IIIIII,, I ,I II III: II I III {III}, {$41,} It“, I,,,I I III III I II....I|II'I; III,,",,I,,..,, IIIIIII’ III III ,I, II III IIII " I H III I III II, :I,,,, IIII II): IIIIII I TN %fi_ “2“.— - _ s «a. r“- III. _.—.. g A. “w“ aw— . III I IIIIIIII 1m II IIIIIIIIIIIIII‘ ,1 IIIIII III! ,III I I III I, III III III III III, IIIIIIIIIIIIIIIIIIIII II,” III M,” I III ,III III, ’IIII‘I I II IIIIII III III I III II ‘- ' . . . I . ’ 'IfI III III III IIIIIIIII III, III I ' I I, IIIIII: f" I' ’; ‘I'I IIII‘II II ,‘IIII‘I I III IIII’IIII III ,I, If??? I I I; W, III III: II,,,I IIIII'1‘1‘II‘ . ,1 IIIII IIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIII’IIIIIII11I III III. II,‘ IIIIIIIII I , I : IIIIIIIIIIIII II I I II .. I ‘1 III: " ,;II II'I.1' III I‘ I » II IIIIII: II I,IIIIII"IIIIIIIIIIIIIII I IIIIIIII IIHIII IIIJIII i I III I III‘I!,,I,,II :-. , III III II II.,I III I II‘IIIIIIII IIIIIIIIIIII I III, II‘IIIIIIIIIIIII, 21%,”, ”,1 IIII,,, -. ..T ‘ III 3 III: I I III III} III II” HIIII IIH ,I' 3' I III, IIIII III III [',IIIIIIIIIII ,II,,I,,, III “III III. 1, I m ! IIIIIIII III I, II III: I‘IIIIIIII IIIIIIIIIIII,IIII III! I, IIIIII IIIII WI”: III'EI m”? ~'.:'r— v“ “w. "3'2?"~ III" ‘ «1":— fl w... ' «m "23-“- IIIIIII --—-. . -3 ”.05.-. - . 4. 1.. .5. III II fi- ‘5 “u” .- . .. - ov b a" A» T.‘ 0...... v ’ . “1 w a... ‘ v - , , .__, ' _ A -_ .- . a< , ‘ . I - ‘p,’ . ..:‘ >' I «a v 1-: .‘J: “4..- ,A .., '. ‘ Jun ‘1 .. . .‘3- ->. a w ' . r}.- “—1 xvii!” , "Iv." ' v - ~ ,t -- .. 1 .1- ‘I in”: ---_‘ 'T‘ . - . -- . .._ .1;th . , .-1. .-.. “— A. -1 1 .. "a 7 _ ”*7 ..1 ‘, --' 1 :- ~' 3.22%.; .. ~ m,»- <. «fir-1:2; W ‘4' 'C V A 4 l v v“ a. I V .. ... 1... ... ,z-."‘ _-.'.:. .1- 0"": .I .v 1 rlrmrr " III. III [III m; .— man's” web-.5:— .g‘. ow 4 IIIIIIIII I I III, IIIII’ IIIII IIIIII, I'IIIIIII ".5: 31 : .-a-w ..p f._:-; ;., '4” I... .— zip _. H... =_:._‘ -:':—‘ ..‘:: — ‘_..1 h... ‘40? III III; III! III III III“ IIIIIIIIII ,II,I,, IIIII IIIIIIII IIIIIIIIIIIII IIIIIIIIII IIIIIIIIII I,,,,,IIII‘ III I ; 33,5, IIIIIoII III III kw, '* I IIIIIII II ’1 I III III II,” ,,,,,, III III I II "II ,III . IIIIHI,, IIIII‘I'III ‘IIIIIIIIII I; III 1 III iHIIIIIIII‘IImIIIIII‘ II III’I'III'IIII I Mia IIIII-WII IIII‘E‘K I: I II III; II " {ti-13:; b n.) ITY LIBRA ES Illllllillllliljlll llllllll 3 1293 020 3 1465 memo»: STATE UNIV LIBRARIES MlCHlGAN STATE UNNERSITY EAST LANSlNG, MlGH. 48824 This is to certify that the thesis entitled A Study of the Fear of Crime Among Undergraduate Students at a Selected University presented by Dan M. Puuri has been accepted towards fulfillment of the requirements for Master of Science degree in Criminal Justice Major péfessor Date JULY 28; 1983 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES —_ RETURNING MATERIALS: Place in book drop to remove this Checkout from your record. FINES will be charged if book is returned after the date stamped below. *xr MAY (2 02100 y>;"fl-.LQ A STUDY OF THE FEAR OF CRIME AMONG UNDERGRADUATE STUDENTS AT A SELECTED UNIVERSITY By Dan M. Puuri A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Criminal Justice 1983 3’ / 3'9" -— u 5 ABSTRACT A STUDY OF THE FEAR OF CRIME AMONG UNDERGRADUATE STUDENTS AT A SELECTED UNIVERSITY By Dan M. Puuri The purpose of the study was to examine the fear of crime with a comprehensive approach. Numerous studies have examined the level of fear and who experiences the fear of crime in terms of personal charac- teristics. However, few have focused on the reasons why individuals have a fear of crime. Three surveys were conducted on the Michigan State University campus during 1979, 1980 and 1981. Data from the surveys were used to examine the fear of crime within the framework of three research ques- tions. For the three surveys, 1.6 percent of the respondents felt unsafe while out alone on campus during the day, compared to 54.3 percent of the reSpondents who felt unsafe at night. In the 1979 study, 86.4 per- cent of the females felt unsafe while out alone at night, compared to 22.3 percent of the males. Environmental characteristics were consist- ently cited as the most important reason for feeling unsafe on campus. This thesis is dedicated to my parents, PAUL and LUCILLE PUURI, who instilled within me the value of education -and- to my devoted wife, LORRAINE, who has given me support throughout the entire study. ii ACKNOWLEDGMENTS I am eSpecially indebted to Tim Bynum, my thesis committee chairman. His time, direction and efforts from the beginning of this project were instrumental in its completion. In terms of education, Tim expanded my knowledge concerning research and analysis. I gratefully thank my other committee members. Jack Greene provided insightful criticisms and recommendations during the "defense." Vince Hoffman, who was asked to participate in the final stage, provided sound guidance to several remaining issues. I would also like to acknowledge the suggestions and words of encouragement from Lou Radelet, Bob Trojanowicz and John Hudzik. Finally, a special thanks is given to Dennis Banas, who provided answers to many questions while I was in the Master's program. TABLE OF CONTENTS LIST OF TABLES .......................... Chapter I. THE PROBLEM ....................... Introduction ...................... Purpose and Research Questions ............. How Prevalent is the Fear of Crime? ....................... Who is Experiencing a Fear of Crime? ....................... What are the Determinants of the Fear of Crime? ................. Definitions ...................... Overview ........................ II. REVIEW OF RELATED LITERATURE ............... Overview ........................ How Prevalent is the Fear of Crime? ........................ Who is Experiencing the Fear of Crime? ...................... What are the Determinants of the Fear of Crime? .................. Direct Victimization ................. Indirect or Vicarious Victimization ................... Vulnerability .................... Formal Versus Informal Control ...................... Perception of Crime ................. Psychological .................... Media Effect ..................... Environment ..................... Summary ........................ III. RESEARCH PROCEDURES ................... Overview ........................ Research Approach ................... Research Design .................... Survey Site ...................... Samples and Population ................. iv \DLOO‘ ON Chapter Data Collection and Recording ............. 78 Analysis ........................ 79 Methodological Assumptions ............... 83 Limitations ...................... 84 Summary ........................ 85 IV. ANALYSIS OF THE DATA ................... 88 HQw_Prevalent is the Fear of Crime?. ........................ 88 Who is Experiencing the Fear of Crime? . .' .................... 92 What -are the De terminantscf * the. Fear of Crime? .................... 108 Summary' ........................ 122 V. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ......... 127 Summary ........................ .127 Purpose ....................... 127 Research Methodology ................. 128 Major Findings .................... 129 Conclusions ...................... 131 Recommendations for Future Research .......... 132 APPENDICES Appendix A. 1979 SURVEY FORMAT .................... 135 B. 1980 SURVEY FORMAT .................... 143 C. 1981 SURVEY FORMAT .................... 152 BIBLIOGRAPHY ........................... 161 .10 .11 .12 .13 .14 .15 .16 LIST OF TABLES Measures of Fear From the Harris and Gallup Polls . . . . Trends in Fear of Crime, 1965-1981 ........... Feeling of Personal Safety When Out Alone in Neighborhood at Night .................. Relationship Between Citizen Perceptions of Crime Seriousness and Crime Rates By Neighborhood ....... Relationship Between Citizen Perceptions of Danger at Night and Crime Rates By Neighborhood .......... Fear Levels for Cities and Neighborhoods ........ Feeling of Safety .................... Fear of Crime and Related Attitudes ........... Summary of the Most Significant Predictors of Fear of Crime Found to be in the Discriminant Analysis as Measured by the Change in RAO V ........... Fear of Crime by Age of the Respondent (In Percent) . . . . Percentage Distribution on Fear of Walking Alone at Night Within One Mile of Home for Sample and Relevant Subsamples, 1973 and 1974 ........... Beta and Beta-Squared Values for Six Explanatory Variables ........................ Estimated Rates of Fear of Crime Among Age, Race and Family Income Groups: Eight Impact Cities Aggregate, 1975 .......................... Neighborhood Safety at Night By Victimization Status Robbery Victimization and Fear ............. Victimization and the Fear of Crime ........... vi 14 15 16 19 20 22 24 26 28 29 3O 31 31 33 34 35 NNNN .17 .18 .19 .20 .21 .22 .23 .24 .25 .26 .27 Estimated Rates of Victimization and Fear of Crime Among Age, Sex, Race and Family Income Groups: Eight Impact Cities Aggregate, 1975 ....... Feeling of Personal Safety When Out Alone at Night with a Selected Experience with Victimization, Eight Impact Cities Aggregate .......... ReSponses By Elderly Residents of Flatbush and Sherman Park to Questions on Perceptions and Fear of Crime .................. Personal Crime: Direct and Indirect Rates of Victimization of the Elderly in the LEAA National and BSL Neighborhood Surveys ...... Women's Vulnerability By Fear Of Crime ..... What Makes Your Neighborhood Safe? ....... Crime Rates (Per 1,000 Residents) By Neighborhood Incivility and the Fear of Crime ........ The Relationship of Fear of Crime to Estimate of Neighborhood Safety .............. . Fear of Crime by Perceived Seriousness of Crime Relative to What the Newspapers and Television Say: Eight Impact Cities Aggregate, 1975 . . . . Reported Impacts Attributed to Street Lighting for Selected Cities ............... Distribution of Surveys and Response Rates (1979, 1980, 1981) ............... DemOgraphic Characteristics of Survey Respondents (1979, 1980, 1981) .................... Comparison of Survey Samples and Student Populations for the Variable Sex (1979, 1980, 1981) .................... Respondent's Perception of Safety During the Day (1979, 1980, 1981) ............... Respondent's Perception of Safety at Night (1979, 1980, 1981) .................... Respondents Expressing Some Degree of Unsafety for MSU Campus and Eight Impact Cities (1979, 1980, 1981) ........................... vii ..... 36 38 4O 41 44 46 47 48 50 55 57 74 76 77 89 89 91 .10 .11 .12 .13 .14 .15 .16 .17 .18 .19 Actions Taken as a Result of Perceptions of Unsafety (Combining Somewhat Unsafe and Very Unsafe), 1980 .................... Perception of Safety at Night by Sex (1979, 1980, 1981) .................... Perception of Safety at Night by Race of Respondent (1979, 1980, 1981) ............... Perception of Safety at Night by Class Level of Respondent (1979, 1980, 1981) ............... Perception of Safety at Night by Residence Complex for Males (1979, 1980, 1981) ............... Perception of Safety at Night by Residence Complex for Females (1979, 1980, 1981) .............. Respondent's Type of Criminal Victimization Experience (1979, 1980, 1981) ............... Perception of Safety at Night by Victimization Experience (1979, 1980, 1981) ............... Perception of Safety at Night by Size of Home Town for Females (1981) .................... Perception of Safety at Night by Size of Home Town for Males (1981) ..................... Perception of Safety at Night on Campus by Perception of Safety at Night in their Home Town for Females (1981) ............... Perception of Safety at Night on Campus by Perception of Safety at Night in their Home Town for Males (1981) ................ Most Important Reason for Feeling Somewhat Unsafe or Very Unsafe While Out Alone on MSU Campus for Female ReSpondents (1979, 1980, 1981) ........... Most Important Reason for Feeling Somewhat Unsafe or Very Unsafe While Out Alone on MSU Campus for Male Respondents (1979, 1980, 1981) ............ Modification of Pedestrian Lighting and Campus Landscape (1980) ..................... Perception of Safety at Night by Perceived Seriousness of Crime Relative to What the Media Say (1980) ..................... 91 94 96 98 100 102 103 104 105 106 107 109 109 112 114 4.20 4.21 4.22 4.23 Respondent's Perception of Safety at Night by Personal Crime Victimization Status (1980) ........ Respondent's Perception of Safety at Night by Knowing a Personal Crime Victim (1980) .......... Perception of Safety at Night by Social Cohesion Index ........... . ........... Perception of Safety at Night by Perception of Chances of Sexual Assault on Campus for Females (1980) ...................... ix 116 118 120 122 CHAPTER I THE PROBLEM Introduction Until very recently, research into the subject of the fear of crime or the feeling of susceptibility to crime has focused primarily upon the personal characteristics of the 'fear victim,‘ such as sex, race, age, socio-economic status and victimization experience. In addi- tion, the extent to which the fear of crime has permeated various sub- groups in society has been a major focus in research. While these foci are important in establishing that a societal problem concerning the fear of crime does exist, the resolution of this problem has not received the attention of concentrated research. The resolution of any problem must be accomplished by studying the characteristics and extent of the Specific problem, and the reasons why the problem exists. The study of the fear of crime, therefore, requires a comprehensive approach which considers the characteristics of the fear victim, the extent of the fear of crime problem, along with the reasons why individuals have a fear of crime. Upon establishing that the fear of crime is a societal problem, it is important to differentiate between the 'fear' of crime and the 'concern' for crime. By differentiating these two concepts, a working definition of the fear of crime can also be substantiated. The concepts of fear and concern have occasionally been used 2 interchangeably, although logical definitions and research have now dem- onstrated that the concepts are different. In developing an exploratory model of the causes and consequences of the fear of crime, James Garofalo, Director of Research Center East of the National Council on Crime and Delinquency, defined the fear of crime as "an emotiong] reacf i tion characterized by a sense of danger and anxiety."1 The sense of J danger and anxiety is considered a response to the threat of physical harm. While this definition may possibly imply that the physical harm or crime is imminent, this doesn't necessarily have to be the situation. As explicitly stated in the opening paragraph, the feeling of suscepti- bility to crime is included in the fear of crime concept. Furthermore, the latter concept will be expanded to include a specific environment in which individuals may feel susceptible to crime. By including the feeling of susceptibility, the fear of crime concept is expanded to include the following: ’1 1. An individual experiences danger and anxiety by observei; ing or experiencing a criminal act. 5 2. An individual experiences danger and anxiety by per- }Y ceiving conditions in a particular environment as pre— carious; conditions which could possibly result in the occurrence of a criminal act. The worry or concern for crime versus the fear of crime has been researched by Frank Furstenberg, Jr., who specialized in family studies, deviant behavior and evaluation research.2 In 1969, the Louis Harris Organization conducted a survey of public reaction to crime in Baltimore, Maryland. As indicated by Furstenberg, the initial results indicated that crime was generally regarded by the respondents as a serious danger to society, but not a personal threat. The data indicate that 89 percent of the respondents believed crime had increased during 3 1968 in the United States, and 80 percent thought crime had risen in Baltimore. When the survey question regarding the fluctuation of the crime rate was directed to the respondent's own neighborhood, only 39 percent felt that crime had risen in their neighborhood over the past year. This wide discrepancy can lead us to assume that the concern for crime in areas outside the neighborhood may be greater than the actual fear of crime in the neighborhood. In Furstenberg's own words, - the “fear'ofmcrimé is usually measured by a person's perception of his L , \ own chances of victimization, and concern by his estimate of the seri- ousness of the crime situation in this country."3 The concern for crime can also be viewed as varying when ranked among other domestic problems. In 1968, the Gallup Poll indicated the concern for crime, labeled in the polls as crime and lawlessness, was ranked second among the most important problems facing the United States.4 One decade later, the concern for crime was displaced by other concerns. While only 3 to 4 percent of the Gallup Poll respondents perceived crime and lawlessness as the most important problem, inflation (33 percent), energy (29 percent), unemployment (17 percent), and inter- national problems (17 percent) were perceived as more important.5 Whereas the concern for crime may fluctuate, national-level data has indicated that the fear of crime has steadily increased in the last seventeen years. In response to a Gallup Poll question, "Is there any area around here--that is, within a mile--where you would be afraid to walk alone at night?" 