditions and to perceive the rendered condition as though it was a real environment. A full description of the instructions given to the subj ects is found in Appendix A. Lighting situations The lighting systems used during the study ha d three different lighting conditions that could be present in a workspace or office sp ace. The footcandle le vels in each of the participants seats were measured (see Table 31). The specifications of the three lighting conditions were as follows: Overhead fluorescent (CW) 91W; 95% Uplight and 5% Downlight; CRI 86 (see Figure 32) Peripheral halogen (WW) downlighting; 50W; CRI 100 (see Figure 3-3) Peripheral daylighting (CW); one window CRI 100 (see Figure 3-4) Table 3-1. List of footcandle levels for each seat during the study. Seat Fluorescent Halogen Daylight 1pm Daylight 2pm Daylight 3pm Daylight 4pm 1 16.4 1.7 19.2 19.8 14.8 6.6 2 18.1 1.2 24.6 24.4 19.5 8.9 3 16.8 6.2 67.2 67.2 62.3 22.3 4 17.9 1.5 89.2 87.3 85.2 29.8 5 23.8 1.9 87.8 78.6 95.2 31.9 6 24.5 1.0 67.1 60.3 61.2 26.5 7 20.6 0.6 26.4 23.2 22.9 10.6 8 17.9 2.0 19.6 19.2 17.6 8.3 9 22.6 1.2 21.8 23.5 23.2 10.9 10 19.8 3.1 18.2 10.2 17.9 7.8 43

PAGE 44

Figure 3-2. Plan of room with location of fluorescent lighting. Figure 3-3. Plan of room with location of halogen lighting. Figure 3-4. Plan of room with location of source of daylight. Rendered situations The two electric lighting situations were rendered in an accurate lighting simulation program (3D Max Design 2009). The photometric data of the lighting fixt ures are supplied by the Illumination Engineering Society (IES files) on the supplier websites. The data is inputted into the rendering program to produce accurate li ghting systems compared to the actual light 44

PAGE 45

settings in the real e nvironment studies. The room was m odeled and assigned materials exactly how it appeared in the real room (see Figures 3-5 and 3-6). The room s were rendered without furniture to have the judgment based only on the lighting design. Figure 3-5. Three-dimensional simula tion of fluorescent light setting. Figure 3-6. Three-dimensional simu lation of halogen light setting. 45

PAGE 46

Dependent Variables Rating scales Depending on the study, subjects evaluated th ree lighting conditions (real environment rating) using a set of bi-polar rating scales. Th e ratings were selected from prior publications, and they were based on relevant adjectives the s ubject groups could understand. The ratings were divided into three groups: perceptual clarity of task, evaluative, and spaciousness and complexity. A seven-point scale was used to allo w a finer grade of judgmen t than the five-point scale, but it did not become too difficult for subjects to grade a nd to allow a neut ral zone, unlike the six-point scale.5 The rating scale survey is shown in Appendix C. Descriptive information Above the rating scales on the survey, subjects filled out descriptive information that could affect their judgments on the sp ace. One factor was the seat number because every seat perceived different light levels, es pecially in the daylight setti ng, that might alter the subjects impressions. Another factor taken into account was whether or not subjects were wearing contacts or glasses to analyze the informati on to see if corrective lenses affected their impressions of the lighting conditions. Subjects A pool of 52 volunteer subjects was recruited who had an architectural background or background in a related field. Al l participants were students. Some who had a background in architecture were currently studyi ng a different discipline unrela ted to architecture. Ultimately, only 38 subjects participat ed; some had scheduling conflicts or were no-shows. Each subject took part in either two or three sessions, de pending on which day the individual participated. There was space for up to 10 subjects per session; the sessions resulted in two to eight subjects. 46

PAGE 47

Nineteen subjects participated in the three real lighting cond itions, and an additional 19 subjects participated in the two rendered light ing conditions. Before pa rticipating, each subject read and signed an informed consent form that had been approved by the University of Floridas Institutional Review Board. The consen t form is provided in Appendix D. Procedure Subjects reported to Computer Lab Room 217 in the McGuire Dance/Theatre Pavilion at the University of Florida for two or three expe rimental sessions. The sessions were conducted on two days, March 19 and 20, 2009. When the subjects ente red the room, they chose to sit at any of the available seats, and each subject read a nd signed the informed consent form, then the experimenter read the instructi ons for the session. The instructions are provided in Appendix A. During each study, subjects observed the space for the appropriate adaptation time discussed in the instructions a nd were given the surveys after th e period of adaptation had ended. The experiment protocol is pr ovided in Appendix B. The experiment of the real lighting conditions did not last longer th an 50 minutes and the experi ment of the rendered lighting conditions did not last longer than 30 minutes. Data Collection During the collection process, each survey wa s reviewed. All the descriptive information and ratings were put into a Microsoft Excel spreadsheet and colorcoded by light source. Limitations It is a good rule of judgment with semantic di fferential scaling to have 6 to 10 times the number of subjects as you have scales.6 This study was conducted using 16 scales, so at a minimum, the study should have had 96 subjects. But the study had only 38 subjects. However, since each subject did at least two sessions, a total of 95 surveys were filled out so the sample size may be too small to draw concrete conclusion s. Also, the light levels at each seat in the 47

PAGE 48

room were measured with a light meter. The meters are older and might not be taking accurate measurements. The overall comparisons of the light levels are correct, bu t the numerical values of the light levels might be inaccurate. Also, in dealing with the ratings of the rendered environments, each participant had a computer mon itor to pull up the images to rate them. Each computer is the same model and, they were checke d to see if all the settin gs were the same. In looking around during the study, clear differences we re apparent in the coloring of the image where some would be perceived as warmer light depending on the computer each subject was using. These differences could be because some computers might be used more than others and the lamp inside the monitor could be wearing out. Data Analysis Method The collection and analysis of data dealt with the semantic differential scale (SDS) used by Flynn during his studies. It is a method used to measure social attitudes in psychology. Semantic differential scaling sti ll appears to be among the best av ailable instruments for measuring multicomponent concepts.6 Using SDS helps to simplify the complex set of data, based on the number of correlations between variables (i.e ., different lighting conditions, variances in footcandle levels in each lighting situation, whether the individual is wearing contacts or glasses, and comparing the real environment to the rende red environment). The means of the ratings for the variables are calculated and graphed to illust rate the relationships with each other pertaining to the scales. The standard devi ation is then calculated to see th e variation of the ratings from the mean.12 For example, the ratings for the halogen lighting pertaining to the scale of glare-no glare are as follows: 3, 7, 7, 4, 2, 7, 7, 3, 7, 3, 7, 3, 6, 7, 5, 5, 4, 4, 1. The mean of this rating is 4.84, but the standard deviation (see Equation 3-1) is 2.01. The higher the number, the more the ratings vary from the mean. The lower number dem onstrates that the ratings are closer to the value of the mean. 48

PAGE 49

(3-1) Taking into account the calculations of th e mean and standard deviations, a 95% confidence interval (see Equation 3-2) wa s calculated for the set of data. (3-2) The interval incorporates a margin of error so it will help to gauge the accuracy of the means. If the intervals between tw o sets of compared data intersect with each other, then it will demonstrate significant statistical data are differentiating them.12 49

PAGE 50

CHAPTER 4 RESULTS Factor One: Analyzing Data of the Three Lighting Conditions The first factor analyzes the overall means of the three different li ghting conditions. The analysis is broken down into the three impression s: perceptual clarity of task, evaluative, and spaciousness and complexity. Perceptual Clarity of Task Analysis The perceptual clarity of task ratings was an alyzed in all three lighting conditions (see Figure 4-1). Reading the figure, the median rating has the value rating of 4 between the two bipolar adjectives. Fluorescent lightin g was perceived as clearer, bright er, produced little glare, and was more focused. Halogen lighting had a little amount of clarity, was dimmer, produced little to no glare, and was focused. Daylighting was cleare r, almost at the median between glare and no glare, was brighter, and was th e least focused light setting. Figure 4-1. Perceptual clarity comparison means of fluorescent, halogen, and daylight lighting. The following conclusions were made: Halogen lighting received the rating for most hazy and the most focused, which might result in the quality of light being hazy, but the light functio ned as a spotlight might result in be considered more focused. Halogen lighting was significantly dimm er than fluorescent and daylight. 50

PAGE 51

Fluorescent lighting produced th e least amount of glare and was rated the clearest, so it seems to be the best lig hting for task lighting. Table 4-1. Standard deviations of the three lighting conditions for perceptual clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 Going deeper into the statistical analysis after finding the standard deviations (see Table 41), the confidence intervals can be charted (see Figur e 4-2). As seen in the figure, most of the intervals overlap to show no real significance in the different ratings. Figure 4-2. Perceptual clarity comparison conf idence intervals of fluorescent, halogen, and daylight lighting. The following information can be observed from the Figure 4.2: Halogen lighting is perceived as the dimmes t; it is the only significant difference in the perceptual clarity of th e three lighting conditions. Fluorescent lighting appears to produce the least amount of gl are but from the intervals there is no real signi ficant difference. Evaluative Analysis The mean evaluative ratings were analyzed in all three lighting condi tions (see Figure 4-3). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives. Fluorescent lighting is almost aligned with the median rating for four out of the five 51

PAGE 52

ratings, and it is considered du ller. Halogen lighting is considered more relaxed, interesting, likeable, and warmer in temperature. Daylighting is seen as more inviting, relaxed, interesting, and likeable. It is also near the me dian cool-warm value in temperature. Figure 4-3. Evaluative compar ison means of fluorescent, ha logen, and daylight lighting. The following information can be observed from Figure 4-3: Daylighting was considered the most inviti ng, relaxed, and likeable of the three lighting conditions. Halogen lighting is the most interesting and warmest of the three lighting conditions. Fluorescent lighting is seen as the dullest, coolest in temperatur e, and most disliked of the three lighting conditions. Table 4-2. Standard deviations of the three lighting conditions for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-2), the confidence intervals can be charted (see Figure 4-4). As s een in the figure, most of the intervals overlap to show no real significance in the different ratings. 52

PAGE 53

Figure 4-4. Evaluative comparis on confidence intervals of fluor escent, halogen, and daylight. The following information can be observed from the figure 4-4: Fluorescent lighting is significantly disliked. Fluorescent lighting is the dullest compared to the other two lighting situations. Spaciousness and Complexity Analysis The mean of the spaciousness and complexity ratings were analyzed in all three lighting conditions (see Figure 4-5). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjec tives. Fluorescent lighting is cons idered smaller, tighter, more simple, private, and quiet. It is also near the median betw een informal-formal and cluttereduncluttered. Halogen lighting is c onsidered more private, quiet, a nd uncluttered. It is also near the median between small-large, tight-open, simp le-complex, and informal-formal. Daylighting is more open, public, informal, unclu ttered, large, simple, and quiet. The following information can be observed in Figure 4-5: Most mean ratings are along the median value for spaciousness and complexity. Daylighting creates the largest, the most in formal and the most public space of the three lighting conditions. 53

PAGE 54

Halogen lighting creates the most complex and private space of the three lighting conditions. Halogen and fluorescent lighting had a similar pattern of ratings compared to daylighting. Figure 4-5. Spaciousness and complexity co mparison means of fluorescent, halogen, and daylight lighting. After calculating the standard de viations during the statistical analysis (see Table 4-3), the confidence intervals can be charted (see Figure 4-6) As seen in the figure, the most significant difference is with the daylight setting. Table 4-3. Standard deviations of the three lighting conditions for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 54

PAGE 55

Figure 4-6. Spaciousness and complexity comp arison confidence interv als of fluorescent, halogen, and daylight lighting. The following information can be observed from Figure 4-6: Daylighting creates as signifi cantly larger and more open and public space than the other light settings. Halogen lighting may create a significantly mo re complex space than fluorescent lighting, but shows no significant difference compared to daylighting. Factor Two: Analyzing Data of the Corrective Lenses vs. None The second factor analyzes th e overall means of weather the subjects were wearing corrective lenses or not. The an alysis is broken down into the three impressions of: perceptual clarity of task, evaluative, and spaciousness and complexity. 55

PAGE 56

Fluorescent Perceptual clarity of task analysis The perceptual clarity of the task ratings was analyzed in whether the subjects were wearing corrective lenses or not (see Figure 4-7). Reading th e figure, the median rating has the value rating of 4 between the two bi-polar adjectives. Those with corrective lenses perceived less glare and the space was considered dimmer than those that were not wearing corrective lenses. Figure 4-7. Perceptual clarity comparison means for fluorescent lighting, whether the subjects were wearing correc tive lenses or not. The following information can be observed from Figure 4-7: Hazy-Clear and Unfocused-Focu sed ratings were similar. Those wearing corrective lenses perceived spaces as dimmer. Those wearing corrective lenses perceived less glare. Table 4-4. Standard deviations for fluorescent li ghting, whether subjects are wearing corrective lenses or not for perceptual clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-4), the confidence intervals can be charted (see Figure 4-8). As s een in the figure, the 56

PAGE 57

confidence intervals are too large to obtain a ny significant difference between those who were wearing corrective le nses to those who were not; possibl y from not having enough subjects or that the difference of ratings between subjects are too high. Figure 4-8. Perceptual clarity comparison confid ence intervals for fluorescent lighting, whether the subjects were wearing corrective lenses or not. Evaluative analysis The mean evaluative ratings were analy zed for whether the subjects were wearing corrective lenses or not (see Figure 4-9). Reading the figure, th e median rating has the value rating of 4 between the two bipolar adjectives. The perceptio n of those wearing corrective lenses is almost down the median rating for four out of the five ratings, and they perceived fluorescent lighting as more host ile. The perception of those not wearing corrective lenses is almost aligned with the median rating for thr ee out of the five ratings, and they perceived fluorescent lighting dulle r and more inviting. The following information can be observed from Figure 4-9: The mean values of the relaxed-tense, dull-in teresting, cool-warm, a nd dislike-like ratings are similar between those wearing corre ctive lenses and those who are not. Those wearing corrective lenses perceive d fluorescent lighting as more hostile. 57

PAGE 58

Figure 4-9. Evaluative compar ison means for fluorescent lighting, whether the subjects were wearing corrective lenses or not. Table 4-5. Standard deviati ons for fluorescent lighting, whet her the subjects were wearing corrective lenses or not for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 After calculating the standard de viations during the statistical analysis (see Table 4-5), the confidence intervals can be charted (see Figure 4-10). As seen in the figure, most the confidence intervals are too large to obt ain any significant difference be tween those who were wearing corrective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings between subjects are too high. But those who do not have small intervals like the inviting-hostile and dislik e-like ratings show that the s ubjects had very similar ratings. The following information can be observed from Figure 4-10: Fluorescent lighting is perceive d significantly more hostile to those wearing corrective lenses. There are no other significant differences when comparing the evaluative intervals. 58

PAGE 59

Figure 4-10. Evaluative comparison confidence intervals for fluorescent lighting, whether the subjects were wearing co rrective lenses or not. Spaciousness and complexity analysis The mean of spaciousness and complexity rati ngs were analyzed in whether the subjects were wearing corrective lenses or not (see Figu re 4-11). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives The overall ratings between subjects wearing corrective le nses and those who did not obtained similar means and did not differ significantly. The following information can be observed Figure 4-11: The mean values of the small-large, tightopen, simple-complex, private-public, and noisyquiet ratings are similar between those weari ng corrective lenses and those who are not. Those wearing corrective lenses rated that fluorescent lightin g was slightly more formal than those who did not wear corrective lenses. Those wearing corrective lenses rated that fluorescen t lighting was slightly more cluttered than those who did not wear corrective lenses. 59

PAGE 60

Figure 4-11. Spaciousness and complexity comparison means for fluorescent lighting, whether the subjects were wearing corrective lenses or not. Table 4-6. Standard deviati ons for fluorescent lighting, whet her the subjects were wearing corrective lenses or not for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-6), the confidence intervals can be charted (see Figure 4-12). As seen in the figure, most the confidence intervals are too large to obtai n any significant difference between those who were wearing corrective lenses to those who were not; possi bly from not having enough subjects or that the difference of ratings between subjec ts are too high. But those which do not have small 60

PAGE 61

intervals like the small-large, tight-open, and simp le-complex ratings show that the subjects had very similar ratings. Figure 4-12. Spaciousness and complexity co mparison confidence intervals for fluorescent lighting, whether the subjects were wearing corrective lenses or not. Halogen Perceptual clarity of task analysis The perceptual clarity of the task ratings was analyzed in whether the subjects were wearing corrective lenses or not (see Figure 4-13). Reading th e figure, the median rating has the value rating of 4 between the two bi-polar ad jectives. The overall ratings between subjects wearing corrective lenses and t hose who did not, differed slightly in some of the ratings. The following information can be observed from Figure 4-13: The mean values of the hazy -clear and dim-bright ratings are similar between those wearing corrective lenses and those who are not. Those wearing corrective lenses pe rceived spaces as more focused. Those wearing corrective lenses perceived more glare. 61

PAGE 62

Figure 4-13. Perceptual clarity comparison m eans for halogen lighti ng, whether the subjects were wearing correc tive lenses or not. Table 4-7. Standard deviations for halogen lighting, whether s ubjects are wearing corrective lenses or not for perceptual clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-7), the confidence intervals can be charted (see Figure 4-14). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference be tween those who were wearing corrective lenses to those who were not; possi bly from not having enough subjects. Figure 4-14. Perceptual clarity comparison confidence intervals for halogen lighting, whether the subjects were wearing co rrective lenses or not. 62

PAGE 63

Evaluative analysis The mean evaluative ratings were analy zed for whether the subjects were wearing corrective lenses or not (see Figur e 4-15). Reading the figure, the median rating has the value of 4 between the two bi-polar adjectives. The overa ll ratings between subjects wearing corrective lenses and those did not, differed sl ightly in some of the ratings. The following information can be observed from Figure 4-15: The mean values of the relaxed-tense, dull-in teresting, and dislike-li ke ratings are similar between those wearing corrective lenses and those who are not. Those wearing corrective lenses percei ved halogen lighting as more hostile. Those not wearing corrective lenses pe rceived halogen lighting as cooler. Figure 4-15. Evaluative comparison means fo r halogen lighting, whether the subjects were wearing corrective lenses or not. Table 4-8. Standard deviations for halogen lighting, whethe r the subjects were wearing corrective lenses or not for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 63

PAGE 64

After calculating the standard de viations during the statistical analysis (see Table 4-8), the confidence intervals can be charted (see Figure 4-16). As seen in the figure, most the confidence intervals are too large to obt ain any significant difference be tween those who were wearing corrective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings between subjects are too high. But those who do not have small intervals like the dull-interesting, c ool-warm, and dislike-like ratings show that the subjects had very similar ratings. Figure 4-16. Evaluative comparison confidence intervals for halogen lighting, whether the subjects were wearing co rrective lenses or not. Spaciousness and complexity analysis The mean of spaciousness and complexity rati ngs were analyzed in whether the subjects were wearing corrective lenses or not (see Figu re 4-17). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives The overall ratings between subjects wearing corrective le nses and those who did not obtained similar means and did not differ significantly. 64

PAGE 65

The following information can be observed Figure 4-17: The mean values of the small-large, ti ght-open, simple-complex, noisy-quiet, and cluttered-uncluttered ratings are similar betw een those wearing correc tive lenses and those who are not. Those wearing corrective lenses rated that halogen lighting was slightly more formal than those who did not wear corrective lenses. Those wearing corrective lenses rated that halogen lighting created a slightly more private space than those who did not wear corrective lenses. Figure 4-17. Spaciousness and complexity comp arison means for halogen lighting, whether the subjects were wearing co rrective lenses or not. Table 4-9. Standard deviations for halogen lighting, whether s ubjects are wearing corrective lenses or not for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 65