41 percent answered in the affirmative in 1973, compared to 34 percent in 1965.6 TheiLaw Enforcement Assistance Administration's public opinion surveys 66 crime used a different survey question to gather information concerning the extent or level of the 4 fear of crime on a national level. In thirteen selected cities, res- pondents were asked the question, "How safe do you feel being out alone in your neighborhood at night?" Over all, the percentage of respondents who answered "somewhat unsafe" or "very unsafe" ranged from a low of 36 percent in Dallas to a high of 58 percent in Newark.7 These general observations enable us to determine that the fear of crime does exist and is conceptually different from the concern for crime. Unfortunately, data at the national level provides little infor- mation concerning policy decision-making at the local level. Because of the wide discrepancy between an individual's perception of crime nationally versus their perception of crime in their neighborhood, fur- ther analysis at the neighborhood level is needed before any strategies to deal with the fear of crime can be undertaken. Marlys McPherson of the Minnesota Crime Prevention Center noted the importance of research at the neighborhood level when she stated: The use of national data, statistics and research results to des- cribe 'the crime problem' belies the critical fact that crime occurs in neighborhoods. Strategies to control crime, as well as to make the residents of a particular area feel safer, must be developed at the local level because crime . . . as well as people's fear levels vary considerably from one neighborhood to another. Nejther national, no: statewide, nor even,citywideuggta can apprOpriately 'EUBstTTUte foFTfUFTFEr analysis at the_neighborh level in pre- paring localized strategies to deal with the problem. Following this line of reason, the purpose and various questions to be investigated by this research can be set forth. Purpose and Research Questions The primary purpose of this research is to examine a specific population residing in a limited environment with regard to their fear of crime. Due to the need for analysis at the neighborhood level, the 5 student population residing on the Michigan State University campus was chosen as the population. In reference to the fear of crime as the major focus, the following three research questions are examined in this study: 1. How prevalent is the fear of crime among the students resid- ing on the campus of Michigan State University? 2. Who is experiencing a fear of crime? (In terms of personal characteristics) 3. What are the determinants of the fear of crime? The nature of these questions indicate the research approach will be primarily descriptive. This approach will systematically des- cribe the facts and characteristics of the population and area of interest, factually and accurately.9 A discussion of each of the three questions will clarify the objectives of this research. How Prevalent is the Fear of Crime? In order to determine if the fear of crime in a Specific popula- tion is a 'problem,' it can be helpful to determine what proportion of the population is experiencing the fear of crime. However, the determination as to what affected pr0portion constitutes a societal problem is quite subjective. It is assumed, though, that those expe- riencing the fear of crime would consider it to be a societal problem. When identifying the problem, it is also helpful to make comparisons between the results obtained from this research with other populations that have been studied. This comparative analysis can determine if the proportion of campus residents experiencing a fear of crime is rel- atively greater or less than in other populations. For this study, information will be collected concerning how safe the students feel 6 when out alone on campus during the day and night. Those students res- ponding "somewhat unsafe” or "very unsafe" will be considered the sub- population that is experiencing the fear of crime. Who is Experiencing a Fear of Crime? This section also contributes to the identification of the prob- lem. Most of the research on the fear of crime has been concerned with a descriptive analysis of the fear victims. The personal character- istics of the fear victims have primarily included sex, race, age, and socio-economic status (income and education).10 In addition to these characteristics, the victimization experi- ence of an individual is sometimes assumed to be a strong predictor of fear. While a large pr0portion of crime victims may have a fear of crime, the number of crime victims is relatively few compared to the number of individuals who have a fear of crime. Although various commentators have described the public's fear of crime as irrational, this label does not make the problem any less real. When the fear of crime is pr0portionately greater than the probability of being victim- ized, this simply demonstrates the apparent need for analysis into the reasons why peOple have a fear of crime. Finally, the community size in which an individual resides has been associated to the fear of crime. Higher levels of fear of crime have been reported in 'urban' areas as compared to 'rural' areas. What are the Determinants of the Fear of Crime? The determinants of the fear of crime are the major focal points for policy-makers in attempting to resolve the problem. In reviewing the existing literature, a variety of potential determinants have been 7 explored. The following eight determinants have been isolated for this study, and a brief discussion of each will follow. Direct Victimization. This factor implies that the individual's fear of crime will be directly related to the criminal victimization experience of the individual. For this study, information will be col- lected concerning the respondent's personal and property crime victimi- zation experience. The personal and property crime victims will be com- pared with nonvictims in regard to the extent of fear for each group. Indirect or Vicarious Victimization. This factor contributes to the 'ripple effect' in which individuals become aware of criminal events that have happened to their friends or acquaintances. Media Effect. The reporting of criminal events in the news media is su5pected of influencing an individual's fear of crime. This is related to indirect victimization, although an individual becomes aware of campus criminal victimization occurring to other peOple by what is read in newspapers and heard on radio or television. Psychological. The psychological approach has often concentrated on the fear of strangers as being a correlate of fear.11 As Henig and Maxfield point out, this psychological approach would apply more to the presence of strangers in residential neighborhoods instead of shapping and business areas where the presence of strangers is normal. Because association may be hard to establish between the fear of strangers and the fear of crime, this study will define the psychological approach differently. This study will attempt to identify those individuals who feel unsafe anywhere and at any time. Perception of Extent of Crime. The psychological approach to fear or the neurosis of fear does not apply when an individual perceives 8 that a lot of crime is occurring in the environment in which they reside. Regardless of the objective crime rate, individuals have per- ceptions of the crime rate and this perception will vary among individ- uals.12 While information was collected concerning the victimization rate on campus, it is the perception of the crime rate that may induce fear. Even though the objective crime rate may be relatively low in the environment compared to other environments, individuals may still perceive the crime rate as being high or a risk to their personal well- being. Vulnerability. This factor refers to the perception by indi- viduals that they are unable to defend themselves against a criminal act. Some individuals may be aware of the criminal element in the environment, but may not fear it because of their physical character- istics or possession of protective arms. Other individuals may or may not perceive the environment as crime ridden, but are still fearful because they perceive themselves as physically vulnerable. Inadequate Formal Control. The reliance on formal controls or official barriers versus informal controls for providing a 'safe com- munity' has been studied by a number of people.13 The formal control refers to law enforcement, whereas the informal controls have been referred to the resident's character, community character and social network.14 Existing research indicates that more reliance for safety is put on the informal controls than on formal law enforcement. Environment. The environment refers to the physical surroundings in which an individual resides. Various physical characteristics of the Michigan State University campus may be associated to the fear of crime. Examples of environmental characteristics are building design, 9 inadequate lighting, and excessive shrubbery, bushes and pillars that create blind spots. Definitions The Fear Victim is defined as an individual who experiences a fear of crime. The fear of crime can result from various factors such as criminal victimization experience, perceiving conditions or cues in the environment as precarious, knowledge and perceptions of objective crime occurrences, and psychological and physical characteristics of an individual. The 'feeling of unsafety' due to crime is used inter- changeably with the fear of crime concept. The Personal Crime Victim is an individual who has experienced one or more incidents of robbery, attempted robbery, physical attack or sexual offense (rape, molestation or exposure). The Property Crime Victim is an individual who has been a victim of theft. The categories of theftincludesthefts from dormitory rooms, thefts from vehicles and thefts from other campus locations (bike racks, laundry rooms, classrooms). may. This study is concerned with examining a Specific p0pulation with regard to their fear of crime. Three research questions have been iden- tified for this study. By providing answers to the questions, a thor- ough understanding of the fear of crime on the Michigan State University campus will result. A review of related literature is presented in the next chapter. The review of literature is presented within the framework of the research questions, and will fully explain the research questions as 10 they are used in this study. In Chapter III the research procedures are discussed. In addi- tion to the description of the research design and approach, sections concerning the samples and population, data collection and recording, data analysis and limitations of the study are presented. Analyses of the data are presented in Chapter IV. Finally, Chapter V provides a discussion of the summary, conclusions and recom- mendations for future research. CHAPTER I FOOTNOTES 1James Garofalo, "The Fear of Crime: Causes and Consequences," The Journal of Criminal Law and Criminology 72 (Summer 1981): 840. 2Frank F. Furstenberg, Jr., "Public Reaction to Crime in the Streets," American Scholar, Vol. 40 (1970/71). 31bid.. p. 603. 4Hazel Erskine, "The Polls: Fear of Violence and Crime," Public Opinion Quarterly, Vol. 38 (Spring 1974). 5George H. Gallup, The Gallup Poll: Public Opinion 1978 (Wilmington, DE: Scholarly Resources Inc., 1979). 6Erskine, "The Polls." 7James Garofalo, Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-l, 1977). 8Marlys McPherson, "Realities and Perceptions of Crime at the Neighborhood Level," Victimology: An International Journal 3 (1978): 320. 9Stephen Isaac and William B. Michael, Handbook in Research and Evaluation (San Diego: EDITS, 1971), p. 18. 10Frank Clemente and Michael B. Kleiman, "Fear of Crime in the United States: A Multivariate Analysis," Social Forces, Vol. 56 (September-June 1977/78). 11Jeffrey Henig and Michael G. Maxfield, "Reducing Fear of Crime: Strategies for Intervention,“ Victimology: An International Journal, Vol. 3 (1978). 11 12 12John E. Conklin, The Impact of Crime (New York: Macmillan, 1976); Young Ja Kim, ”The Social Correlate of Perceptions of Neighbor- hood Crime Problems and Fear of Victimization," Working Paper, Reactions to Crime Project, Center for Urban Affairs, Northwestern University, 1976; Michael J. Hindelang, Michael R. Gottfredson, and James Garofalo, Victims of Personal Crime (Cambridge, MA: Ballinger, 1978). 13Jane Jacobs, The Death and Life of Great American Cities (New York: Vintage Books, 1961); L. Bickman, S. K. Green, J. Edwards, S. Shane-DuBow, P. J. Lavrakas, N. North-Walker, and S. Borkowski, "Towards Increasing Citizen Responsibility, Surveillance, and Reporting of Crimes,“ Prepared for NILECJ, Law Enforcement Assistance Administration, Department of Justice; Sarah L. Boggs, "Formal and Informal Crime Control: An Exploratory Study of Urban, Suburban, and Rural Orienta- tions," The Sociological Quarterly 12 (Summer 1971): 319-327. 14Boggs, "Crime Control." CHAPTER II REVIEW OF RELATED LITERATURE 0v_e:_v1e_w The purpose of this chapter is to review the literature pertain- ing to the research questions that were briefly discussed in Chapter I. This review will fully explain the research questions as they are used in this study. The research questions are: 1. How prevalent is the fear of crime? 2. Who is experiencing the fear of crime? 3. What are the determinants of the fear of crime? How Prevalent is the Fear of Crime? The prevalency or level of fear has been examined mainly by national and urban public Opinion polls. In addition to reported crime --. 