PAGE 66

When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-9), the confidence intervals can be charted (see Figure 4-18). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference between those who were wearing corrective lenses to those who were not; possi bly from not having enough subjects. Figure 4-18. Spaciousness and complexity compar ison confidence intervals for halogen lighting, whether the subjects were wear ing corrective lenses or not. Daylight Perceptual clarity of task analysis The perceptual clarity of the task ratings was analyzed in whether the subjects were wearing corrective lenses or not (see Figure 4-19). Reading th e figure, the median rating has the value rating of 4 between the two bi-polar ad jectives. The overall ratings between subjects wearing corrective lenses and t hose who did not obtained similar means and did not differ significantly. 66

PAGE 67

The following information can be observed Figure 4-19: The mean values of the hazy -clear and unfocused-focused ratings are similar between those wearing corrective lens es and those who are not. Those wearing corrective lenses rated that dayligh ting produced slightly more glare than those who did not wear corrective lenses. Those wearing corrective lenses rated that daylighting was sli ghtly brighter than those who did not wear corrective lenses. Figure 4-19. Perceptual clarity comparison mean s for daylighting, whether the subjects were wearing corrective lenses or not. Table 4-10. Standard deviations for daylighting, whether subjects are wearing corrective lenses or not for perceptual clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-10), the confidence intervals can be charte d (see Figure 4-20). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference be tween those who were wearing corrective lenses to those who were not; possi bly from not having enough subjects. 67

PAGE 68

Figure 4-20. Perceptual clarity comparison conf idence intervals for daylighting, whether the subjects were wearing co rrective lenses or not. Evaluative analysis The mean evaluative ratings were analy zed in whether the subjects were wearing corrective lenses or not (see Figure 4-21). Reading the figure, the median rating has the value rating of 4 between the two bipolar adjectives. The overall rati ngs between subjects wearing corrective lenses and those who did not obtaine d similar means and did not differ significantly. Figure 4-21. Evaluative comparison means for da ylighting, whether the subjects were wearing corrective lenses or not. 68

PAGE 69

The following information can be observed Figure 4-21: The mean values of the relaxed-tense, dullinteresting, and cool-warm ratings are similar between those wearing corrective lenses and those who are not. Those wearing corrective lenses rated that daylighting created a slightly more inviting space than those who did not wear corrective lenses. Those wearing corrective lenses rated that daylighting created a slightly more likable space than those who did not wear corrective lenses. Table 4-11. Standard deviations for daylighting, whether subject s are wearing corrective lenses or not for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-11), the confidence intervals can be charted (see Figure 4-22). Figure 4-22. Evaluative comparison confidence in tervals for daylighting, whether the subjects were wearing correctiv e lenses or not. As seen in the figure, most of the confid ence intervals are too large to obtain any significant difference between those who were wear ing corrective lenses to those who were not; 69

PAGE 70

possibly from not having enough s ubjects or that the difference of ratings between subjects are too high. But those who do not have small interv als like the inviting-hostile and relaxed-tense ratings show that the subjects had very similar ratings. Spaciousness and complexity analysis The mean of spaciousness and complexity rati ngs were analyzed in whether the subjects were wearing corrective lenses or not (see Figu re 4-23). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives The overall ratings between subjects wearing corrective le nses and those who did not obtained similar means and did not differ significantly. Figure 4-23. Spaciousness and complexity co mparison means for daylighting, whether the subjects were wearing co rrective lenses or not. The following information can be observed Figure 4-23: The mean values of the small-large, tightopen, private-public, noisy-quiet, and cluttereduncluttered ratings are similar between those wearing corrective lens es and those who are not. 70

PAGE 71

Those wearing corrective lenses rated that daylighting created a slightly more simple space than those who did not wear corrective lenses. Those wearing corrective lenses rated that daylighting created a slightly more informal space than those who did not wear corrective lenses. Table 4-12. Standard deviations for daylighting, whether subjects are wearing corrective lenses or not for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-12), the confidence intervals can be charted (see Figure 4-24). Figure 4-24. Spaciousness and complexity comp arison confidence intervals for daylighting, whether the subjects were wear ing corrective lenses or not. 71

PAGE 72

As seen in the figure, most the confidence inte rvals are too large to obtain any significant difference between those who were wearing corr ective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings between subjects are too high. But those who do not have small intervals like the small-large, noisy-quiet, and cluttereduncluttered ratings show that the subjects had very similar ratings. Factor Three: Analyzing Data of the Different Light Levels The third factor analyzes the overall mean of the different light levels of halogen lighting and daylighting. The analysis is broken down into the three impressions of: perceptual clarity of task, evaluative, and spaciousness and complexity. Halogen Perceptual clarity of task analysis The perceptual clarity of task ratings was analyzed in whether subjects were wearing corrective lenses or not (see Figure 4-25). Read ing the figure, the median rating has the value rating of 4 between the two bi -polar adjectives. The overall ratings between subjects who wearing corrective lenses and t hose who did not obtained similar means and did not differ significantly. Figure 4-25. Perceptual clarity comparison means for different light levels of halogen lighting. 72

PAGE 73

The following information can be observed Figure 4-25: The mean values of the glare-no glare and unf ocused-focused ratings are similar between different halogen levels. Subjects who experienced a lower light level perceived a more hazy space. Subjects who experienced a lower light level perceived a dimmer space. Table 4-13. Standard deviations for different light levels of halogen lighting for perceptual clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 After calculating the standa rd deviations during the statistical analysis (see Table 4-13), the confidence intervals can be charted (see Figure 426). As seen in the figure, the confidence intervals are too large to obt ain any significant difference be tween those who were wearing corrective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings betw een subjects are too high. Figure 4-26. Perceptual clarity comparison confid ence intervals for different light levels of halogen lighting. 73

PAGE 74

Evaluative analysis The mean evaluative ratings were analy zed in whether the subjects were wearing corrective lenses or not (see Figure 4-27). Reading the figure, the median rating has the value rating of 4 between the two bipolar adjectives. The overall rati ngs between subjects wearing corrective lenses and those who did not obtaine d similar means and did not differ significantly. Figure 4-27. Evaluative comparison means for different light levels of halogen lighting. The following information can be observed from Figure 4-27: The mean values of the cool-warm and dislik e-like ratings are similar between different halogen levels. Subjects who experienced a lower light le vel perceived a more hostile space. Subjects who experienced a lower light level perceived a more tense space. Subjects who experienced a lower light le vel perceived a more interesting space. Table 4-14. Standard deviations for different light levels of halogen lighting for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 74

PAGE 75

When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-14), the confidence intervals can be charte d (see Figure 4-28). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference be tween those who were wearing corrective le nses to those who were not; possibl y from not having enough subjects or that the difference of ratings between subjects are too high. Figure 4-28. Evaluative comparison confidence in tervals for different light levels of halogen lighting. Spaciousness and complexity analysis The mean of spaciousness and complexity rati ngs were analyzed in whether the subjects were wearing corrective lenses or not (see Figu re 4-29). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives The overall ratings between subjects wearing corrective le nses and those who did not obtained similar means and did not differ significantly. The following information can be observed Figure 4-29: The mean values of the small-large, tightopen, simple-complex, private-public, informalformal, and cluttered-uncluttered ratings ar e similar between different halogen levels. Subjects who experienced a lower light level perceived a more quiet space. 75

PAGE 76

Figure 4-29. Spaciousness and complexity comp arison means for different light levels of halogen lighting. Table 4-15. Standard deviations for different light levels of hal ogen lighting for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-15), the confidence intervals can be charte d (see Figure 4-30). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference be tween those who were wearing corrective le nses to those who were not; possibl y from not having enough subjects or that the difference of ratings between subjects are too high. 76

PAGE 77

Figure 4-30. Spaciousness and complexity compar ison confidence intervals for different light levels of halogen lighting. Daylight Perceptual clarity of task analysis The perceptual clarity of the task ratings was analyzed in whether the subjects were wearing corrective lenses or not (see Figure 4-31). Reading th e figure, the median rating has the value rating of 4 between the two bi-polar ad jectives. The overall ratings between subjects wearing corrective lenses and t hose who did not obtained similar means and did not differ significantly. The following information can be observed Figure 4-31: The mean values of the hazy-c lear and glare-no glare ratings are similar between different daylight levels. Subjects who experienced a lower dayli ght level perceived a dimmer space. Subjects who experienced a lower daylight level perceived a more unfocused space. 77

PAGE 78

Figure 4-31. Perceptual clarity comparison mean s for different light levels of daylighting. Table 4-16. Standard deviations for different light levels of daylig hting for perceptu al clarity of task. Rating Fluorescent Halogen Daylight Hazy-Clear 1.17 1.78 1.54 Glare-No Glare 1.41 2.01 1.67 Dim-Bright 1.2 1.45 1.55 Unfocused-Focused 1.79 1.34 1.42 After calculating the standa rd deviations during the statistical analysis (see Table 4-16), the confidence intervals can be charted (see Figure 432). As seen in the figure, the confidence intervals are too large to obt ain any significant difference be tween those who were wearing corrective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings betw een subjects are too high. Figure 4-32. Perceptual clarity comparison confid ence intervals for different light levels of daylighting. 78

PAGE 79

Evaluative analysis The mean evaluative ratings were analy zed in whether the subjects were wearing corrective lenses or not (see Figure 4-33). Reading the figure, the median rating has the value rating of 4 between the two bipolar adjectives. The overall rati ngs between subjects wearing corrective lenses and those who did not obtained similar means and did not differ significantly. The following information can be observed Figure 4-33: The mean values of the dull-in teresting, cool-warm, and dislike-like ratings are similar between different daylight levels. Subjects who experienced a lower daylight level perceived a more inviting space. Subjects who experienced a lower daylight level perceived a more relaxed space. Figure 4-33. Evaluative comparison means fo r different light leve ls of daylighting. Table 4-17. Standard deviations for different light levels of daylighting for evaluative. Rating Fluorescent Halogen Daylight Inviting-Hostile 1.39 1.89 1.31 Relaxed-Tense 1.26 1.82 0.96 Dull-Interesting 1.13 1 1.47 Cool-Warm Dislike-Like 1.46 1.08 1.22 1.07 1.49 1.21 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-17), the confidence intervals can be charted (see Figure 4-34). 79

PAGE 80

Figure 4-34. Evaluative comparison confidence interv als for different light le vels of daylighting. As seen in the figure, the confidence inte rvals are too large to obtain any significant difference between those who were wearing corr ective lenses to those who were not; possibly from not having enough subjects or that the difference of ratings between subjects are too high. Spaciousness and complexity analysis The mean of spaciousness and complexity rati ngs were analyzed in whether the subjects were wearing corrective lenses or not (see Figu re 4-35). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives The overall ratings between subjects wearing corrective le nses and those who did not obtained similar means and did not differ significantly. The mean values of the small-large, tightopen, simple-complex, informal-formal, noisyquiet, and clutter-uncluttered ratings are similar between different daylight levels. Subjects who experienced a lower daylight level perceived a more private space. 80

PAGE 81

Figure 4-35. Spaciousness and complexity comp arison means for different light levels of daylighting. Table 4-18. Standard deviations for different light levels of daylighting for spaciousness and complexity. Rating Fluorescent Halogen Daylight Small-Large 1.21 1.5 1.21 Tight-Open 1.42 1.27 1.46 Simple-Complex 1.25 1.33 1.29 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.6 1.73 1.54 1.41 1.45 1.33 0.96 1 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-18), the confidence intervals can be charte d (see Figure 4-36). As seen in the figure, the confidence intervals are too large to obtain a ny significant difference be tween those who were wearing corrective le nses to those who were not; possibl y from not having enough subjects or that the difference of ratings between subjects are too high. But those who do not have small intervals like the cluttered-uncluttered ratings sh ow that the subjects ha d very similar ratings. 81

PAGE 82

Figure 4-36. Spaciousness and complexity compar ison confidence intervals for different light levels of daylighting. Factor Four: Analyzing the Data of Real vs. Rendered Environment The second factor analyzes the overall mean s of the two lighting conditions between the real and rendered environments. The analysis is broken down into th e three impressions of perceptual clarity of task, evaluativ e, and spaciousness and complexity. Fluorescent Perceptual clarity of task analysis The perceptual clarity of task ratings were analyzed in the fluorescent lighting condition. The data was calculated and compared from the real environment and the rendered environment (see Figure 4-37). Reading the figure, the median rating has th e value rating of 4 between the two bi-polar adjectives. Fluorescent rendered lighting image seemed to be perceived as producing more glare than the real environment. 82

PAGE 83

Figure 4-37. Perceptual clarity of task comparison means of fluorescent real and rendered environment The following can be observed from Figure 4-37: The only change in the percep tual impression between the re ndered and real environment is the glare-no glare rating. The clear-hazy rating was the same value betw een the rendered and real environment. Bright-dim and focused-unfocus ed have similar results. Table 4-19. Standard deviations of the fluorescent real and rendered conditions for perceptual clarity of task. Rating Real Rendered Hazy-Clear 1.17 1.35 Glare-No Glare 1.41 1.48 Dim-Bright 1.2 1.35 Unfocused-Focused 1.79 1.7 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-19), the confidence intervals can be charte d (see Figure 4-8). As seen in the figure, all the intervals overlap to show no real significance in the different ratings. 83

PAGE 84

Figure 4-38. Perceptual clarity of task comparison confidence intervals of fluorescent real and rendered environment The following information can be observed from Figure 4-38: The results between the real and re ndered environment are similar. A slight difference occurs between the two for the glare-no glare rating, but it is not significant enough. Evaluative analysis The evaluative ratings were analyzed in th e fluorescent lighting condition. The data were calculated and compared from the real environm ent and the rendered environment (see Figure 439). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives. The two conditions ar e consistent, but the rendered environment received slightly higher ratings. The following information can be observed from the figure: The rating of the values between the rende red and real environment are similar. The ratings of the rendered values were sl ightly higher than the real environment. 84

PAGE 85

Figure 4-39. Evaluative comparison means of fluorescent real and rendered environment Table 4-20. Standard deviations of the fluorescent real and rendered conditions for evaluative. Rating Real Rendered Inviting-Hostile 1.39 1.4 Relaxed-Tense 1.26 1.48 Dull-Interesting 1.13 1.38 Cool-Warm Dislike-Like 1.46 1.08 1.47 1.12 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-20), the confidence intervals can be charte d (see Figure 4-40). As seen in the figure, all the intervals overlap to show real significance in the different ratings. The following information can be observed from Figure 4-40: The results between the real and render ed environment are almost identical. Even though the intervals ar e overlapping, it appe ars the rendered environment appears more likeable. 85

PAGE 86

Figure 4-40. Evaluative comparison confidence intervals of fluorescent real and rendered environment Spaciousness and complexity analysis The spaciousness and complexity ratings we re analyzed in the fluorescent lighting condition. The data were calculated and compared from the real environment and the rendered environment (see Figure 4-41). Reading the figur e, the median rating has the value rating of 4 between the two bi-polar adjectives. The fluorescent rendered image is not as uniform in ratings as the real environment. The following information can be observed from Figure 4-41: The rendered image is perceived as larger, more open, simpler, quieter, and less cluttered than the real environment. The rendered image did not have any furniture added, as in the real environment, which could have affected those ratings. 86

PAGE 87

Figure 4-41. Spaciousness and complexity comp arison means of fluorescent real and rendered environment. Table 4-21. Standard deviations of the fluorescent real and rendered conditions for spaciousness and complexity. Rating Real Rendered Small-Large 1.21 1.56 Tight-Open 1.42 1.73 Simple-Complex 1.25 0.96 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.62 1.8 1.54 1.57 1.85 1.82 1.48 1.75 After calculating the standa rd deviations during the statistical analysis (see Table 4-21), the confidence intervals can be charted (see Figure 4-42). As seen in the figure, most of the intervals overlap to show no real signifi cance in the different ratings. The following information can be observed from Figure 4-42: The rendered environment is rated significantly less cluttered than the real environment. All the other ratings are similar and show no significant difference when the confidence intervals are analyzed. 87

PAGE 88

Figure 4-42. Spaciousness and complexity compar ison confidence intervals of fluorescent real and rendered environment Halogen Perceptual clarity of task analysis The perceptual clarity of task ratings were analyzed in the haloge n lighting condition. The data was calculated and compared from the real environment and the rendered environment (see Figure 4-43). Reading the figure, the median ra ting has the value rating of 4 between the two bipolar adjectives. The halogen rendered image was dimmer, less focused, produced less glare, and hazier than the real environment. The following information can be observed from Figure 4-43: The real and rendered environment varies significantly in most of the ratings. The glare-no glare rating varies th e least out of the four ratings. 88

PAGE 89

Figure 4-43. Perceptual clar ity of task comparison means of halogen real and rendered environment Table 4-22. Standard deviations of the halogen real and rend ered conditions for perceptual clarity of task. Rating Real Rendered Hazy-Clear 1.73 1.65 Glare-No Glare 2.01 1.48 Dim-Bright 1.45 0.85 Unfocused-Focused 1.34 1.68 When going deeper into the stat istical analysis after finding th e standard deviations (see Table 4-22), the confidence intervals can be charted (see Figure 4-44). As seen in the figure, only one of the intervals overl aps to show a significant diffe rence between the ratings. Figure 4-44. Perceptual clarity of task comparison confidence intervals of halogen real and rendered environment 89

PAGE 90

The following information can be observed from Figure 4-44: Only the glare-no glare rating obtained similar results. The rendered environment was perceived sign ificantly hazier, dimmer, and less focused than the real environment. Evaluative analysis The evaluative ratings were analyzed in the halogen lighting condition. The data were calculated and compared from the real environm ent and the rendered environment (see Figure 445). Reading the figure, the median rating has the value rating of 4 between the two bi-polar adjectives. The two conditions are consistent, but the rendered environment is irregular compared to the real environments slightly high ratings. The rendered environment received lower ratings: inviting-hostile, relaxed-tense, dull-int eresting, and dislike-like. The rendered environment also received a higher value in th e cool-warm rating. The dislike-like and dullinteresting have the highest value diff erences between the two environments. Figure 4-45. Evaluative comparison means of halogen real and rendered environment. The following information can be observed from Figure 4-45: The rating of the values between the rende red and real environment are similar. 90

PAGE 91

The rendered environment seems significantly more disliked and duller than the real environment. Table 4-23. Standard deviations of the halogen real and rend ered conditions for evaluative. Rating Real Rendered Inviting-Hostile 1.89 2.04 Relaxed-Tense 1.82 1.76 Dull-Interesting 1 1.72 Cool-Warm Dislike-Like 1.22 1.07 1.32 1.74 After calculating the standa rd deviations during the statistical analysis (see Table 4-23), the confidence intervals can be charted (see Figure 4-46). As seen in the figure, all of the intervals overlap to show no real signifi cance in the different ratings. The following information can be observed from Figure 4-47: The results between the real and rendered envi ronments are within ea ch others confidence intervals. The rendered environment has a much larger c onfidence interval than the real environment within the dull-interesting and dislike-like ratings. This shows more of a fluctuation in values in the rendered environment than in the real environment. Figure 4-46. Evaluative comparison confidence intervals of halogen real and rendered environment. 91