3 rates, the level offearhas been used as an indicator of the extent of the crime problem. Several measures of fear of crime used by the Harris and Gallup Polls as presented in Table 2.1, appear to indicate that the national level of fear has fluctuated but has generally increased since the middle sixties.1 As indicated by Skogan and Maxfield, the increase in fear during the later 1960s and early 19705 closely matches the 'crime wave' that occurred during the same time period.2 This general observation indicates that the objective crime rate and level of fear may be related. An increase in reported crime during the late '605 and early '705 is marked by an increase in measures 13 14 of fear. As indicated in Table 2.1, 49 percent of the respondents from a Harris poll felt more uneasy in 1966 than in the previous year. In addition, a 1967 Harris poll indicated that 46 percent of the respond- ents felt that crime was increasing over the previous year. During the early 19705, a range of 53 to 65 percent of the respondents felt more uneasy than in a previous year. During the 19705, the percentage of respondents who felt crime was increasing ranged from a low of 48 per- cent in 1973 to a high of 70 percent in 1975. Table 2.1.--Measures of Fear From the Harris and Gallup Polls % Feeling % Feeling % Reporting More Uneasy that Crime More Crime Years than a Year is Increasing in Area than Ago1 Over the a Year Ago2 Previous Year1 1966 49 1967 46 1968 53 1969 55 1970 65 62 1971 55 1972 51 1973 51 48 1975 55 7O 50 1977 49 58 1Harris Poll aniiup Poll A frequently used measure of fear of crime used by the Gallup Organization asks individuals if they are afraid to walk alone at night in a nearby area.3 An analysis of data extending from 1965 to 1981 indicates the previously mentioned observation, that the national level of fear had increased but has leveled off since the middle seventies. Table 2.2 illustrates this trend in fear of crime from 1965 to 1981. 15 Thirty-four percent of the poll respondents were afraid to walk at night in 1965, compared to forty-five percent in 1974. Since 1974, the per- centage has remained at about forty-five. Table 2.2.-—Trends in Fear of Crime, 1965-1981 45-h 40-- 35..- 30.. 2544 20-- 15-- 10-- o I l l J J L l 1 l J J l 1 J 1 l I 4 65 6% 67 at 65 7b 73 7% i3 it 75 f6 77 f5 75 15b 51 Adapted from Wesley G. SkOQan, "On Attitudes and Behaviors," Reactions to Crime (Beverly Hills, Sage Publications, 1981), p. 28. At the urban level, the most widely analyzed data pertaining to the fear of crime was collected in a series of surveys known as the National Crime Survey (NCS).4 Conducted by the Bureau of Census for the Law EnforCEment Assistance Administration, the NCS consisted of a victimization survey and an attitudinal survey. In addition to the eight large cities participating in LEAA's High Impact Crime Reduction Program, the nation's five largest cities were selected for the NCS. To measure the personal feelings of safety of the reSpondents in their own neighborhood, the question "How safe do you feel or would you feel 16 being out alone in your neighborhood at night?" was asked. The percent- age of reSponses to this question from each of the cities indicates that the responses vary between cities. It can be seen in Table 2.3 that a low of 36 percent of the respondents in Dallas and a high of 58 percent of the re5pondents in Newark felt somewhat unsafe or very unsafe. Table 2.3.--Feeling of Personal Safety When Out Alone in Neighborhood at Night Very Safe Reasonably Safe Somewhat Unsafe Very Unsafe Percent Percent Percent Percent Atlanta 17 36 22 24 Baltimore 14 34 22 29 Cleveland 12 38 21 28 Dallas 23 40 19 17 Denver 24 39 20 17 Newark 11 30 24 34 Portland 25 38 21 17 St. Louis 14 38 21 26 Chicago 15 39 24 21 Detroit 11 34 24 28 Los Angeles 16 39 23 20 New York 12 38 25 23 Philadelphia 15 37 22 25 Adapted from James Garofalo, Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-I, 1977). The use of aggregate data as presented in Table 2.3 provides for general observations of the differences in fear of crime between city residents in large cities, but tells little of the differences in fear within those cities. Cities are commonly broken down into com- munities or neighborhoods, and various physical and social character- istics of each community can affect the level of fear. One particular study by John E. Conklin consists of an analysis of fear in two Boston 17 communities.5 The two communities, Port City and Belleville, were sel- ected mainly because of the difference in their reported crime rates. While Belleville was reported to have a relatively low crime rate, Port City had nine times the number of crimes against persons than in Belle- ville.6 A systematic sample of two hundred residents over the age of twenty was selected from each community. One hundred thirty-eight resi- dents were interviewed in Belleville and 128 residents were interviewed in Port City. To measure the fear of crime, the residents were asked the ques- tion ”15 there any area around here--that is, within a mile--where you would be afraid to walk alone at night?" In Belleville, 33 percent answered this question in the affirmative, whereas 52 percent expressed fear in Port City. These data are important in demonstrating that the use of aggregate data on the city or national level may hide the fact that fear may not be distributed evenly throughout an urban area. A more recent study by McPherson supports the same finding, that fear of crime can vary within cities and that citizens can have a quite accurate perception of the seriousness of crime in their neighborhoods.7 Furthermore, the rates for all types of personal crime were found to be significantly correlated with the citizens' perceptions of danger. The data for her study was originally collected by a team of researchers that were defining the crime problem in Minneapolis prior to developing a comprehensive crime prevention program.8 Data pertaining to percep- tions of neighborhood crime, fear of victimization and actual victimiza- tion over the previous year was collected in an attitudinal survey.9 The crime rates (number per 100,000 population) for residential bur- glary, personal robbery, purse snatch and pickpocket, stranger-to- 18 stranger assault, sexual assault or rape and vandalism were calculated for each neighborhood. These crime measurements are considered as the independent variables, whereas individual perceptions and fear of crime are the dependent variables. For the neighborhoods in Minneapolis, the percentage of reSpondents who felt it was dangerous to walk through their neighborhood at night ranged from a low of 5.1 percent in Univer- sity to a high of 29 percent in Central (see Table 2.5). In McPherson's article, Pearson Product-Moment correlations (r)_‘ii were presented to show the relationship between the reported crime rates and the citizens' perception of the seriousness of the selected crimes. The 'seriousness of the crime' was measured by combining the responses "a big problem" and “some problem." As indicated in Table 2.4, all of the selected crimes were significantly related with the citizens' perceptions of the seriousness of the selected crime. Generally, the neighborhood which had the highest crime rate for each crime type also had the highest percentage of respondents who indicated that the par- ticular crime was a 'problem.' For example, the Near North neighborhood had the highest rate (per 100,000 population) of residential burglary (534.1) and the highest percentage of respondents who perceived burglary as a neighborhood problem (81.9 percent). Conversely, the Longfellow and Northeast neighborhoods had the lowest burglary rates and the lowest percentage of respondents indicating that burglary was a problem. The citizens' fear of crime was measured by asking the respondent how dangerous they felt various situations were. With a scale from zero (no danger) to ten (very dangerous), the t0p three points on the scale (8, 9, 10) were combined to represent the citizens' fear of walk- ing through their neighborhood at night. Table 2.5 shows that the .mmm .a .Amumfiv m .po> ._mcL:oa .m:o_um=me:~ =< "amoposwuuw> =.Pw>m4 woosLon;m_mz one an mEpLu Lo mcowpnmULma ccm mm_p_.mmm= .comLogauz mxmez soLL umuqmu< 83 83 83 So 83 :53 853:8; ALV =o_pm—mLLou efiow. u L mmmw. u L mafia. u L emom. u L mean. u L ucwsozuauauoLm commea um.- H.~m ao.m~ o.ooH x¢.mfi m.m~ x¢.mH e.eo~ no.mm H.H~ ummmzuLoz am.- N.om a~.o~ m.mm aH.mN o.Hm RH.mm “.mN nm.mm m.mo~ m_soxoz am.m~ “.me nw.- H.-~ xH.NN o.m NH.~N H.Hm a¢.¢e H.mm zo__mem:og xm.~o ~.omH nm.- m.wo~ am.o~ o.oH um.c~ ¢.Nm ae.- o.~HH ABPmLm>w== am.m~ o.o~ x~.mH o.m¢ u~.o~ ~.oH RH.oH m.om ue.mm ~.omH ummzzuaom mu xo.m~ m.o~ am.m~ m.m- am.o~ o.¢a um.o~ o.mm no.mo ~.m¢~ :mcEmu am.oe o.-~ N~.om m.omH n~.~m m.- RH.~m w.~o~ um.m~ ~.mm~ mmmeucaonpmu am.mm o.o- u~.mm m.mm~.~ Ro.m~ m.¢HN Ro.m~ ~.~m¢.fi aw.wm ¢.m~m FmLacmu x~.m¢ ~.m- xm.~m o.mw~ xo.mm m.mo fio.mm ¢.m~¢ a~.wo m.-m cLoszuzoa xm.om e.o- ao.~e o.em~ um.~m o.om am.~m o.m~e am.~w H.mm¢ cuLoz L$2 EmpnoLa Aooo.oofl Em—noLa Aooo.ooH Em_noLa Aooo.ooH EmpnoLa Aooo.ooH Em_noLa Aooo.ooH mP—oamm::_z =_ m Lmav m u_:mm Lmav m men Lmav m aLwa Lmav m me_m Lwav muoogLonnumz mama as? mama -m< new menu unom mew mama -aom ac_ apex -Lam mew mama . lucoammm luconmmm nucoammm -ucoqmmm nucoammm a x n R a Amaomv upzmmm< mecmLam amxuoaxuwa menaox pmcomLma amemLam upsmmm< —o=xmm -ouumecmLum \gUpmcm mmLaa _mwu=mu_mom voogLon;m_mz Am mmuom me_Lu ucm mmocm=o_me ms_Lu Lo mcowuamome :mNLHPu cmmzumm n_;m:opuupm¢--.e.~ manh .Nmm .a .AmRmHV m ._o> .LeeLsoa Le=OLSeeLaS=L eq "auopos_uuw> =.~m>m4 noozLonzmwmz wcp pm mewLu Lo mzowuamULma use mm_u_~mw¢= .comegau: mame: EoLL umuamu< 20 mo.w Lo.w Ho.w Ho.w Fo>o4 ooeeoLLL=mLm ALL :o_uwmeLou ammo. u L wowm. u L Nomm. n L mfimw. u L acmEozuauauoLa :ommea RR.R ~.~m RR.R o.oo~ RR.R m.mH RR.R ¢.eoH ummmnwLoz Re.m ~.om R¢.o m.mR Re.o m.- Re.o R.mR m_soxoz R~.m H.~e R~.m ~.RHH Rm.m o.m Rm.m ~.Hm zeppmemcog RH.m N.om~ RH.m m.mo~ RH.m m.oH RH.m e.RR >a_mLm>_:: RH.R o.o~ RH.R o.me RH.R R.o~ RH.R w.om ummzzgaom Rm.wH m.o~ Rm.m~ m.mHH Rm.wH o.e~ Rm.wH m.mm cmcsmu Rm.HH o.-~ Rm.H~ R.om~ Rm.~H m.- Rw.~H m.Ro~ mmpmmuczo;_ou Ro.m~ o.m- Ro.m~ m.RoN.~ Ro.mm m.e- Ro.m~ R.Rm¢.~ FmLucmu Re.om “.mmfi Re.o~ o.mm~ R<.om m.mm R¢.o~ e.m~e :Losznzoa Rm.- ¢.eHH Rm.- o.ew~ Rm.- o.ow Rm.~m o.mme guLoz Lmmz o;m_z Se “coo.oofi SemLz Se Aooo.ooH S;m_z Se Acoo.ooH Sem_z he Loco.oofi mL_oaeoeeLz =L maomecmo mev maomecmo mev mzomecmo Lmav maomecmo Lmav muoogLongmwmz mcwucoammz mama mcwccoammm mung mcwuconmmm mgmm acwucoqmmm mama R R R R Amnmmv upammm< pmxuoaxowa xLonnom chomLma u_:mmm< ~u=xmm mecmLumnou-me=mLum \cupmcm mmLza uoosLonsmmmz am mmumm meRLu can usuwz um mecmo Lo mcoppamULma :mNRRRU cmmzumm awzmcopumpmmuu.m.~ mpnmh 21 citizens' fear of crime was significantly correlated with the selected crimes presented. Neighborhoods which had the highest crime rates also had the highest levels of fear. Central, with the highest rates of personal robbery, purse snatch/pickpocket, assault and sexual assault, had the highest percentage of respondents who felt it was dangerous to walk through their neighborhood at night. It is also interesting to note that neighborhoods such as Southwest, University, Longfellow, Nokomis and Northeast had the lowest crime rates relative to the other neighborhoods. Corresponding with the low crime rates for these neigh- borhoods were relatively low levels of fear of crime. A study by Skogan and Maxfield provides additional support that there can be important differences in levels of fear within cities.10 Ten neighborhoods within the cities of San Francisco, Chicago and Phila- delphia were selected for this study that examined the relationships among crime, fear of crime and things that people do in response to crime. The neighborhoods were selected primarily on the basis of being predominantly black neighborhoods, racially heterogenous neighborhoods or predominantly white neighborhoods.11 From each neighborhood, random digit dialing was used to select a sample which ranged in size from 200 to 450 respondents. In reference to the fear of crime, the respondents were asked, "How safe do you feel, or would you feel, being out alone in your neighborhood at night--very safe, somewhat safe, somewhat unsafe, or very unsafe?" It can be seen in Table 2.6 that the average scores for the three cities do not vary much with each other. However, when comparing the neighborhoods within cities, substantial differences are apparent. Woodlawn and Wicker Park had the highest average scores within Chicago, and Lincoln Park had one of the lowest average scores. 22 Table 2.6.--Fear Levels for Cities and Neighborhoods Somewhat __ Unsafe 3’0 - Wicker Park (Chicago) -*- - Woodlawn (Chicago) -- Visitacion Valley (San Francisco) -"' 2.5 — Mission (San Francisco) Chicago -- , , F---West Philadelphia (Philadelphia) -- . . )— Back of the Yards (Chicago) Philadelphia ‘— Sunset (San Francisco) San Francisco --- Lincoln Park (Chicago) )— Logan (Philadelphia) Somewhat __ 2.0 Safe - South Philadelphia (Philadelphia) Very Safe db- 1.0 Adapted from Wesley G. Skogan and Michael G. Maxfield, Caping With Crime (Beverly Hills: Sage Publications, 1981): p. 54. 23 Within the city of San Francisco, a noticeable difference also exists between the neighborhoods of Visitacion Valley and Sunset. According to Skogan and Maxfield, when taking into account the sampling variation, the differences between the average scores of the neighborhoods are highly significant statistically, while the city-level scores are vir- tually identical. It is also interesting to note that four neighborhoods with the lowest average scores were characterized as predominantly white neigh- borhoods. The neighborhoods were described as being a working-class Italian community (South Philadelphia), an Irish and Eastern European community (Back of the Yards), an affluent area consisting of young and white residents (Lincoln Park) and a white, middle-class, homeowner neighborhood (Sunset).~ The Logan neighborhood was described as being ethnically diverse, with low family incomes and a substantial amount of unemployment. The three neighborhoods which were ranked as the most unsafe were Wicker Park, Woodlawn, and Visitacion Valley. Woodlawn was referred to as a classic ghetto slum, whereas Wicker Park and Visitacion Valley were heterogenous neighborhoods consisting of whites, blacks and hispanics. In 1971, Sarah L. 80995 of the University of Missouri reported that urban dwellers were more fearful than residents of suburban and rural areas.12 The data for her study was obtained by the University of Missouri Public Opinion Survey Unit during the Spring of 1968. A total of 842 respondents were selected for the study: 270 central city residents, 212 suburban residents and 360 small town and rural resi- dents. Among a series of questions concerning neighborhood crime and its deterrence, the respondents were asked: "How safe from crime would 24 you say your neighborhood is--very safe, safe, unsafe, very unsafe, or don't know?" Presented in Table 2.7 are the percentages of responses elicited by the 'feeling of safety' question. Readily apparent in Table 2.7 is the large percentage of reSpondents feeling very safe and safe in rural/small towns (95.8) and suburbs (95.3), compared to the percentage of central city respondents (75.2). Conversely, more central city res- pondents (24.1 percent) indicated they felt unsafe or very unsafe. Table 2.7.--Feeling of Safety Rural/Small Central Total Safety Town Suburbs City State Very Safe 33.3 34.0 17.8 28.5 Safe 62.5 61.3 57.4 60.6 Unsafe 1.1 2.8 13.7 5.6 Very Unsafe 2.0 1.4 10.4 4.5 Don't Know or Not Ascertained 1.1 .5 .7 .8 Total 160—6 153.7, 135 m Adapted from Sarah L. Boggs, "Formal and Informal Crime Control: An Exploratory Study of Urban, Suburban and Rural Orientations," The Socio- logical Quarterly 12 (Summer 1971): p. 324. In the discussion thus far, a variety of survey questions have been stated which are presumed to measure the fear of crime. It is a concern as to whether the survey questions are actually measuring the fear of crime, versus some other trait of the reSpondent. Garofalo discusses the potential drawbacks of the indicator that was used by the NCS.13 The major concern is that the question does not directly 25 link the 'feeling of unsafety' or fear to crime. Because the NCS did immediately precede the attitudinal survey with a victimization survey, it is unlikely that the respondent's perceptions of fear were concerned with anything else other than crime. While this statement is still an assumption, Skogan and Maxfield provide quantitative support that the fear indicator does measure the fear of crime.14 Skogan_and Maxfield re-examined cumulative data from the General Social Surveys of public opinion that were conducted between 1973 and 1977 by the National Opinion Research Center. In these surveys, the following ques- tion was asked: "Is there any area right around here--that is, within one mile-~where you would be afraid to walk alone at night?" This question, being very similar to the question used by the NCS, was cor- related with other questions that supposedly measured the reSpondents' distrust, suSpicion and anxiety about change. The data presented in Table 2.8 was computed by Skogan and Maxfield. The multiple R25 indi- cate that the fear related question is very weakly correlated with the other indicators. A conclusion drawn by the authors is that the mea- sure of fear is independent of the other measures listed in the follow- ing table. Who is Experiencing the Fear of Crime? This section is limited to the description of who is afraid in terms of demographic or personal characteristics. This discussion will focus upon the sex, age, race, income, and education of the fear victim, along with the city size of the individual's residence. While the vic- timization experience has often been used to describe who is afraid, this discussion will be included in the next section dealing with the 26 Table 2.8.--Fear of Crime and Related Attitudes Survey Questions Multiple R2 With All Other Items Is there any area right around here--that is, within 1 mile-- where you would be afraid to walk alone at night? Generally speaking, would you say that most pe0ple can be trusted, or that you can't be too careful in dealing with peOple? Would you say that most of the time pe0ple try to be helpful, or that they are mostly just looking out for themselves? Do you think most pe0ple would try to take advantage of you if they got a chance, or would they try to be fair? In spite of what some pe0ple say, the lot of the average man is getting worse, not better. Most people don't really care what happens to the next fellow. These days a person doesn't really know whom he can count on. (Number of Cases) .02 .34 .34 .31 .18 .34 .24 (2807) Adapted from Wesley G. Skogan and Michael G. Maxfield, C0ping with Crime (Beverly Hills: Sage Publications, 1981). 27 determinants of fear. The most powerful predictor of fear of crime has often been attributed to the sex of the individual. Aggregate data of the eight impact cities reveals wide margins between males' and females' feeling of personal safety at night.15 While 27 percent of the males felt some- what unsafe or very unsafe, 59 percent of the females felt unsafe. In a recent study by Ronald W. Toseland, twenty independent variables were divided into three categories: demographic, psychosocial and crime-related.16 The data for his study were obtained from the 1976 General Social Survey. Discriminant analysis was used to identify the independent variables that were significantly correlated with the fear of crime. Table 2.9, adapted from Toseland's article, indicates that sex was the most powerful predictor of the respondent's fear of crime. In the study, 60.9 percent of the females were fearful, compared to 23.1 percent of the males. In the analysis, twelve demographic, psychosocial and crime— related variables were found to be significantly related to the fear of crime. As presented in Table 2.9, sex_was the most important pre— dictor, with a standardized discriminant coefficient of .83593. As indicated in Table 2.9, size of place was also significantly related to fear of crime. Approximately 60 percent of the respondents residing in large cities were fearful, compared to 50 percent of the residents in small cities. The percentage of respondents who were fear- ful of crime in small towns and rural areas dropped to 23 percent and 27 percent, respectively. Toseland also reported a positive relationship between age and fear of crime. As age increased, the percentage of respondents who 28 ._o> .ou_»m:o chws_Lo yo PmcLaoo =.m_omme_=> umoz m_ on: .mON .a ”Awweoo OH "mewLo Lo Lawn: .ucmpmmoh .3 upmcom 50L; owuamu< omH. omeo.oem oooo. RoHR.~ omofifi. ompLLmz «ma. oomm.oem oooo. ~oo~.~ ooemo. mono ooo. oNHH.e¢m oooo. oomm.fi oeooo. unmoconmmm on“ cap: o:_>_o mcomLmo mmH. oooo.~¢m oooo. omoR.~ Roooo. mpooma Lo mmmcpsmopmz ooo. oomo.omm oooo. RHoN.N came“. omzoowz ooo. “moR.omm oooo. eooH.m omooo. me—oLzo ooo. «www.mmm oooo. mmom.m Rmomfi. ummLLmz Lm>mz mmo. omoo.omm oooo. emmo.¢ momoo. zupmw: ooo. ommm.o~m oooo. Ro~¢.om efifiofi. mo< ooo. momm.oo~ oooo. oo-.o~ moemm. ooosLonno_mz ms» now: copaumomwumm ooo. ooe¢.mom oooo. NRH~.oN omemm. ooo.; Lo mnwm ooo. oonm.~m~ oooo. mm~m.~m~ mommo. xmm > o Amunsmo mx_wzo Pw>wo mu:m_u_mmmoo use: cw mocmzo oo o<¢ m me u acmcmspLummo mpamem> mucmuwowco_m mucmuR$_:oRm omNPuLmocmum > o<¢ cw mocmzo ms» xn umLsmmmz mm mwma_mc< “cmcpswLum_o mg» cw on o» venom mEmLo oo Lam; $0 mLouuRomLa acmuvypco_m umoz on» we messsmui.o.m open» 29 indicated fear of crime also increased according to the findings pre- sented in Table 2.10. Table 2.10.-~Fear of Crime by Age of the Respondent (In Percent) Fear 0-24 25-44 45-64 65-74 75+ Fearful 42.6 39.1 45.4 50.6 57.3 Not Fearful 57.4 60.9 54.6 49.4 42.7 Number of Cases 202 578 441 157 117 x2 = 17.2, p < .002 Adapted from Ronald W. Toseland, "Fear of Crime: Who is Most Vulner- able," Journal of Criminal Justice, Vol. 10 (1982): p. 205. Clemente and Kleiman analyzed data from the 1973 and 1974 General Social Surveys to determine the explanatory power of five independent variables on the fear of crime.17 The independent variables were sex, race, age, socioeconomic status and size of community. As reported previously, sex was the strongest predictor of the fear of crime. Sixty-one percent of the females were afraid to walk alone at night in their neighborhood, compared to 22 percent of the males. In addition to sex, the distributions of the variables race, age, income, education and community size on fear are shown in Table 2.11. In terms of race and community size, blacks and residents of larger cities displayed fear more often than whites and residents of less populated areas. The elderly, lower-income respondents and res- pondents with less formal education were more fearful, but the differ- ences within these variables are not as great as within the variables sex and community size. 30 Table 2.11.--Percentage Distribution on Fear of Walking Alone at Night Within One Mile of Home for Sample and Relevant Subsamples, 1973 and 1974 Category N % Afraid Category N % Afraid Sample 2,700 42 Income (3): 6,999 821 48 Sex: 7.000-9,999 429 41 Males 1,291 22 10,000-14,999 665 43 Females 1,409 61 15,000 plus 785 35 Race: Education: White 2,386 40 (HS 938 44 Black 314 57 HS 898 44 >HS 864 38 Age: 18-34 972 41 Community Size: 35-49 721 40 Large City 621 60 50-64 611 43 Medium City 320 51 65 plus 396 50 Suburb 706 41 Small Town 520 36 Rural 533 24 Adapted from Frank Clemente and Michael B. Kleiman, "Fear of Crime in the United States: A Multivariate Analysis," Social Forces 56 (Sept.- June 1977/78): p. 527. The presentation of the beta and beta-squared values for each of the six independent variables support what is indicated by the per- centage distributions. Sex and city size have relatively high explana- tory power compared to race, age, income and education. The variables and findings of the analysis are presented in Table 2.12. James Garofalo, in his extensive evaluation of the aggregate data for the eight impact cities, also reported the distribution of fear of crime within age, race and family income.18 Higher age groups, blacks and lower-income respondents indicated feeling somewhat unsafe or very unsafe more often than their counterparts (see Table 2.13). 31 Table 2.12.--Beta and Beta-Squared Values for Six Explanatory Variables Variable Beta Beta-Squared Sex .39 .153 Race .05 .002 Age .09 .008 Income .06 .003 Education .02 .000 Size .24 .058 R = .48, R2 = .230 Adapted from Frank Clemente and Michael B. Kleiman, "Fear of Crime in the United States: A Multivariate Analysis," Social Forces 56 (September-June 1977/78): p. 528. Table 2.13.--Estimated Rates of Fear of Crime Among Age, Race and Family Income Groups: Eight Impact Cities Aggregate, 1975 Percent Responding Percent ReSponding Somewhat Unsafe or Somewhat Unsafe or Very Unsafe Very Unsafe Age Race 16-19 37 White 41 20-24 38 Black 54 25-34 37 35-49 43 Family Income 50-64 50 Less than $3,000 62 65 or older 63 $3,000-$7,499 53 $7.500-$9.999 45 Sex $10.000-$14.999 39 Male 26 $15,000-$24.999 34 Female 60 $25,000 or more 30 Adapted from James Garofalo, "Victimization and the Fear of Crime," Journal of Research in Crime and Delinquency, 16, 1979, p. 85. 32 What are the Determinants of the Fear of Crime? This section discusses the eight determinants of fear as they are used in this study. Each determinant will be discussed separately; however, it will be necessary at times to discuss the relationship bet- ween the various determinants. It is accepted that the determinants of fear are probably interrelated and not mutually exclusive. Direct Victimization This determinant reflects the approach that criminal victimiza- tion contributes to the feelings of fear within a population by nega- tively affecting the individuals who have been personally victimized or who have experienced direct victimization. The data to support this approach has been substantiated; however, direct victimization generally affects a relatively small portion of a population. In reviewing the aggregate data for the eight impact cities, the percentage of respondents feeling unsafe (combining somewhat unsafe or very unsafe) when out alone in their neighborhood at night did not vary depending on their victimization status. The data presented in Table 2.14 serves two purposes. First, it indicates that a sizeable proportion of crime victims responded some- what unsafe or very unsafe. Forty-five percent of the victim responses fell into these two categories. It is important to note, however, that approximately the same percentage of nonvictims responded somewhat unsafe or very unsafe also. Second, it is apparent that the number of nonvictims (2,058,170) is almost twice the number of victims (1,050,510). These two observa- tions initially indicate that direct victimization can account for only 33 a portion of fear of crime in the general population. In addition, because of the large number of nonvictims feeling unsafe, the search for determinants of fear of crime must extend beyond the direct victim- ization approach. Table 2.14.--Neighborhood Safety at Night By Victimization Status Victimization Reasonably Somewhat Status Very Safe Safe Unsafe Very Unsafe Victim 18% 37% 21% 24% (187,410) (386,470) (217,790) (258.840) Nonvictim 17% 37% 21% 23% (360,470) (772,460) (442,930) (482,310) Adapted from James Garofalo, Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-l, 1977): p. 58. Despite this contention, the direct victimization approach is generally used to explain the differences in the level of fear between victims and nonvictims. Victims of personal crimes such as rape, rob- bery and personal theft (pickpocketing and purse snatching) generally show more fear than nonvictims. Skogan and Maxfield illustrate this point by displaying the responses of nonvictims and robbery victims for the fear of crime question. In Table 2.15, a higher percentage of robbery victims felt very unsafe than nonvictims. The type of criminal victimization may have an effect on the level of fear. Skogan has recently shown moderately strong correlations between victimization type, feelings of safety, and no change in activ- ity.19 Using a 50 percent sample from the 1973 attitudinal survey of 34 the Nation's five largest cities, Skogan correlated the responses of various victims with their responses to feelings of safety and change in activity. Table 2.15.--Robbery Victimization and Fear Fear Nonvictims Robbery Victims Very Safe 13.5% 10.1% Reasonably Safe 38.6 35.4 Somewhat Unsafe 24.5 23.0 Very Unsafe 23.3 31.5 Total 99T9 TOOTO (28,472) (1,629) Adapted from Wesley G. Skogan and Michael G. Maxfield, Coping With Crime (Beverly Hills: Sage Publications, 1981): p. 61. As illustrated in Table 2.16, the crimes of rape, robbery and personal theft (pickpocketing and purse snatching) had the highest cor- relations with feelings of safety and no change in activity. The nega- tive correlations indicate the direction of the relationships are towards feelings of unsafety and change in activity. Consequently, victims of rape, robbery and personal theft were more likely to feel unsafe and to change or limit their activities. Particularly surprising is the low correlation between assault and fear of crime. Skogan suggests that this may be due to inadequacies in the measurement of assault or to the large component of inter-family, nonstranger crime in this category. In addition, burglary, which is v generally classified as a property crime, had a relatively high correla- tion when compared with assault or larcency. The intrusion into an 35 individual's personal domain may account for the negative correlation in addition to the fact that the burglary is sometimes of an occupied dwelling.20 Table 2.16.--Victimization and the Fear of Crime Correlation (Gamma) With: Victimization Type Feelings of Safety No Change in Activity Rape -.25 -.51 Robbery -.22 -.32 Personal Theft -.34 -.40 Assault -.01 -.09 Burglary —.13 -.14 Larceny .04 .01 Auto Theft .02 -.06 (n = 23,022) (n = 23,033) Adapted from Wesley G. Skogan, "Public Policy and the Fear of Crime in Large American Cities," Public Law and Public Policy, Edited by: John A. Gardiner (New York: Praeger Publishers, 1977): p. 7. The direct victimization approach loses support when data reveals that those who least fear crime are those who are most likely to be victimized. For age and sex, the lower age groups and males experience higher rates of personal victimization, yet these groups also have a lower rate of fear of crime. Garofalo provides the estimated rates of victimization and fear of crime among age, sex, race and family income groups for the eight impact cities.21 The personal victimiza- tions consisted of rape, robbery, assault and larceny with victim/ offender contact. The data in Table 2.17 also shows that blacks and lower income groups experience higher rates of victimization and higher 36 Table 2.17.--Estimated Rates of Victimization and Fear of Crime Among Age, Sex, Race and Family Income Groups: Eight Impact Cities Aggre- gate, 1975 Rates of Personal Percent Responding Victimization per Somewhat Unsafe or 1,000 Persons Very Unsafe Age 16-19 125 37 20-24 105 38 25-34 76 37 35-49 51 43 50-64 42 50 65 or older 34 63 Sex Male 90 26 Female 54 60 Race White 69 41 Black 72 54 Family Income Less than $3,000 93 62 $3.000-$7,499 78 53 $7.500—$9,999 7O 45 $10,000-$14,999 64 39 $15.000-$24,999 59 34 $25,000 or more 56 30 Adapted from James Garofalo, ”Victimization and the Fear of Crime," Journal of Research in Crime and Delinquency, 16, 1979, p. 85. 