PAGE 92

Spaciousness and complexity analysis The spaciousness and complexity ratings were analyzed in the halogen lighting condition. The data were compared and calculated from th e real environment and the rendered environment (see Figure 4-47). Reading the figure, the median rating has th e value rating of 4 between the two bi-polar adjectives. The halogen rendered imag e is not as uniform in ratings as the real environment. The following information can be observed from Figure 4-47: The rendered image is perceived larger, more open, simpler, quieter, and less cluttered than the real environment. The rendered image did not have any furniture added, as in the real environment, which could have affected those ratings. Figure 4-47. Spaciousness and complexity comp arison means of halogen real and rendered environment. 92

PAGE 93

After calculating the standa rd deviations during the statistical analysis (see Table 4-24), the confidence intervals can be charted (see Figure 4-48). As seen in the figure, all of the intervals overlap to show no real signifi cance in the different ratings. Table 4-24. Standard deviations of the halogen real and rend ered conditions for spaciousness and complexity. Rating Real Rendered Small-Large 1.5 1.64 Tight-Open 1.27 1.42 Simple-Complex 1.33 1.61 Private-Public Informal-Formal Noisy-Quiet Cluttered-Uncluttered 1.6 1.73 1.54 1.41 1.26 1.57 1.55 1.72 Figure 4-48. Spaciousness and complexity compar ison confidence intervals of halogen real and rendered environment. The following information can be observed from Figure 4-48: All of the ratings are similar and there is no significant difference when the confidence intervals are analyzed. 93

PAGE 94

The real environment has a much larger confidence interval than the rendered environment within the private-public rating. This shows more of a fluctuat ion in values in the real environment than in the real environment. 94

PAGE 95

CHAPTER 5 SUMMARY AND CONCLUSIONS Conclusions Compared to Flynns Impressions As discussed in the literature review, the seven factors in th e subjective ratings included visual clarity, spaciousness, preference, glare, spatial comple xity, public or private, and relaxation or tension. This study incorporated more bi-polar adjectives to create three subjective rating groups that deal with some of the six factors from Flynns impressions (see Table 5-1). Table 5-1. Comparison of Flynns impressions. Gillars Impressions Flynns Impressions Perceptual Clarity of Task Visual Clarity Glare Evaluative Relaxation or Tension Preference Spaciousness and Complexity Spaciousness Spatial Complexity Public or Private Flynn reported that uniform overhead systems seem to increase the impressions of clarity with central overhead cool white fluorescents. This result was al so supported in this study, but the difference between the ot her lighting conditions was not significant enough. During Flynns study (1975), less glare issues occu rred with non-uniform cool white fluorescents and more glare issues occurred with a central light fixture. This study did not have a central light fixture, but the daylight condition was from one wall, which pr esented a glare issue, and the cool white fluorescents had the least gl are issues (see Table 5-2). Flynn reported that non-uniform peripheral systems are more relaxing, and central overhead systems are perceived as more tense. This study supported th at theory, though the difference between the values is not significant enough. Daylighti ng was actually seen as more relaxing. Flynn reported that warm color tones and wall-lighting are preferred. His theory was supported in this study, and the difference between the values was significa nt (see Table 5-2). 95

PAGE 96

Flynn reported that uniform lighting systems a nd central overhead systems were perceived as more spacious than non-uniform peripheral sy stems. In this study, the daylight setting was considered the most spacious. Peripheral lig hting and overhead ligh ting with halogen and fluorescent lighting were rated similar and pe rceived as smaller and tighter spaces. Flynn reported that peripheral non-uniform settings had greater comple xity ratings. This result was supported in this study, and the difference betw een the values was si gnificant. Flynn reported that central overhead systems were perceived as public space, and peri pheral lighting systems were perceived as more private, which was supported in this study. However, the difference between the values was not significant, and daylig hting was rated significantly more public than the other two lighting co nditions (see Table 5-2). Table 5-2. Comparison of results to Flynns results. Flynns Impressions Flynns Results Gillars Results Visual Clarity Increase clarity with central overhead cool fluorescent lighting Supported, but difference was not significant Glare Less glare with cool fluorescent lighting More glare with central lighting Supported More glare with daylighting Relaxation or Tension Non-uniform peripheral lighting seen as more relaxing than central overhead lighting Supported, but difference was not significant Daylighting most relaxing Preference Warm color tones and peripheral lighting more preferred Supported Spaciousness Uniform and central overhead lighting more spacious Uniform and central overhead lighting smaller Daylighting perceived as larger and more open Spatial Complexity Peripheral non-uniform lighting more complex Supported Public or Private Overhead lighting perceived as public and peripheral more private Supported, but difference was not significant 96

PAGE 97

Three Lighting Conditions There were significant differences that are noted during the analysis of the three different lighting conditions. This data can result in what lighting condition should be used for certain spaces. Fluorescent lighting appeared to produce the least amount of glare, even though it was not a significant difference, but coul d be used for the classroom and office setting. But it is also considered significantly dull and highly disliked out of the three lighti ng conditions. Fluorescent lighting should probably be paired with another lighting condition to crea te a more interesting and well-liked space. Halogen lighting was perceived significan tly dimmer than the other two lighting conditions, which shows that warm halogen lighting should not be used dur ing tasks that need higher light levels like reading. It also cr eates a significantly more complex space than fluorescent lighting, so it should be chosen to light galleries and restaurants. Daylighting was perceived the most positive of the all the lighting conditions. Although the difference between the ratings was not significant when deali ng with the confidence interval, daylighting was perceived as the most invi ting, relaxed, and likeable lighting condition. Daylighting was perceive d significantly larger, as well as mo re open and public from the other lighting conditions. The overall impression of day lighting is perceived positively and should be incorporated into all spaces with additional li ghting. There should be further testing with different daylighting co ntrol technologies. Corrective Lenses vs. None There was only one significant difference that was noted during the analysis of subjects wearing corrective lenses to thos e that were not wearing any co rrective lenses. This could be because there is a coating on lenses that blocks ou t blue light to prevent glare. The results from 97

PAGE 98

this test are not valid because there were not enough subjects involved. Further investigation should be implemented with more subjects to create valid results. Different Light Levels During the data analysis comparing the ratings between different light levels of the halogen lighting and daylighting, there is no real significant difference. It could have been possible that the different levels in daylighting would have had differences because of the extreme range of light levels (10.6 95.2 fc). Some of the subjects were si tting in direct sun light, receiving more heat and possibly more glare, but that did not affect their ratings. Also depending on the time that the subjects participate d, some experienced darker c onditions because it became more cloudier and their ratings were not affected. People seemed to be rating the entire room it did not matter how much light was at th eir individual seat but how much light wa s perceived in the overall room. The results from this test are not valid because there were not enough subjects involved. Further investiga tion should be implemented with more subjects to create valid results. Real vs. Rendered Environments Flynn did a small study to compare the results from the real environment study to the results from rating a series of high quality slide images of the same lighting conditions. The overall comparison between the two mean valu es had significant differences. This study between the real environment and a three-dimens ional rendering of the environment illustrated similar ratings. The perceptual clarity of task factor was th e only one out of the three factors that had significant differences. No task was shown in the rendered image, so the clarity factor would be difficult to calculate. The rendered environment of the halogen lighting condition was perceived as less cluttered than the real environment, but that could result from th e fact that there was no furniture rendered in the lighting conditions. 98

PAGE 99

The overall values between the evaluative and sp aciousness factors are similar. From this study, using the IES files of the lamp type or potential lamp type, one can judge the subjective impressions of the evaluative and spaciousness factors of the space by a rendered image. An issue that should be addressed is screen differences. Ev en though every screen was set to the same setting, some of the images appe ared differently; this could be because some computers could be getting more used than others so that some of the lamps are not as bright. This issue might have affected the overall ratings of the rendered environment. Recommendations The next six steps in creating a more comp lex study would be to include the following goals: 1) to obtain more volunteers and have them be a normal sample of the population, not just those with architectural background; 2) to survey the bi-polar ad jectives that are used in the study to see what people think when they see a certa in set of adjectives to see if it can properly relate to the lighting conditions a nd to see if the adjectives are va lid in the study; 3) to combine and compare the three different lighting conditions to see the types of ratings that would result in the three lighting conditions that were tested; 4) to analyze different age groups to see the different preferences depending on age group; 5) to create different ratios of brightness from the sources of the light and see how it effects the impressions; and 6) to test during different time periods of the day to see if significant differen ces occur in daylighti ng qualities because the morning presents a warm light and the afternoon presents a cool light. A study of comparisons between the real and rendered environments s hould be continued with more complex lighting conditions, including daylighting to see if similar results could still be obtained. This study was based on a room with existing lighting conditions. If there was more time, resources, and money, it would interesting to use a room with a blank canvas in need of lighting and create innovative and different methods of lighting that ma y not be found in most spaces 99

PAGE 100

(i.e., lighting integrated into the walls and floor). Perhaps differe nt lighting methods could create new parameters for different lightin g distribution strategies. Also, it would be useful if more task performances were tested. This test was in a classroom setting using computers. It would be interesting to find out if there are different perceptions to spa ces if you were in a restaurant setting, an office setting, or a hospital setting to name a few. Many new technologies are availabl e in the lighting industry that have not been tested for subjective impressions (i.e., fiber optics, light emitting diodes [LEDs], compact fluorescents [CFLs]). It would be interesting to see how th ese lighting types affect occupants subjective impressions. These lighting conditio ns are already known to be more sustainable and efficient in energy usage. The lighting types may also be mo re efficient in receiving the desired impression of the space. The type of occupants who are using the space can suggest the lighting design of a space. Their culture can strongly affect the way they pe rceive space. For example, those in Western cultures appreciate light and clarity, while Asia n cultures appreciate the shadows and subtleties in a space.13 It would be interesting to study a variety of cultures to see if the parameters of lighting differ. 100

PAGE 101

APPENDIX A INSTRUCTIONS TO SUBJECTS Real Environment Rating I would like to thank every one of you for volunteering an hour of your day to help me with my research. Today, you will be rating three different lighting conditions. First, I would like you to read and fill out the participant consent form explaining the basis of the research study and stating that your name s will remain anonymous in the results of the study. During each light study, I will give your eyes ti me to adapt to each lighting condition. The range of adaptation time is anywhere between 2 milliseconds and 10 minutes I will be giving your eyes a five-minute period of adaptation for each condition. During this time of adaptation, I would like you to partake in a task on the computer in front of you because part of the survey is rating the clarity of your task. The task that I would like you to do is so mething interactive like solitaire. After the five-minute period of adaptation, I will be handing out the survey that I would like you to complete. The top portion is the descriptive information; your seat number is located on top corner of the computer screen. Next ther e are the rating scales. Pl ease go with your first instinct of your rating and do not change any previously answered questions. Once everyone is done with the survey, I will collect them and ch ange to the next lighting condition, and I will distribute another surv ey after the five-minute adaptation time. Now for the second lighting condition, again, please do an intera ctive task on your computer during the five-minute adaptation time, and when it has been five minutes, I will begin to hand out the same survey to be filled out for this lighting condition. Please fill out the descriptive information for this survey and pro ceed to the rating scales of the survey. Once 101

PAGE 102

everyone is done with the survey, I will collect them and we will move on to the last lighting condition, and I will distribute another survey after the five-minute adaptation time. Now for the third and final lighting condition, please continue doing th e interactive task on your computer during the five-mi nute adaptation time, and when it has been five minutes, I will begin to hand out the same survey to be filled out for this lighting cond ition. Please fill out the descriptive information for this survey and pro ceed to the rating scales of the survey. Once everyone is done with the survey, I will collect th em and that will be the conclusion of my study. The study is complete. Thank you very much for your participation. Render Environment Rating I would like to thank every one of you for volunteering an hour of your day to help me with my research; this study will be taking onl y 30 minutes of your time. Today, you will be rating two different rendered images re lating to real life situations. First, I would like you to read and fill out the participant consent form explaining the basis of the research study and stating that your name s will remain anonymous in the results of the study. During each study, I will give your eyes time to adapt to looking at each image on the computer screen in front of you. I feel that an adaptation time of two minutes will be enough to adjust to the different images on the computer screen. The range of adaptation time in a real lighting situation is anywhere between 2 millisec onds and 10 minutes. I feel that an adaptation time of two minutes will be enough to adjust to the different images on the computer screen. During this time of adaptation, I would like you to continue to l ook at the computer screen and image that this is a real lighting situation. After the two-minute period of adaptation, I w ill be handing out the survey that I would like you to complete. The top portion is the descriptive information; your seat number is located 102

PAGE 103

103 on top corner of the computer screen. Next ther e are the rating scales. Pl ease go with your first instinct of your rating and do not change any previously answered questions. Once everyone is done with the survey, I will collect them and we will move on to the second and final rendered lighting condition, and I will distribute another survey after the twominute adaptation time. Now for the second and final rendered lighting condition, again, please continue to look at the image on the computer in front of you and imag e that this is a real lighting situation during the two-minute adaptation time, and when it has been two minutes, I will begin to hand out the same survey to be filled out for this rendere d lighting condition. Please fill out the descriptive information for this survey and proceed to the ra ting scales of the survey. Once everyone is done with the survey, I will collect them and that will be the c onclusion of my study. The study is complete. Thank you very much for your participation.

PAGE 104

APPENDIX B EXPERIMENT PROTOCOL Two different types of experiments will be cond ucted to test different lighting techniques. Experiment A Experiment A will take place in room 217 of the McGuire Theater and Dance Pavilion. The participants will be given an interactive task on the desk top computers that are already provided in the room and will complete a handwritte n survey. The test will show the efficiency of lighting in the room when given a task that ad mits light back to the participant (the computer screen). They will fill out a survey fo r each lighting situation presented to them. The following is the timeline of the experiment: The subjects arrive in the computer lab, choose a seat in front of a computer out of the 10 that are provided, then are given instruc tions on how the experiment will be conducted, and the first lighting condition will already be present. (5 min.) The subjects will do an interactive task on the computer during the five-minute period of adaptation. (5 min.) The experimenter will begin to hand out the survey and subjects will fill out the descriptive information and begin go rate the space. (5 min.) The experimenter will collect the surveys and change to the second lighting condition. (2 min.) Steps 2 to 4 will be repeated for each light setting. (12 min.) Experiment B Experiment B will take place in room 217 of the McGuire Theater and Dance Pavilion. The participants will be presente d with two digital light renderings of the same light qualities of Experiments A. Since these studies will be done by looking at images at the computer screen, we will not need as long of a period of adapta tion as Experiment A. The test will show the preference of certain light settings and see if a relation exists to the pattern of preference to the 104

PAGE 105

light settings in Experiment A. They will fill out a survey for each lighting situation presented to them. The following is the timeline of the experiment: The subjects arrive in the computer lab, choose a seat in front of a computer out of the 10 that are provided, then are given instructi ons on how the experiment will be conducted and the first lighting condition will already be present. (5 min.) The subjects will open the first rendered image on the computer and continue to study it during the two-minute period of adaptation. (2 min.) The experimenter will begin to hand out the survey and subjects will fill out the descriptive information and begin go rate the space. (5 min.) The experimenter will collect the surveys and change to the second lighting condition. (2 min.) Steps 2 to 4 will be repeated for the next rendered light setting. (9 min.) 105

PAGE 106

APPENDIX C SAMPLE OF RATING SCALE SURVEY Descriptive Information: Seat Number: ________________________________________ Major: ______________________________________________ Age: _________ Are you wearing contacts: Circle Yes No Are you wearing glasses: Circle Yes No Do you feel you are sensitive to light Yes No Rating: Perceptual Clarity of Task Hazy 1 2 3 4 5 6 7 Clear Glare 1 2 3 4 5 6 7 No Glare Dim 1 2 3 4 5 6 7 Bright Unfocused 1 2 3 4 5 6 7 Focused Evaluative Inviting 1 2 3 4 5 6 7 Hostile Relaxed 1 2 3 4 5 6 7 Tense Dull 1 2 3 4 5 6 7 Interesting Cool 1 2 3 4 5 6 7 Warm Dislike 1 2 3 4 5 6 7 Like Spaciousness and Complexity Small 1 2 3 4 5 6 7 Large Tight 1 2 3 4 5 6 7 Open 106

PAGE 107

Simple 1 2 3 4 5 6 7 Complex Private 1 2 3 4 5 6 7 Public Informal 1 2 3 4 5 6 7 Formal Noisy 1 2 3 4 5 6 7 Quiet Cluttered 1 2 3 4 5 6 7 Uncluttered 107

PAGE 108

APPENDIX D PARTICIPANT CONSENT FORM Dear Student: I am a graduate student at the University of Florida. For my thesis, I am conducting a study to advance the knowledge of percep tual relationships of lighting in architecture. I am asking you to participate in this study because your educa tional background can be helpful in my study. The purpose of my study is to collect data to find pa rameter needs for certain lighting conditions to be efficient and to create a positive effect on one s perception of space and to make the occupant more comfortable. Participants will be asked to participate in a series of lighting situations lasting no longer than one hour. Th ere are two different types of studies and you can only be a part of one. Your identity will be kept conf idential to the extent provided by law and your identity will not be revealed in my thesis. There are no anticipated risks, compensation, or other direct benefits to you as a participant in this study. You are free to withdraw your c onsent to participate and may discontinue your participation in the study at a ny time without consequence. If you have any questions about this research protocol, please cont act me at (954)439-3017 or my faculty supervisor, Martin Gold at ( 352)392-0205 x 209. Questions or concerns about your rights as a research participant ri ghts may be directed to the IRB02 office, University of Florida, Box 112250, Gainesville, FL 32611; (352) 392-0433. By signing this letter, you give me permissi on to report your responses anonymously in my thesis that will be submitted to the University of Florida as my final requirements of my degree to be submitted to my faculty supervisor as part of my coursework. Kimberly Gillar ___________________________________________________ 108

PAGE 109

109 I have read the procedure described above for thesis research. I voluntarily agree to participate in the study. ________________________________________________________________________ Signature of participant Date ________________________________________________________________________ Participants Name ________________________________________________________________________ E-mail Address Major and Year _____________________________________ Check here if preferred method of contact Phone Number

PAGE 110

APPENDIX E STATISTICAL DATA Factor One: Three Lighting Conditions Fluorescent Lighting n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.47 1.17 (4.91, 6.03) Glare-No Glare 6 1.41 (5.32, 6.68) Dim-Bright 4.89 1.2 (4.31, 5.37) Unfocused-Focused 5.26 1.79 (4.4, 6.12) Inviting-Hostile 3.58 1.39 (2.91, 4.25) Relaxed-Tense 3.37 1.26 (2.76, 3.98) Dull-Interesting 3.05 1.13 (2.51, 3.59) Cool-Warm 3.63 1.46 (2.93, 4.33) Dislike-Like 4.05 1.08 (3.53, 4.57) Small-Large 3.37 1.21 (2.79, 3.95) Tight-Open 3.37 1.42 (2.69, 4.05) Simple-Complex 3 1.25 (2.4, 3.6) Private-Public 3.21 1.62 (2.43, 3.99) Informal-Formal 4.16 1.8 (3.29, 5.03) Noisy-Quiet 5.16 1.54 (4.42, 5.9) Uncluttered-Cluttered 4.42 1.57 (3.66, 5.18) Halogen Lighting n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 4.89 1.73 (4.06, 5.72) Glare-No Glare 4.84 2.01 (3.87, 5.81) Dim-Bright 3.11 1.45 (2.41, 3.81) Unfocused-Focused 5.68 1.34 (5.03, 6.33) Inviting-Hostile 3.58 1.89 (2.67, 4.49) Relaxed-Tense 3.26 1.82 (2.38, 4.14) Dull-Interesting 5.32 1 (4.84, 5.8) Cool-Warm 5.05 1.22 (4.46, 5.64) Dislike-Like 5.47 1.07 (4.95, 5.99) Small-Large 3.58 1.5 (2.86, 4.3) Tight-Open 3.79 1.27 (3.18, 4.4) Simple-Complex 4.26 1.33 (3.62, 4.9) Private-Public 2.63 1.6 (1.86, 3.4) Informal-Formal 4.26 1.73 (3.43, 5.09) Noisy-Quiet 5.32 1.54 (4.58, 6.06) Uncluttered-Cluttered 5 1.41 (4.32, 5.68) 110