37 rates of fear. While the relationship between these two variables is not specified, Garofalo suggests that the relationship between fear and victimization for these groups may result from the convergence of these factors in particular geographic locations. Indirect or Vicarious Victimization Indirect victimization is an aftereffect that occurs once the direct or personal victimization has occurred. Indirect victimization contributes to the 'ripple effect' whereby individuals learn or vicar- iously experience victimizations that have occurred to friends or acquaintances. An individual becomes aware of criminal events through 'word of mouth' and not through the media or by direct experience. The NCS was one of the first studies to indicate that the fear of crime was related to indirect victimization. Respondents were asked to reSpond to various experiences with victimization; one experience which simply involved residing within a household in which another mem- ber was a victim of a personal crime. Of the 3,129,190 respondents from the eight impact cities, 228,840 or 7.3 percent of the respondents indicated they resided in such a household. These respondents were also asked their perception of neighborhood safety at night. The res- ponses were crosstabulated with the experience with victimization and the results are presented in Table 2.18. In reference to indirect victimization, 42 percent of the res- pondents felt somewhat or very unsafe while also knowing a household member who was a victim of a personal crime. This does not obviously represent cause and effect, but the data does indicate a relationship exists between the two variables. It should be noted that of the total respondents who had no contact with crime, 45 percent felt somewhat 38 or very unsafe, which again suggests that additional determinants, other than direct or indirect victimization must account for the respondent's perception of unsafety. Table 2.18.--Feeling of Personal Safety When Out Alone at Night with a Selected Experience with Victimization, Eight Impact Cities Aggregate Experience with Reasonably Somewhat Victimization Very Safe Safe Unsafe Very Unsafe Other Household 20% 38% 21% 21% Member Victim of a Personal Crime (44,520) (87,990) (47,080) (47,770) Adapted from James Garofalo, Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-l, 1977): p. 61. Nevertheless, one particular study indicates that the rates for indirect victimization exceeds the direct victimization rates.22 In other words, individuals are more likely to experience victimization indirectly than directly. The Behavioral Sciences Laboratory (BSL) of the University of Cincinnati conducted a telephone survey in November 1977. A sample of 1,600 citizens, aged sixty and over were selected from eight neighborhoods in the cities of New York, Los Angeles, Mil- waukee and New Orlenas. The reSpondents were asked a series of ques- tions concerning socio-demographic characteristics, their experiences with victimization, their perceptions and fear of crime and their aware- ness and use of crime prevention measures. The BSL team discovered several important findings. First, with the use of multiple regression analysis, several demographic variables were discovered related to the fear of crime. Women, blacks, lower-income groups and older age groups were more afraid of crime than their counterparts. This parallels the 39 finding of Clemente and Kleiman, Toseland, Garofalo and others. A relationship was also found between the fear of crime and the reported crime rate. Utilizing the demographic data, two neighborhoods that closely matched each other were identified: Sherman Park in Mil- waukee and Flatbush in Brooklyn. DeSpite the demographic similarities, it is important to point out that the environment of each neighborhood was different. While Sherman Park was described as resembling a suburb with 86 percent of its elderly living in houses or duplexes, Flatbush was located within the city. Flatbush consisted of commercial and resi- dential establishments with 80 percent of its elderly living in apart- ment buildings. In reference to the fear of crime, each respondent was asked four questions and the percentage of respondents reSponding affirmatively is presented in Table 2.19. The percentages in Table 2.19 obviously Show that considerably more elderly residents from Flatbush were fearful than residents from Sherman Park. According to the direct and indirect victimization approaches, this finding would be expected as long as Flatbush had higher rates of direct victimization and indirect victimization. Resi- dents were asked to report their personal experiences concerning robbery or attack in addition to their knowledge of their friends and acquaint- ances being robbed or attacked. The results are highly significant. Converting the number of direct and indirect victimizations into rates per thousand, Flatbush had much higher rates than Sherman Park. Not only is it important to mention the relationship of high victimization rates (direct and indirect) with a higher level of fear for Flatbush, the indirect rates in Table 2.20 are significantly greater than the direct victimization rates. This observation may indicate that among 40 the elderly, indirect victimization might have a stronger impact on the fear of crime within an environment than direct victimization. Table 2.19.--Response5 By Elderly Residents of Flatbush and Sherman Park to Questions on Perceptions and Fear of Crime Percentage of Elderly Responding Affirmatively Response Flatbush Sherman Park They felt "very unsafe" alone in their neighborhoods at night 40 14 They felt "very unsafe" alone in their neighborhood during the day 19 2 They avoid certain places in the neighborhood at night because of crime 76 30 They avoid certain places in the neighborhood during the day because of crime 58 18 Adapted from Victoria H. Jaycox, "The Elderly's Fear of Crime: Rational or Irrational?," Victimology: An International Journal, Vol. 3 (1978): p. 330. It is also important to point out that the indirect rates are comparable between Flatbush and Sherman Park. However, the direct per- sonal victimization rate for Flatbush is approximately four times the direct victimization rate for Sherman Park. With a relatively large indirect victimization rate for Sherman Park, it is possible that the informal control, as discussed in a later section, is greater among the residents of Sherman Park than residents of Flatbush. Data to support this implication was presented by Jaycox. It was reported that while 69 percent of Sherman Park's elderly "feel a part" of their neighborhood, 41 Table 2.20.--Personal Crime: Direct and Indirect Rates of Victimization of the Elderly in the LEAA National and BSL Neighborhood Surveys Rate Per Personal Victimizations Thousand Direct: Rate in Sherman Park for Elderly (60 years and older) for robbery or attack on the street in 1977. (BSL survey) 38.0 Rate in Flatbush for elderly (60 years of age and older) for robbery or attack on the street in 1977. (BSL survey) ' 164.0 Indirect: Rate in Sherman Park for elderly's friends and acquaint- ances who were robbed or attacked on the street "in the past couple of years." (BSL survey) 520.0 Rate in Flatbush for elderly's friends and acquaintances who were robbed or attacked on the streets "in the past couple of years." (BSL survey) 640.0 Adapted from Victoria H. Jaycox, "The Elderly's Fear of Crime: Rational or Irrational?," Victimology: An International Journal, Vol. 3 (1978): p. 332. 42 only 30 percent of Flatbush's elderly subscribed to these feelings. A higher level of informal control or feelings of stability may result in higher levels of indirect victimization. Vulnerability It is quite apparent from crime statistics that certain popula- tion groups are more vulnerable to crime than others. Referring back to Table 2.17, objective rates of personal victimization indicated that younger age groups, males and lower-income groups were most vulnerable to crime. However, it was also shown that the crime rates did not cor- respond with the 'fear rate' for the same groups. This negative rela- tionship between victimization and fear of crime indicates that certain groups may perceive themselves as vulnerable to crime, which therefore increases their fear of crime. This perception exists deSpite what the objective crime rate is for a particular group. The concept of vulnerability has been primarily concerned with the demographic indicators of vulnerability to personal attack. The indicators of sex and age have been associated with one main type of vulnerability: physical vulnerability. Physical vulnerability is referred to as the inability to resist attack or the inability to defend themselves.23 Due to their physical conditions, women and the elderly are generally associated with physical vulnerability. In a study by Riger, Gordon and LeBailly, respondents were asked if they thought they could defend themselves successfully against attack.24 While 54 percent of the males reSponded they could defend themselves against attack, 41 percent of the females reported they could. As expected, more women who perceived themselves as vulnerable also had a higher level of fear. 43 It is important to point out that the difference here is statistically significant (p S .0004). However, a large percentage (59 percent) of the females did not perceive themselves as physically vulnerable. While the percentage of females who felt somewhat or very unsafe (43.6 percent) is similar to the percentage of females who felt vulnerable (59 percent), a relationship between vulnerability and fear of crime has not been established. Riger, Gordon and LeBailly took a step in this direction by show- ing that women who feel very unsafe also feel considerably more vulner- able. Respondents were asked to compare themselves in strength and running speed to the average woman and man. A "vulnerability index" was compiled and among women, significant differences in feelings of vulnerability are associated with different levels of reported fear. The mean scores indicate that women who felt considerably more vulner- able also felt very unsafe. As indicated in Table 2.21, a similar level of vulnerability is observed for women who felt very safe, reasonably safe or somewhat unsafe. Although the concept of vulnerability is related to the fear of crime, without adequate control variables we do not know the impact that vulnerability has on the fear of crime. The elderly are also asso- ciated with being more fearful, as previously indicated by Toseland and Kleiman and Clemente. Furthermore, it has been consistently reported that the size of community, generally distinguished by rural, suburban or urban, is strongly related to fear of crime. The concern that arises in regard to the concept of vulnerability is the relation- ship between vulnerability and other variables, such as city size and perception of crime. 00 women feel more vulnerable in urban areas as 44 a result of their perception of crime or perception of actual risk, or have they been 'socialized' into feeling vulnerable as some people contend? Questions like these that refer to the interaction and rela- tion between determinants of fear remain unanswered. Table 2.21.--Women's Vulnerability* By Fear Of Crime Fear Level Mean Very Safe 2.25 Reasonably Safe 2.22 Somewhat Unsafe 2.29 Very Unsafe 2.48 *Vulnerability was measured by questions ask- ing the respondent to compare him/herself in strength and running speed to the average woman and man. Scores range from O to 4. Adapted from Stephanie Riger, Margaret T. Gordon and Robert LeBailly, "Women's Fear of Crime: From Blaming to Restricting the Vic- tim," Victimology: An International Journal, Vol. 3 (1978). A second type of vulnerability, as discussed by Skogan and Max- field, is social vulnerability to crime. This refers to the frequency at which pe0ple are exposed to the threat of victimization. Because blacks and the poor are generally victimized disprOportionately, the contention is that these groups fear crime as a result of their increased exposure to crime. This concept seems to be an overlap of two previous determinants: direct and indirect victimization. If blacks and the poor are disproportionately victimized, it would seem logical that these groups would have a high rate of indirect and direct 45 victimization. Formal Versus Informal Control One of the earliest commentators on the fear of crime was Jane Jacobs.25 She comments, . . . if a city's streets are safe from barbarism and fear, the city is thereby tolerably safe from barbarism and fear. Jacobs attributes a perception of safety in the streets to 'an intri- cate, almost unconscious network of voluntary controls and standards among the people themselves, and enforced by the people themselves."26 While the police are regarded as a necessary formal control to crime and subsequently to fear, the informal control is what maintains a con- stant 'watch' over what happens on a city's streets and sidewalks. In addition to being well—used, Jacobs contends that streets must con- tain people who are known to belong on the streets or natural proprie- tors. Shop owners, people going about their daily business, or pe0ple out on the streets for perceived purposes other than crime or trouble must frequent the streets and sidewalks to maintain the aura of safety. The National Crime Commission made reference to the informal control by noting in 1967: A man who lives in the country or in a small town is likely to be conspicuous, under surveillance by his community so to speak, and therefore under its control. A city man is often almost invisible, socially isolated from his neighborhood and therefore incapable of being controlled by it.27 The study conducted by Sarah L. 80995 in 1971 clearly shows that a citizen's perception of neighborhood safety can be more dependent on informal controls than on formal controls.28 In this study, respon- dents were asked 'how safe from crime would you say your neighborhood is?’ As indicated back in Table 2.7, the percentage of respondents 46 indicating unsafe or very unsafe ranged from 3.1 percent in rural areas to 24.1 percent in central cities. Shown in Table 2.22 are the responses to the question, "What makes your neighborhood safe?" Although the reSponses were limited in nature, the objective was to determine if reliance was placed on the informal controls versus the formal control. The data presented clearly indicates heavy reliance on informal controls, although a greater percentage of reSpondents from suburbs and central cities relied on formal control (law enforcement) than rural/small town respondents. Table 2.22.--What Makes Your Neighborhood Safe? Rural/Small Central Total Town Suburbs City State (N = 360) (N = 212) (N = 270) (N = 842) Informal Controls: 82.9% 70.3% 67.9% 75.8% Residents' Character 37.1 32.6 37.4 36.0 Community Character 33.3 24.8 18.7 27.4 Social Network 12.5 12.9 11.8 12.4 Formal Control (Law Enforcement) 10.4 20.2 17.7 15.1 No Past Crime 4.1 4.5 7.4 7.4 Other Reason --- 1.5 1.0 .7 Don't Know or Not Ascertained 2.6 3.5 6.0 1.0 Total 100.0 100.0 100.0 100.0 Adapted from Sarah L. Boggs, "Formal and Informal Crime Control: An Exploratory Study of Urban, Suburban and Rural Orientations," The Sociological Quarterly 12 (Summer 1971): p. 324. The informal control which consisted mainly of the residents' character and community character refers to whether neighbors were 47 perceived as decent, law-abiding, middle-class citizens, and also to the physical character of the neighborhood. These informal controls are related to Lewis' and Maxfield's concept of incivility.29 The data for their study consisted of field observations, telephone surveys and official crime rates. The neighborhoods involved in the study consist of the four Chicago neighborhoods discussed earlier in this chapter: Lincoln Park, Wicker Park, Woodlawn, and Back of the Yards. Incivility was measured by several categories consisting of the resident's concern for teenagers hanging out on the streets, presence of abandoned build- ings, people using illegal drugs and evidence of vandalism. By summing the objective crime rates for each neighborhood found in Table 2.23, a ranking of the neighborhoods from highest crime rate to lowest would be as follows: 1. Woodlawn (59.9) 2. Lincoln Park (51.4) 3. Wicker Park (45.0) 4. Back of the Yards (25.0) Table 2.23.