PAGE 111

Daylighting n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.37 1.54 (4.63, 6.11) Glare-No Glare 4.32 1.67 (3.52, 5.12) Dim-Bright 4.79 1.55 (4.04, 5.54) Unfocused-Focused 4.63 1.42 (3.95, 5.31) Inviting-Hostile 2.53 1.31 (1.9, 3.16) Relaxed-Tense 2.42 0.96 (1.96, 2.88) Dull-Interesting 4.79 1.47 (4.08, 5.5) Cool-Warm 4.11 1.49 (3.39, 4.83) Dislike-Like 5.53 1.47 (4.82, 6.24) Small-Large 5.16 1.21 (4.58, 5.74) Tight-Open 5.37 1.46 (4.65, 6.09) Simple-Complex 3.11 1.29 (2.49, 3.73) Private-Public 5 1.45 (4.3, 5.7) Informal-Formal 3.11 1.33 (2.47, 3.75) Noisy-Quiet 5.37 0.96 (4.91, 5.83) Uncluttered-Cluttered 5 1 (4.52, 5.48) Factor Two: Corrective Lenses vs. None Fluorescent Corrective Lenses n=8 t=2.365 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.5 1.31 (4.41, 6.59) Glare-No Glare 6.5 0.53 (6.05, 6.95) Dim-Bright 5.38 0.74 (4.75, 6) Unfocused-Focused 5.25 1.67 (3.85, 6.65) Inviting-Hostile 4.5 0.93 (3.73, 5.27) Relaxed-Tense 3.63 1.51 (2.37, 4.89) Dull-Interesting 3.38 2 (1.71, 5.04) Cool-Warm 3.37 1.19 (2.38, 4.37) Dislike-Like 4 1.07 (3.11, 4.89) Small-Large 3.13 0.99 (2.3, 3.95) Tight-Open 3.25 1.58 (1.93, 4.57) Simple-Complex 2.88 0.99 (2.05, 3.7) Private-Public 3.25 1.83 (1.72, 4.78) Informal-Formal 4.63 2 (2.96, 6.29) Noisy-Quiet 5.38 1.69 (3.97, 6.78) Uncluttered-Cluttered 4.13 1.89 (2.55, 5.7) 111

PAGE 112

Fluorescent None n=11 t=2.228 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.45 1.13 (4.69, 6.21) Glare-No Glare 5.6 1.75 (4.42, 6.78) Dim-Bright 4.55 1.37 (3.63, 5.47) Unfocused-Focused 5.27 1.95 (3.96, 6.58) Inviting-Hostile 2.9 1.3 (2.03, 3.77) Relaxed-Tense 3.18 1.08 (2.45, 3.91) Dull-Interesting 3.09 1.22 (2,27, 3.91) Cool-Warm 3.82 1.66 (2.7, 4.94) Dislike-Like 4.09 1.14 (3.32, 4.86) Small-Large 3.55 1.37 (2.63, 4.47) Tight-Open 3.45 1.37 (2.53, 4.37) Simple-Complex 3.09 1.44 (2.12, 4.06) Private-Public 3.18 1.53 (2.15, 4.21) Informal-Formal 3.82 1.66 (2.7, 4.94) Noisy-Quiet 5 1.48 (4.01, 5.99) Uncluttered-Cluttered 4.64 1.36 (3.73, 5.55) Halogen Corrective Lenses n=8 t=2.365 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.13 1.89 (3.55, 6.7) Glare-No Glare 4.38 2.45 (2.33, 6.42) Dim-Bright 3 1.77 (1.52, 4.48) Unfocused-Focused 6.25 0.46 (5.86, 6.64) Inviting-Hostile 4 2.39 (2, 6) Relaxed-Tense 3.63 2.33 (1.68, 5.57) Dull-Interesting 5.38 0.92 (4.61, 6.14) Cool-Warm 5.75 1.28 (4.69, 6.82) Dislike-Like 5.38 1.19 (4.38, 6.37) Small-Large 3.25 1.49 (2.01, 4.49) Tight-Open 3.38 1.51 (2.12, 4.63) Simple-Complex 4.5 1.85 (2.95, 6.05) Private-Public 2.13 1.13 (1.18, 3.07) Informal-Formal 3.5 1.93 (1.89, 5.11) Noisy-Quiet 5 1.69 (3.59, 6.41) Uncluttered-Cluttered 5 1.69 (3.59, 6.41) 112

PAGE 113

Halogen None n=11 t=2.228 Rating Mean SD 95 % Confidence Interval Hazy-Clear 4.73 1.68 (3.6, 5.86) Glare-No Glare 5.18 1.66 (4.06, 6.3) Dim-Bright 3.18 1.25 (2.34, 4.02) Unfocused-Focused 5.27 1.62 (4.18, 6.36) Inviting-Hostile 3.27 1.49 (2.27, 4.27) Relaxed-Tense 3 1.41 (2.05, 3.95) Dull-Interesting 5.27 1.1 (4.53, 6.01) Cool-Warm 4.55 0.93 (3.93, 5.17) Dislike-Like 5.55 1.04 (4.85, 6.25) Small-Large 3.82 1.54 (2.79, 4.85) Tight-Open 4.1 1.04 (3.4, 4.8) Simple-Complex 4.1 0.83 (3.54, 4.66) Private-Public 3 1.84 (1.77, 4.23) Informal-Formal 4.82 1.4 (3.88, 5.76) Noisy-Quiet 5.64 1.43 (4.68, 6.6) Uncluttered-Cluttered 5 1.26 (4.15, 5.85) Daylight Corrective Lenses n=8 t=2.365 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.63 1.51 (4.37, 6.88) Glare-No Glare 3.75 2.05 (2.03, 5.47) Dim-Bright 5.25 1.49 (4.01, 6.49) Unfocused-Focused 4.5 1.31 (3.41, 5.59) Inviting-Hostile 1.88 0.83 (1.18, 2.57) Relaxed-Tense 2 1.07 (1.11, 2.89) Dull-Interesting 5.25 1.16 (4.28, 6.22) Cool-Warm 4.5 1.69 (3.09, 5.91) Dislike-Like 6.13 0.83 (5.43, 6.82) Small-Large 5.5 0.93 (4.73, 6.27) Tight-Open 5.75 0.89 (5.01, 6.49) Simple-Complex 2.5 1.2 (1.5, 3.5) Private-Public 4.88 0.99 (4.05, 5.7) Informal-Formal 2.63 1.19 (1.63, 3.62) Noisy-Quiet 5.13 0.83 (4.43, 5.82) Uncluttered-Cluttered 4.75 1.28 (3.68, 5.82) 113

PAGE 114

Daylight None n=11 t=2.228 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.18 1.6 (4.11, 6.25) Glare-No Glare 4.73 1.27 (3.88, 5.58) Dim-Bright 4.45 1.57 (3.4, 5.5) Unfocused-Focused 4.73 1.56 (3.68, 5.78) Inviting-Hostile 3 1.41 (2.05, 3.95) Relaxed-Tense 2.73 0.79 (2.2, 3.26) Dull-Interesting 4.45 1.63 (3.36, 5.54) Cool-Warm 3.82 1.33 (2.93, 4.71) Dislike-Like 5.09 1.7 (3.95, 6.23) Small-Large 4.91 1.38 (3.98, 5.84) Tight-Open 5.09 1.76 (3.91, 6.27) Simple-Complex 3.5 1.21 (2.74, 4.36) Private-Public 5 2 (3.66, 6.34) Informal-Formal 3.45 1.37 (2.53, 4.37) Noisy-Quiet 5.55 1.04 (4.85, 6.25) Uncluttered-Cluttered 5 1 (4.33, 5.67) Factor Three: Different Light Levels Halogen 0-1.4 n=12 t=2.201 Rating Mean SD 95 % Confidence Interval Hazy-Clear 4.5 1.71 (3.4, 5.6) Glare-No Glare 5 1.73 (3.65, 6.35) Dim-Bright 2.75 1.38 (1.89, 3.61) Unfocused-Focused 5.58 1.44 (4.89, 6.27) Inviting-Hostile 3.92 1.06 (2.75, 5.09) Relaxed-Tense 3.58 1.73 (2.48, 4.68) Dull-Interesting 5.58 0.79 (5.08, 6.09) Cool-Warm 5.08 1.16 (4.34, 5.82) Dislike-Like 5.67 1.03 (4.51, 5.82) Small-Large 3.42 1.62 (2.39, 4.45) Tight-Open 3.58 1.08 (2.89, 4.27) Simple-Complex 4.08 1.24 (3.3, 4.87) Private-Public 2.75 1.91 (1.53, 3.97) Informal-Formal 4.33 1.97 (3.08, 5.58) Noisy-Quiet 5.67 1.44 (4.75, 6.58) Uncluttered-Cluttered 5 1.41 (4.1, 5.9) 114