--Crime Rates (Per 1,000 Residents) By Neighborhood Wicker Lincoln Back of Crime Type Woodlawn Park Park the Yards Burglary 28.0 24.4 39.5 15.5 Rape and Sexual Assault 2.7 1.0 2.2 0.8 Robbery 21.6 12.5 6.7 6.0 Assault 8.8 7.5 4.0 3.0 Adapted from Dan A. Lewis and Michael G. Maxfield, “Fear in the Neigh- borhoods: An Investigation of the Impact of Crime," Journal of Research in Crime and Delinquency 17 (1980): p. 174-178. 48 The interesting finding that occurs when incivility and fear is analyzed is that the ranking of the neighborhoods changes, in spite of the objective crime rates. Table 2.24 indicates that Wicker Park residents showed the most concern for incivility and also had the high- est level of fear. This is especially surprising when Table 2.23 indi- cated a lower objective crime rate for Wicker Park. One possible explanation for this occurrence is the use of reported crime rates instead of victimization survey rates. The discussion earlier concern- ing indirect victimization proposed that knowledge of crime is not generated simply from crimes reported to the police. Indirect victim- ization would also include 'non-reported' crimes. Using this as an assumption, it may be possible that the Wicker Park total crime rates (including non-reported crime) is much higher than the official reported crime rates indicate. Therefore, while the fear of crime does appear to correspond with the concern about incivility, caution should be used until non-reported crime rates are taken into account. Table 2.24.--Incivility and the Fear of Crime Concern About Incivility a Fear Range = Big Problem 0-100% Wicker Park 2.0 54.6% Woodlawn 1.83 50.4% Lincoln Park 1.74 29.2% Back of the Yards 1.62 29.1% Incivility Range = 0-4; 3.0 = Great Big Problem; 1 = No Problem Adapted from Dan A. Lewis and Michael G. Maxfield, "Fear in the Neigh- borhoods: An Investigation of the Impact of Crime," Journal of Research in Crime and Delinquency 17 (1980): p. 186. 49 Perception of Crime As mentioned briefly already, direct victimization may have a positive affect on indirect victimization; that is, as the amount of actual crime increases, more people directly and indirectly experience crime. If the public is aware of the increase in crime, the fear of crime may also increase. However, this discussion involves a determi- nant consisting of perceptions which are not necessarily dependent on objective crime rates. The perception of the crime level may be based upon direct victimization experiences, indirect victimization experi- ences, or the media. Whether or not the perception is accurate, as compared to the objective crime rate, is not of concern. The issue is to determine the impact on fear, taking into account an individual's perception of crime. In the study discussed earlier by Frank F. Furstenberg, Jr., the relationship between fear of crime and the estimate of neighborhood safety was determined by crosstabulating the responses to these two variables. An index was established for the fear of crime which con- tained three levels: low level of fear, medium level of fear, and high level of fear. Respondents were also asked to estimate their neighbor- hood safety with the response categories consisting of: less safe than most (neighborhoods), about average, and more safe than most (neighbor- hoods). As observed in Table 2.25, a relationship existed between the perception of safety and the fear of crime. Respondents who perceived their neighborhood as less safe than most neighborhoods, were more likely to have a high fear of crime than respondents who perceived their neighborhood as being more safe. 50 Table 2.25.--The Relationship of Fear of Crime to Estimate of Neighbor- hood Safety Less Safe About More Safe Fear of Crime Index Than Most Average Than Most Low 13% 30% 57% Medium 32 35 24 High 55 35 20 x2 = 183.534, 4 d.f., p S .001 Adapted from Frank F. Furstenberg, Jr., ”Public Reaction to Crime in the Streets,“ American Scholar 40 (1970/71): p. 607. An individual's perception of crime can be an accurate assessment of the actual crime rate of the neighborhood. As discussed earlier, Marlys McPherson found residents were able to accurately identify which crimes were a 'problem' in their neighborhood. John E. Conklin, in his study of two communities in an eastern metr0politan area, compared the reported crime rates of a suburban neighborhood with an urban neigh- borhood.30 Residents of the urban neighborhood, which had nine times the amount of reported crime than the suburban neighborhood, were more likely to perceive the neighborhood as unsafe. Although the perception of crime was greater in the urban neighborhood, the perception corres- ponded with the actual level of crime. The relationship between the perception of crime and other deter- minants of fear has been largely unexplored. As mentioned before, direct victimization, indirect victimization, and the media may lead an individual to gain an accurate perception of the crime level. On the other hand, perceptions of the crime level may be blown out of 51 'proportion, which could lead to an unjustified increase in the fear »of crime. The perception of crime may also have little impact on the fear level. Individuals may or may not perceive the level of crime as being high, but their perceived physical vulnerability to crime may increase their fear. Finally, as we will discuss shortly, environmental factors may be related to a perception of unsafety or unsafety. Psychological The psychological determinant of fear generally results from misinformation where an individual's fear of crime far exceeds the objective probability of being victimized. Most often, the psychologi- cal correlates of fear have centered around the fear of strangers or xenophobia. John E. Conklin suggests that individuals have a tendency to blame strangers for crime, which allows them to maintain a sense of security, believing that their own neighbors are not a personal threat to them.31 This has been indirectly supported by data that Shows individuals perceiving crime affecting others, but not themselves. Respondents in the Census Bureau Surveys were asked about limiting or changing their activities because of fear of crime. Eight-seven percent of the respondents thought 'pe0ple in general' limited or changed their activities. When asked whether people in their neighborhood did this, 67 percent responded in the affirmative. Furthermore, only 48 percent of the respondents indicated they limited their activities because of crime.32 The fear of strangers per se is of little value if the link bet- ween this concept and crime cannot be established. While individuals may fear neighborhoods where many strangers are present, the findings 52 are somewhat limited due to the many situations in which individuals may be exposed to strangers. In shopping malls and business areas. virtually everybody may be a stranger to each other, yet the level of fear in most of these situations may be low. Besides the fear of stran- gers as a psychological determinant of fear, individuals that fear crime regardless of the objective circumstances is a concern. Some individ- uals, without necessarily having a neuroses, may fear crime because of the serious consequences victimization has for them.33 In addition to women being subject to sexual assault, women and older people may be physically vulnerable to crime, as discussed previously. Being unable to defend themselves to any substantial degree, women and older people may suffer physically and take longer to heal in the aftermath of the victimization. Consequently, certain individuals, primarily women and older people, may fear crime at any time and in any place. Although this determinant is sometimes attributed to being a neurosis or irrational, the perceived consequences of crime to these individuals is undoubtedly real. Media Effect In addition to direct victimization and indirect victimization, the news media is the third method of obtaining knowledge of crime occurring in the environment. It was previously indicated that direct victimization affects relatively few individuals and indirect victimiza- tion has a greater impact, but it is probably not the primary source of fear. Research concerning the sources of information about neighbor- hood crime indicates that the media is a major source of information. The term media is generally referred to as newspapers, radio and 53 television. In this section, emphasis is placed on obtaining informa- tion about the local neighborhood crime occurrence. In the study by Skogan and Maxfield, respondents were asked what was their best source of information concerning neighborhood crime. They report that newspapers, television and personal conversation accounted for approximately 86 percent of all responses. The newspapers appear to be the major source of crime news in this particular study. Forty-five percent of the respondents indicated they had just recently (the day before they were surveyed) read in a newspaper or watched on a television news show, a report concerning crime. In a separate question, the respondents were asked what was their best source of information about crime in their neighborhood. Thirty-one percent indicated that a major daily newspaper or a local community paper was their most important source of local crime news. This finding is not particularly surprising when we look at var- ious newspapers and find how much emphasis is placed on the subject of crime. Gordon and Heath recently reported that in eight major news- papers in Philadelphia, Chicago and San Francisco, 19.1 percent of the major topics of the top two lead stories concerned violent crime. Classifying the major topics, Politics (31 percent) was the most common tapic followed by Crime (19.1 percent), Sports/Human Interest (17.2 percent), Foreign (12.6 percent), Domestic (11.8 percent), and Natural Disaster (8.4 percent).34 With emphasis placed on crime news, it would be hard for an indi- vidual not to be exposed to crime on a local level, let alone a city- wide level. In the NCS questionnaire, one item was related to the impact that crime news in the media has on an individual's perception 54 of crime locally. Respondents were asked whether crime was less serious, more serious or about as serious than the newspapers and TV say. Believing that the media is publicizing a vast majority of the criminal events that occur, approximately 9 percent of the respondents perceived crime as being less serious than what the newSpapers and TV portray. Forty-two percent of the respondents in the five largest cities thought that crime was more serious. Surprisingly, 41 percent perceived crime as being about as serious relative to what the newspapers and TV say. This particular finding points out that the media does play an important role in influencing individuals' perception about crime. Accepting the fact that the media actually reports an insignificant number of all the crime that is reported to the police, it is amazing that such a large percentage of respondents would view crime 'about as serious.‘35 It would seem logical that the level of fear would increase if the indi- viduals responding 'about the same' knew that the actual crime level was much higher or more serious. James Garofalo crosstabulated the media-related question to the fear of crime. As indicated in Table 2.26, individuals who perceived crime as 'more serious' rather than 'about the same' or 'less serious,’ were more likely to feel somewhat or very unsafe. Of the reSpondents who thought crime was more serious than what the media indicates, 51 percent felt somewhat or very unsafe. Forty-three percent of the respondents who indicated 'about the same' felt unsafe, and only 29 percent who thought crime was less serious than reported in the media felt unsafe. 55 Table 2.26.--Fear of Crime by Perceived Seriousness of Crime Relative to What the Newspapers and Television Say: Eight Impact Cities Aggregate, 1975 Fear of Crime Seriousness of Crime Relative Very Reasonably Somewhat Very Estimated to What Media Say Safe Safe Unsafe Unsafe Number Less Serious 28% 42% 16% 13% 261,623 About the Same 15% 42% i 23% 20% 1,514,438 More Serious 13% 36% 23% 28% 1,292,171 Estimated Number 474,850 1,208,236 683,365 701,781 3,068,232 Adapted from James Garofalo, "Victimization and the Fear of Crime," Journal of Research in Crime and Delinquency 16 (1979): p. 89. Environment The final determinant to be discussed concerns the effects that the physical environment may have on an individual's fear of crime. Based upon the works of Jane Jacobs, Oscar Newman and others, the National Institute of Law Enforcement established a program entitled Crime Prevention Through Environmental Design.36 It was assumed at the beginning of this program that high risk of victimization and high levels of fear may be dependent upon the characteristics of the physical environment. Poor lighting, landscape design, blind spots, places in which perpetrators could hide, and building design were considered to be prime environmental characteristics which could increase the level of fear. Of central concern in this study was the potential effects of poor lighting and landscape design which would provide blind spots and 56 hiding places. In the review of literature, numerous studies have been conducted concerning these two specific environmental characteristics. The most comprehensive study of street lighting and its impact on the fear of crime and level of crime was reported by James M. Tien, et al.37 Tien et al. reviewed forty-one street lighting projects and evaluated fifteen of the leading studies. While the authors indicated that meth- odological problems existed in all of the studies, empirical evidence supported the authors' conclusion that "there is a strong indication that increased lighting--perhaps lighting uniformity--decreases the fear of crime."38 The projects which addressed the fear of crime issue were con- ducted in Baltimore, Milwaukee, Tucson, Denver, Portland and Norfolk. Because the study by Tien was quite extensive, a summary of the findings will be presented. The reported impacts on the fear of crime attributed to street lighting consisted of a reported change in attitude. The attitude impact was concerned with measuring the changes in feelings of safety and/or related attitudes. Although the measurement methods varied and were not usually well-defined, the survey question reflecting change in attitude was typically, "Since the addition of the new street lights, do you generally feel safer, the same, or not as safe?" In Table 2.27, a summary of the reported impacts attributed to street lighting is presented. Although caution should be used in interpreting the results, the summary does support the proposition that street light- ing reduces the fear of crime. As for the effects of landscape design on the fear of crime, very few studies have established the relationship between these vari- ables. Most researchers have studied the effects of the physical 57 environment on criminal behavior. Oscar Newman examined the relation- ship between crime rates and a number of building characteristics.39 In terms of the fear of crime, a survey of 425 tenants in 7 New York City public housing projects indicated that the fear of crime was rela- tively high in public housing developments. Approximately 16 percent of the residents of high-rise developments rated their buildings as safe or fairly safe compared to 32 percent of the residents of low- rises (three stories). Table 2.27.--Reported Impacts Attributed to Street Lighting for Selected Cities City Reported Change in Attitude Baltimore 66% of residents "feel safe." Denver 43% of residents were unaware of “additional street lighting.“ Of residents aware of street lighting improvement, over 67% "feel much safer." Milwaukee 82% of residents feel safer. Additionally, 71% of resi- dents perceived a decrease in crime and 90% were "generally satisfied.” Norfolk An unspecified percentage of "test subjects" reported increased sense of security. Portland 25% of target area residents were aware of increased street lighting. No impact on residents' feelings of safety. Tucson UnSpecified percentage of residents felt "substantially safer," and reported "less fear" walking through alleys at night. Adapted from James M. Tien, et al., Street Lighting Projects: National Evaluation Program, Phase 1, Final Report (Cambridge, MA: Public Sys- tems Evaluation, Inc., 1977): p. 67-69. In 1975, a study was conducted by Barton-Aschman Associates to assess the security and crime problems in Allentown, Pennsylvania. 58 It was reported that well-lighted streets and sidewalks promoted feel- ings of safety. Furthermore, feelings of insecurity were associated with areas that contained numerous trees and shrubs.40 The environ- mental characteristics of poor street lighting and the presence of large shrubs and potential for concealment are two major elements of the social surveillance rationale, as discussed by Rubenstein, et al.41 This rationale suggests that the design of the physical environment can help individuals detect suspicious behavior or crimes. Subsequently, the fear of crime should be reduced once the level of crime is reduced. In 1976, Thomas Molumby analyzed crime in a single housing development and found that crimes were associated with opportunity for conceal- ment.42 A victimization survey was conducted which collected informa- tion concerning number and type of crimes, where and when it occurred, method of entry and several other factors. Information pertaining to the Spatial aspects of the dwelling units included building type, place- ment, lighting and traffic flow. The most notable finding was that the highest victimization “break-in" rate was associated with the town- houses in which visibility was hindered by overgrown shrubbery, trees and inadequate or nonexistent lighting. This was contrary to the find- ing for the apartments in which 95 percent of the crimes occurred out- side the apartments and items were generally taken from the yards, vehicles or breezeways. In the townhouses, 35 percent of the crimes were "break-ins." Although this study did not discuss the fear of crime, it is logical that the fear of crime could be higher in build- ings which afforded areas for concealment. 59 Summary The first section of this chapter discussed the levels of fear of crime from several perspectives. National level data indicates that the fear of crime had risen during the 19605 and has leveled off during the 19705. Surveys conducted in large cities throughout the United States indicate that the levels of fear do vary between cities. At the neighborhood or community, variation in levels of fear also exists. This section also discussed the relationship between the level of fear within neighborhoods and the citizens' perceptions of danger. In addi- tion, data supports the relationship between personal crime rates and the citizen's perception of danger. Finally, a comparison of the levels of fear by community size was presented. A higher level of fear was found among residents of central cities as compared to residents of rural/small towns and suburbs. The second section was concerned with identifying various sub- groups that have higher levels of fear of crime. This section focused upon such characteristics as sex, race, income and education of the fear victim. Sex of an individual was reported to be the most powerful predictor of the fear of crime. In several studies presented, there were wide margins between males' and females' feeling of personal safety at night. Age and race have also been associated to the fear of crime with elderly and black individuals reporting higher levels of fear than their counterparts. In addition, the variation between income groups and education groups concerning the fear of crime is not so substantial as in sex and race. The determinants of the fear of crime were discussed in the final section. The review of literature exposed eight major reasons why 60 individuals perceive themselves as potential crime victims. It was reported that individuals may fear crime as a result of being a victim of crime previously. In addition to directly experiencing crime, indi- viduals may experience crime indirectly or vicariously by being aware of victimizations that have occurred to friends or acquaintances. It was reported that the news media also informs individuals about crime that is occurring within the community or neighborhood. It was pr0posed that these three determinants may affect an individual's perception of crime in his or her neighborhood. Despite the objective crime rate, individuals may have a fear of crime if they perceive that a lot of crime is occurring in the neighborhood. Research has also purported that individuals rely on local law enforcement agencies for protection. When an individual perceives the law enforcement as inadequate in providing protection, the individual may have a fear of crime. In contrast to formal control is a concept referred to as informal control, or the 'cohesiveness' of the neighbor- hood. Individuals who perceive high social cohesion within a neighbor- hood may have a lower level of fear of crime. Physical vulnerability to crime may also affect the level of fear. The vulnerability to crime refers to the inability to defend oneself against a criminal attack. Somewhat related to this determinant is a determinant labeled as psychological. This determinant was used to identify individuals who have a fear of crime anywhere at any time. Finally, environmental characteristics that have an affect on the fear of crime were discussed. Building design, excessive shrubbery and bushes that create blind spots and poor lighting conditions may affect an individual's perception of safety at night. UMBERII FOOTNOTES 1George Gallup and Associates, "Fear of Crime, Victimization Now Common to Many Americans," The Gallup Opinion Index, Vol. 124 (October 1975 . 2Wesley G. Skogan and Michael G. Maxfield, Coping With Crime (Beverly Hills: Sage Publications, 1981). See also James Alan Fox, Forecasting Crime Data: An Econometric Analysis (Lexington, MA: 0.0. Heath, 1978). 3Gallup Organization, The Gallup Opinion Index (Princeton, NJ: Gallup Organization Inc., 1981). 4James Garofalo, Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-l, 1977). 5John E. Conklin, The Impact of Crime (New York: Macmillan, 1975). 6Crimes against person were listed as murder, rape, robbery, aggravated assault, burglary, larceny over $50 and auto theft. 7Marlys McPherson, "Realities and Perceptions of Crime at the Neighborhood Level," Victimology: An International Journal 3 (1978): 320. 8Douglas Frisbie, Glenn Fishbine, Richard Hintz, Mitchell Foelson, and Julia Brown Nutter, Crime in Minneapolis: Proposals for Prevention (Minneapolis: Minnesota Crime Prevention Center, 1977). 9McPherson notes that the survey was conducted in the fall of 1975 and consisted of 1,541 adult residents. The confidence that per- centages based upon the sample are within i 3.5 percentage points of the actual population percentages is 99 percent. Information concerning the consequences of fear, crime prevention measures taken and attitudes toward police was also collected. 105kogan and Maxfield, Coping With Crime, p. 54. 11For a brief description of each neighborhood, see pages 23-26 in Skogan and Maxfield, Coping With Crime, 1981. 61 62 12Sarah L. Boggs, "Formal and Informal Crime Control: An Explor- atory Study of Urban, Suburban and Rural Orientations," The Sociological Quarterly 12 (Summer 1971): 319-327. 13James Garofalo, "Victimization and the Fear of Crime," Journal of Research in Crime and Delinquency 16 (1979): 80. 14Skogan and Maxfield, Coping With Crime, p. 55-58. 15Garofalo, Public Opinion. 16Ronald W. Toseland, "Fear of Crime: Who is Most Vulnerable," Journal of Criminal Justice 10 (1982): 199. 17Frank Clemente and Michael B. Kleiman, "Fear of Crime in the United States: A Multivariate Analysis," Social Forces, Vol. 56 (September- June 1977/78). 18Garofalo, "Victimization." 19wesiey G. Skogan, "Public Policy and the Fear of Crime in Large American Cities," Public Law and Public Policy, ed. John A. Gardiner (New York: Praeger Publishers, 1977): p. 7. 20Skogan briefly mentions the psychological role that a dwelling plays in maintaining one's sense of security and order. For a further discussion of this tapic, see Lee Rainwater, "Fear and the House-as- Haven in the Lower Class," Journal of the American Institute of Planners 32 (January 1966): 23-31. 21Garofalo, “Victimization." 22Victoria H. Jaycox, "The Elderly's Fear of Crime: Rational or Irrational?," Victimolqu: An International Journal 3 (1978): 329. 23Skogan and Maxfield, Coping With Crime, p. 69. 24Stephanie Riger, Margaret T. Gordon, and Robert LeBailly, "Women's Fear of Crime: From Blaming to Restricting the Victim," Victimology: An International Journal 3 (1978): 274. 25Jane Jacobs, The Death and Life of Great American Cities (New York: Random House, 1961). 261bid.. pp. 31-32. 27The President's Commission on Law Enforcement and Administra- tion of Justice, The Challenge of Crime in a Free Society (Washington, DC: United State Government Printing Office, 1967), p. 6. 28Boggs, "Crime Control." 63 29Dan A. Lewis and Michael G. Maxfield, "Fear in the Neighbor- hoods: An Investigation of the Impact of Crime," Journal of Research in Crime and Delinquency 17 (1980): 160-189. 30John E. Conklin, "Dimensions of Community Response to the Crime Problem," Crime and Justice (New York: AMS Press, 1970/71), pp. 27-38. 31Conkiin, The Impact of Crime. 32Wesley G. Skogan, “On Attitudes and Behaviors," Reactions to Crime, ed. Dan A. Lewis (Beverly Hills: Sage Publications, 1981), p. 19. 33Stephanie Riger, "On Women,“ 1981. 34Margaret T. Gordon and Linda Heath, "The News Business, Crime, and Fear," Reactions to Crime, ed. Dan A Lewis (Beverly Hills: Sage Publications, 1981), p. 227. 351t has been accepted by several researchers that media por- trayals of crime do not accurately refelct the total amount and type of crime that actually occurs. Se James F. Davic, "Crime News in Colo- rado Newspapers," American Journal of Sociology 57 (November, 1952): 325-330; Joseph R. Dominick, “Crime and Law Enforcement on Prime Time Television," Public Opinion Quarterly 37 (Summer 1973): 241-250; Bob Roshier, "The Selection of Crime News by the Press,“ The Manufacturer of News, eds. S. Cohen and J. Young (Beverly Hills: Sage Publications, 1573), pp. 28-39. 36Jeffrey Henig and Michael G. Maxfield, "Reducing Fear of Crime: Strategies for Intervention," Victimology: An International Journal 3 (1978): 297. 37James M. Tien, Vincent O'Donnell, Arnold Barnett and Pitu Mirchandani, Street Lighting Projects: National Evaluation Program, Phase 1, Final Report (Cambridge, MA: Public Systems Evaluation, Inc., 1977). 381bid., p. 93. 39Oscar Newman, Defensible Space: Crime Prevention Through Urban Design (New York: Macmillan, 1972); Architectural Design for Crime Prevention (New York: New York University, March 1973), U.S. Govern- ment Printing Office. 40This article was reported in ”Reducing Fear of Crime: Strate- gies for Intervention," by Jeffrey Henig and Michael G. Maxfield. , 41Herb Rubenstein, Charles Murray, Tetsuro Motoyama, and W. V. Rouse, The Link Between Crime and the Built Environment, Vol. 1, (National Institute of Justice: U.S. Government Printing Office, 1980). 64 42Thomas Molumby, "Patterns of Crime in a University Housing Project," American Behavioral Scientist 20 (Beverly Hills: Sage Pub- lications, 1976): 274-279. CHAPTER 111 RESEARCH PROCEDURES gem One purpose of this chapter is to describe the research approach and design. The approach will describe the basic method in which the data for this study were collected, whereas the design will define the variables as dependent, independent or controlling (personal character- istics). In addition, the research design section will include Opera- tional definitions, the numerical codes that were used in the data coding stage. Operational definitions will aid in data interpretation by indicating the direction of the relationship between variables. A section dealing with the survey site and samples and popula- tion follows the research approach and design. The survey site was the Michigan State University campus. The campus is the location for which the perceptions and experiences concerning the fear of crime are being studied. The demographic composition of the sample will be presented, along with a comparison to the known papulation characteristic. In addition, the residence halls that comprise each complex will be pre- sented. The data collection and recording section consists of an overview of the survey package, follow-up postcard, survey distribution, and data coding. This section naturally leads to the methods of data analysis. Here we are concerned with providing a brief description of how the 65 66 analysis of the data Chapter is laid out. Finally, the methodological assumptions and the limitations of the study are addressed. Research Approach The primary sources of information for this study were three mailed surveys. The first survey was distributed in January of 1979, the second in January of 1980, and the third in January of 1981. Var- ious limits such as time, money and personnel excluded the use of per- sonal interviewing in the research approach. Due to the personal nature of several questions in the surveys, it was presumed that a more honest or accurate response would be received if the survey was anonymous. Herbert H. Hyman, a reCOgnized authority on the interviewing technique, found that due to anonymity, respondents are more willing to provide answers to questions in surveys than in personal interviews.1 According to Hyman, questionnaires also give respondents more time to think, and may actually reduce bias in situations where the respondent may want to impress the interviewer. Disadvantages that generally accompany a mailed survey are low return rates and nonresponse to individual questions. To minimize these disadvantages, anonymity guarantees, follow-up postcards and design of the survey format were stressed in this study.2 Research Design The dependent variable is considered to be the perception of the fear of crime by each individual or respondent. The question that was used by the National Crime Panel Surveys to measure the fear of crime was modified slightly to reflect the 'MSU campus' as the respondent's 'neighborhood.‘ Responses to the question, "How safe do you feel it 67 is to be out alone on the MSU campus at night?" were very safe (coded as 1), reasonably safe (coded as 2), somewhat unsafe (coded as 3), and very unsafe (coded as 4). A similar question followed which asked the respondents their perception of safety during the day. Respondents who indicated somewhat unsafe or very unsafe were considered to have a fear of crime on the MSU campus. These two foils are sometimes combined to represent the 'feeling of unsafety' due to crime or the 'perception of unsafety.