PAGE 115

Halogen 1.5-7 n=7 t=2.447 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.57 1.62 (4.07, 7.07) Glare-No Glare 4.57 1.9 (2.81, 6.33) Dim-Bright 3.71 1.5 (2.33, 5.1) Unfocused-Focused 5.86 1.77 (4.22, 7.5) Inviting-Hostile 3 2 (1.15, 4.85) Relaxed-Tense 2.71 1.98 (0.89, 4.54) Dull-Interesting 4.86 1.21 (3.73, 5.98) Cool-Warm 5 1.41 (3.69, 6.31) Dislike-Like 6 1 (5.08, 6.92) Small-Large 3.86 1.35 (2.61, 5.1) Tight-Open 4.14 1.57 (2.69, 5.6) Simple-Complex 4.57 1.51 (3.17, 5.97) Private-Public 2.43 0.98 (1.53, 3.33) Informal-Formal 4.14 1.35 (2.9, 5.39) Noisy-Quiet 4.86 1.68 (3.31, 6.41) Uncluttered-Cluttered 5 1.53 (3.59, 6.41) Daylight 10-30 n=12 t=2.201 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.25 1.71 (4.16, 6.34) Glare-No Glare 4.5 1.73 (3.4, 5.6) Dim-Bright 4.42 1.38 (3.54, 5.3) Unfocused-Focused 4.33 1.44 (3.42, 5.24) Inviting-Hostile 2.25 1.06 (1.58, 2.92) Relaxed-Tense 2.09 0.94 (1.49, 2.69) Dull-Interesting 4.67 1.5 (3.72, 5.62) Cool-Warm 4.33 1.56 (3.34, 5.32) Dislike-Like 5.33 1.67 (4.27, 6.39) Small-Large 5.08 1.51 (4.12, 6.04) Tight-Open 5.33 1.83 (4.17, 6.49) Simple-Complex 3.08 1.24 (2.06, 4.1) Private-Public 4.75 1.6 (3.73, 5.77) Informal-Formal 3.25 1.54 (2.27, 4.23) Noisy-Quiet 5.25 0.87 (4.7, 5.8) Uncluttered-Cluttered 4.92 1.16 (4.18, 5.66) 115

PAGE 116

Daylight 60-100 n=7 t=2.447 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.57 1.27 (4.4, 6.74) Glare-No Glare 4 1.63 (2.49, 5.51) Dim-Bright 5.43 1.72 (3.84, 7.02) Unfocused-Focused 5.14 1.35 (3.89, 6.39) Inviting-Hostile 3 1.63 (1.49, 4.51) Relaxed-Tense 2.86 0.9 (2.03, 3.69) Dull-Interesting 5 1.53 (3.58, 6.42) Cool-Warm 3.71 1.38 (2.43, 4.99) Dislike-Like 5.86 1.07 (4.87, 6.85) Small-Large 5.29 0.49 (4.84, 5.74) Tight-Open 5.43 0.53 (4.94, 5.92) Simple-Complex 3.14 1.46 (1.79, 4.49) Private-Public 5.43 1.13 (4.38, 6.48) Informal-Formal 2.86 0.9 (2.03, 3.69) Noisy-Quiet 5.57 1.13 (4.52, 6.62) Uncluttered-Cluttered 5.14 0.69 (4.5, 5.78) Factor Four: Real vs. Rendered Environment Fluorescent Real n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.47 1.17 (4.91, 6.03) Glare-No Glare 6 1.41 (5.32, 6.68) Dim-Bright 4.89 1.2 (4.31, 5.37) Unfocused-Focused 5.26 1.79 (4.4, 6.12) Inviting-Hostile 3.58 1.39 (2.91, 4.25) Relaxed-Tense 3.37 1.26 (2.76, 3.98) Dull-Interesting 3.05 1.13 (2.51, 3.59) Cool-Warm 3.63 1.46 (2.93, 4.33) Dislike-Like 4.05 1.08 (3.53, 4.57) Small-Large 3.37 1.21 (2.79, 3.95) Tight-Open 3.37 1.42 (2.69, 4.05) Simple-Complex 3 1.25 (2.4, 3.6) Private-Public 3.21 1.62 (2.43, 3.99) Informal-Formal 4.16 1.8 (3.29, 5.03) Noisy-Quiet 5.16 1.54 (4.42, 5.9) Uncluttered-Cluttered 4.42 1.57 (3.66, 5.18) 116

PAGE 117

Fluorescent Rendered n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 5.47 1.35 (4.82, 6.12) Glare-No Glare 4.74 1.48 (4.03, 5.45) Dim-Bright 4.53 1.35 (3.88, 5.18) Unfocused-Focused 5 1.7 (4.18, 5.82) Inviting-Hostile 3.79 1.4 (3.12, 4.46) Relaxed-Tense 3.74 1.48 (3.03, 4.45) Dull-Interesting 3.37 1.38 (2.7, 4.04) Cool-Warm 3.79 1.47 (3.08, 4.5) Dislike-Like 4.53 1.12 (3.99, 5.07) Small-Large 4 1.56 (3.25, 4.75) Tight-Open 4.32 1.73 (3.49, 5.15) Simple-Complex 2.53 0.96 (2.07, 2.99) Private-Public 3.26 1.85 (2.37, 4.15) Informal-Formal 3.89 1.82 (3.01, 4.77) Noisy-Quiet 5.74 1.48 (5.03, 6.45) Uncluttered-Cluttered 6.05 1.75 (5.21, 6.89) Halogen Real n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 4.89 1.73 (4.06, 5.72) Glare-No Glare 4.84 2.01 (3.87, 5.81) Dim-Bright 3.11 1.45 (2.41, 3.81) Unfocused-Focused 5.68 1.34 (5.03, 6.33) Inviting-Hostile 3.58 1.89 (2.67, 4.49) Relaxed-Tense 3.26 1.82 (2.38, 4.14) Dull-Interesting 5.32 1 (4.84, 5.8) Cool-Warm 5.05 1.22 (4.46, 5.64) Dislike-Like 5.47 1.07 (4.95, 5.99) Small-Large 3.58 1.5 (2.86, 4.3) Tight-Open 3.79 1.27 (3.18, 4.4) Simple-Complex 4.26 1.33 (3.62, 4.9) Private-Public 2.63 1.6 (1.86, 3.4) Informal-Formal 4.26 1.73 (3.43, 5.09) Noisy-Quiet 5.32 1.54 (4.58, 6.06) Uncluttered-Cluttered 5 1.41 (4.32, 5.68) 117

PAGE 118

118 Halogen Rendered n=19 t=2.101 Rating Mean SD 95 % Confidence Interval Hazy-Clear 3.05 1.65 (2.25, 3.85) Glare-No Glare 5.74 1.48 (5.03, 6.45) Dim-Bright 1.79 0.85 (1.38, 2.2) Unfocused-Focused 3.47 1.68 (2.66, 4.28) Inviting-Hostile 3.47 2.04 (2.49, 4.45) Relaxed-Tense 3.11 1.76 (2.26, 3.96) Dull-Interesting 4.95 1.72 (4.12, 5.78) Cool-Warm 5.21 1.32 (4.57, 5.85) Dislike-Like 4.42 1.74 (3.58, 5.26) Small-Large 3.58 1.64 (2.79, 4.37) Tight-Open 3.84 1.42 (3.16, 4.52) Simple-Complex 4.42 1.61 (3.64, 5.2) Private-Public 2.42 1.26 (1.81, 3.03) Informal-Formal 3.68 1.57 (2.92, 4.44) Noisy-Quiet 5.21 1.55 (4.46, 5.96) Uncluttered-Cluttered 4.95 1.72 (4.12, 5.78)

PAGE 119

LIST OF REFERENCES 1 Plummer, Henry. Masters of Light. First Volume. Twentieth-Century Pioneers / Henry Plummer; Masters of Light. First Volume Twentieth-Century Pioneers / Henry Plummer. T ky, 2003. 2 Egan, M. David, Victor Ol gyay, and Victor M. Olgyay. Architectural Lighting 2nd ed. Boston: McGraw-Hill, 2002. 3 Stein, Benjamin. Mechanical and Electrical Equipment for Buildings 10th ed. Hoboken, N.J.: Wiley, 2006. 4 Boyce, P. R. Human Factors in Lighting. New York: Macmillan, 1981. 5 Snider, James G., Charles Egert on Osgood, and Charles Egerton Osgood. Semantic Differential Technique; a Sourcebook Chicago: Aldine-Atherton, 1969. 6 Flynn, Spencer, Martyniuk, Hendrick. IERI Project 92 the Effect of Light on Human Judgment and Behavior. University Park, PA: Illumination Laboratory, 1975. 7 Murdoch, Joseph B., and Carol C. Caughey. "J ohn Flynn and the Psychological Effects of Lighting." Lighting Design and Application August 2004 (2004): 69-73. 8 Steffy, Gary R. Architectural Lighting Design 2nd ed. New York: John Wiley, 2002. 9 Veitch JA. "Psychological processe s influencing lighting quality." Journal of the Illuminating Engineering Society 2001; 30: 124. 10 Zimmons, Paul M. The Influence of Lighting Quality on Presence and Task Performance in Virtual Environments (Doctor of Philosophy University of North Carolina at Chapel Hill, 2004). Retrieved from http://proquest .umi.com/pqdweb?did=775177031&sid=2&Fmt =2&clientId=20179&RQT=309&VName=PQD 11 Davis, Robert G. Cognitive and Perceptual Factors in Lighted Architectural Environments (Doctor of Philosophy University of Colora do, 2006). Retrieved from http://proquest.umi. com/pqdweb?did=1208145201&sid=1&Fmt= 2&clientId=20179&RQT=309&VName=PQ D 12 Agresti, Alan, and Christine A. Franklin. Statistics: The Art and Science of Learning from Data Upper Saddle River, N.J.: Pearson Prentice Hall, 2007. 13 Tanizaki, Junichir et al. In Praise of Shadows. New Haven, Conn.: Leete's Island Books, 1977. 119

PAGE 120

BIOGRAPHICAL SKETCH Kimberly Ann Gillar was born in 1985 in Sunrise, Florida. She is one of triplets and was raised in Sunrise. While attending Bair Mi ddle School, her technology teacher, Samuel Burton, sparked an interest in architecture. This inte rest extended into high sc hool where she took three levels of drafting. In 2003, she was accepted into the School of Architecture at the University of Florida. Throughout her undergraduate years, a growing awareness in architectural lighting influenced many of her designs. In 2007, she gra duated with a Bachelor of Design degree cum laude. Later that year she started graduate school in the School of Architecture in the College of Design, Construction and Planning at the Univers ity of Florida to re ceive her Master of Architecture degree in 2009. 120