‘ The determinants of the fear of crime are also referred to as the independent or predictor variables. A statement expressing each determinant was developed. In response to the question, "If you feel unsafe, what are the reasons?" the respondent was asked to check all of the reasons that apply. The following statements were intended to represent the determinants. Direct Victimization--"Past criminal incidents that have occurred to me." Indirect or Vicarious Victimization--"Because of criminal incidents that have happened to friends." Media Effect-~"From what I've read in newspapers, heard on radio or television." Psychological--"I'm afraid of being victimized anywhere." Vulnerability--"I feel unable to defend myself." Inadequate Formal Control--"I feel the police are inade- quate in providing protection," or "I don't see the police on campus very often when walking from building to build- ing." Environment--"There are various places on campus that are poorly lit and have an excess of shrubbery and bushes." Perception of Extent of Crime--"The general feeling and attitude among students is that a lot of crime occurs on the MSU campus." 68 To determine the relative strength or importance of each determinant, the respondents were also asked to indicate the most important reason why they felt unsafe. This required the respondent to indicate the major cause of their fear from among the foils already checked. Information concerning sex, race, age, Class level, home town and current residence was collected as personal characteristics or control- ling variables. Sex was naturally broken down as male (coded as 1) and female (coded as 0). While race was coded as White (1), Black (2), Latin American (3), Chinese (4), American Indian (5), Japanese (6) and other (7), this variable was dichotomized to include only White (recoded as 1) and Nonwhite (recoded as 2) because of the relatively few respon- dents falling into the latter six categories. However, the collapsing of Nonwhite categories did not come close to equaling the size of the White category. Class level consisted of four categories: freshman (1), sopho- more (2), junior (3), and senior (4). This variable is used to deter- mine if first-term students' (generally freshmen) perceptions differ from perceptions of students who have resided on campus for more than one school year. It was assumed that class level would identify more accurately than age those respondents who were residing on the MSU cam- pus for their first school year. For the purposes of analysis, Class level was dichotomized into freshman (recoded as 1) and sophomore- senior (recoded as 2). In the 1981 study, the reSpondent was asked to classify their home town as urban (coded as 1), suburban (coded as 2) or rural (coded as 3). This variable was important in determining if the respondent's perception of campus safety could be affected by their home town 69 experiences and perceptions. The respondents were also asked a question concerning their perception of safety at night in their home town. This variable is used to determine if a greater (or lesser) percentage of respondents feel more (or less) safe while on campus, compared to their home town. The current residence hall in which the respondent resided was used to determine if the fear of crime was more prevalent in one com- munity or complex than another. Twenty-six residence halls were divided into five distinct complexes. The distinction is based primarily upon geographical location. The surveys also contained a section dealing with the actual cam- pus victimization experience of the respondent. A reference period for the occurrence of the criminal event was defined as the prior 'fall term,‘ and the MSU campus was required for the location of the incident. Although the boundaries of the campus were not specifically defined for the respondents, the location of the incident was collected and used for analysis in determining if the criminal event occurred on campus. The reference period for the campus victimization experience was limited to fall term (approximately three and one-half months) for several reasons. If the period was lengthened, this would have limited input from the freshman class. Freshmen were entering their second term at the time the questionnaires were distributed. Second, fall term is very identifiable to a majority of the students. This term is the first term for freshmen, or generally the first term of the school year for most students. Third, a relatively short reference period would not hinder the recall of recent victimization experiences. Victim recall, or the ability of respondents to remember incidents that have occurred 70 to them, has been a concern in victimization surveys which have an extended reference time frame. A related concern has been the phenom- enon known as telescoping. Telescoping occurs when the respondents include incidents that have actually occurred outside the reference time frame. By limiting the reference time frame to include only fall term, victim recall and telesc0ping is most likely minimized. Respondents of the 1980 survey were also asked if they had "ever been a victim of a personal crime." This question was used to further test the direct victimization determinant by lengthening the reference period. It was assumed that respondents would recall the incidents that were of significant nature to them. The victimization experience was collected by asking the respon- dent to check 'yes' or 'no' to whether they were a victim to a robbery, attempted robbery, or physical attack. Three questions concerning property crimes were also asked. The 'yes' responses were coded as one (1) and the 'no' responses were coded as zero (0). The following ques- tions were used to collect the basic criminal victimization data. Personal Crimes: Robbery--"During fall term, did anyone actually take some- thing directly from you by using force, such as a stickup, mugging or threat?" Attempted Robbery--"During fall term, did anyone tgy to take something directly from you by using force or threat- ening to harm you?" Physical Attack--"During fall term, did anyone physically attack you, knock you down or beat you up?“ Property Crimes: Theft from Residence Room--"During fall term, was anything stolen from your dorm room while you were away?" 71 Theft from Other Campus Location--"During fall term, was anything (else) stolen from you on campus?“ Theft from Vehicle--"During fall term, was anything stolen from your vehicle while it was parked on campus?" Without the use of personal interviewing, specific data concern- ing the criminal incident had to be omitted. For example, the degree to which the person was physically attacked was omitted, which excluded differentiating between simple and aggravated assault. In reference to thefts from residence rooms, the 'element of force' could not be ade- quately identified in the survey approach. This disallowed the distinc- tion between such crimes as burglary and larceny. For the purposes of this study, these distinctions are somewhat unimportant. The major goal was to identify the respondents who perceived themselves as victims of either a personal crime or a property crime. The legal technicalities of whether they were actual victims is not essential as long as their perception of victimization is established. For the analyses, personal crime victims were recoded as one (1), pr0perty crime victims were recoded as two (2) and nonvictims were recoded as nine (9). Survey Site The survey site for this study is the Michigan State University campus. With an on-campus population of approximately 17,000 under- graduate students, the campus is the smallest populated community in the area. To the north and east of the campus is the city of East Lansing, with an approximate p0pulation of 48,000. Located to the west of campus is the city of Lansing, with an approximate p0pulation of 130,000. Although the campus is located adjacent to an urban area, several environmental characteristics depict the campus as a distinct 72 community. The campus is physically located on the perimeter of the urban area, unlike some college campuses that are located within an urban area. With no major highways or streets directly connecting the interior of the campus with the urban area, the daily traffic on campus is generally related to campus activity. General observations of the campus environment characterize it as a park. An abundance of trees, wooded areas, a small river and numerous flower gardens on campus dis- tinguish it from the adjacent urban area. In addition, the mode of transportation for the campus residents is walking or bicycling. Finally, the campus and the adjacent cities have a relatively low reported crime rate. In 1980, the number of reported Part I crimes on campus was 2,571.3 East Lansing had approximately 1,432 reported Part I crimes, while Lansing reported 9,515 Part I crimes. While the Part I crime index includes property related crimes such as motor vehicle theft, larceny/theft and burglary, the number of violent crimes for each area is substantially lower. Combining the crimes of murder and non- negligent manslaughter, forcible rape, robbery and aggravated assault, the 1980 Uniform Crime Report reported only 53 violent crimes on campus, 44 in East Lansing and 735 in Lansing. It should be noted that in addition to the residence halls, a number of other campus buildings, consisting of classrooms and offices, are also present. A vast majority of these structures are located in the north half of campus, while the south part of campus primarily con- sists of farm and grazing land. Finally, it is important to note that Michigan State University has its own Department of Public Safety. Included within the Department of Public Safety is the Police Services Division, commonly known as the 73 campus police department. Samples and Population The samples were chosen from a p0pulation consisting of under- graduate students residing on the Michigan State University campus dur- ing the fall terms of 1978, 1979, and 1980. The undergraduate residence halls were grouped by geographic location into five complexes. Each complex and the halls that comprise them are listed below. Brody Complex: Armstrong, Bailey, Bryan, Emmons, Butter- field, and Rather West Circle Complex: Campbell, Landon, Mayo, Williams, Gilchrist, and Yakeley Red Cedar Complex: Mason, Abbot, Phillips, Snyder, and Shaw East Complex: Akers, Fee, Holmes, Hubbard, and McDonel South Complex: Case, Wilson, Wonders, and Holden With the campus stratified into five complexes, a random sample of each complex was generated. Complexes that had smaller populations were sampled at a higher rate in order to obtain an adequate number of respondents. Table 3.1 provides the p0pulation and sample of each com- plex for the 1979, 1980, and 1981 studies. Because the samples were actually chosen from the prior fall terms, the p0pulation shown is for the fall terms. Also included in Table 3.1 are the number of returns and return rates. Out of the 3,000 surveys distributed, 1,850 usable responses were returned from the 1979 sample, for a 61.7 percent return rate. A similar return rate of 62.3 percent was obtained from the 1980 sample. Although these somewhat high return rates were welcomed, the potential bias due to the nonresponse rate cannot be ignored. This 74 Table 3.1.--Distribution of Surveys and Response Rates (1979, 1980, 1981) Complex Category By West Red Survey Year Brody Circle Cedar East South TOTAL Population 1979 2,832 1,592 2,620 5,193 4,598 16,835 1980 2,889 1,620 2,597 5,295 4.719 17.120 1981 2,906 1,660 2,660 5,314 4,778 17,318 Sample 1979 550 400 550 800 700 3,000 1980 550 400 550 800 700 3.000 1981 550 400 550 800 700 3,000 Sample Return 1979 322 252 336 493 420 1,850* 1980 342 262 345 508 413 1,892* 1981 255 214 290 384 363 1,551* Return Rate 1979 58.5% 63.0% 61.1% 61.6% 60.0% 61.7% 1980 62.2% 65.5% 62.7% 63.5% 59.0% 62.3% 1981 46.4% 53.5% 52.7% 48.0% 51.9% 51.7% *Complex sample returns do not equal TOTAL sample return due to missing data. 75 issue will be addressed in a following section. The 1981 study had a somewhat lower return rate of 51.7 percent. A change in the survey format may have affected the return rate. For the surveys conducted in 1979 and 1980, the format consisted of a seven-page survey with items printed on only one side of the page. The 1981 survey format consisted primarily of the same questions. However, due to limited finances, the size of the print was reduced and the sur- vey items were printed on both sides of the pages. These cost-cutting measures reduced each survey to only two 11")<14" pages. In general, the clarity and neatness of the survey format was reduced by condensing the survey items. The demographic composition of the three samples are presented in Table 3.2. The variables sex, race and class level are presented as they are used in the analysis of data. Race was dichotomized into White and Nonwhite, and class level was dichotomized into Freshman and Sophomore-Senior. While age is not used in any analysis, it is pre- sented in Table 3.2 for informational purposes. A few students failed to reSpond to specific items on the questionnaire. These are referred to in the table as 'Blanks.’ The blanks are included as raw data (N) to account for the total number of respondents in each study. However, in the percentages and in the analyses on the following pages, missing data were excluded. A review of the data presented in Table 3.2, in conjunction with the review of literature, indicates that the use of the variable sex should be observed with caution. Females were often highly associated with the fear of crime or feeling of unsafety, and the composition of Sex in the samples is more females than males. To offset any possible 76 Table 3.2.--Demographic Characteristics of Survey Respondents (1979, 1980, 1981) Survey Year Demographic 1979 1980 1981 Characteristics (N) Percent (N) Percent (N) Percent Sex Male (793) 43.4% (822) 43.8% (611) 39.4% Female (1,033) 56.6% (1,054) 56.2% (940) 60.6% Blank (24) (16) (0) Race White (1,723) 94.7% (1,770) 94.8% (1,435) 93.4% Nonwhite (96) 5.3% (98) 5.2% (101) 6.6% Blank (31) (24) (15) Class Level Freshman (667) 36.6% (750) 40.1% (569) 36.9% Sophomore- Senior (1,115) 63.4% (1,118) 59.9% (971) 63.1% Blank (68) (24) (11) Age 17 (9) .5% (12) .6% (9) .6% 18 (483) 26.5% (544) 29.1% (395) 25.9% 19 (620) 33.9% (628) 33.6% (562) 36.8% 20 (399) 21.9% (380) 20.3% (316) 20.7% 21 (211) 11.6% (196) 10.5% (155) 10.2% 22 (65) 3.6% (68) 3.6% (59) 3.9% 23 and over (37) 2.0% (40) 2.3% (31) 1.9% Blank (26) (24) (24) 77 effects that sex may have on various analyses, sex will be used as a control variable where appropriate. The only known population characteristic used in comparison with the sample characteristics is sex. In Table 3.3, the percentage of male and female students in the student population residing on campus is com- pared with the sample distributions. The undergraduate student popula- tion residing on campus during the fall terms of 1978, 1979, and 1980 were the sampling frames. The Characteristic sex was distributed quite evenly throughout the population for each time period, with females and males each accounting for approximately 50 percent of the population. As indicated by the survey samples, females are overrepresented compared to the student populations. This could be a problem if inferential statistics were being used; however, the analysis of data was basically limited to summing, percentages and bivariate analysis, while control- ling for the variable sex in many instances. Table 3.3.--Comparison of Survey Samples and Student P0pulations for the Variable Sex (1979, 1980, 1981) Survey Year By Sex Survey Sample . Student Population 1979 Study Male 43.4% 50.7% Female 56.6% 49.3% 1980 Study Male 43.8% 49.9% Female 56.2% 50.1% 1981 Study Male 39.4% 48.8% Female 60.6% 51.2% 78 Data Collection and Recording During this phase of the study, the surveys were distributed, received and prepared for analysis. Along with the survey, a self- addressed return envel0pe and a letter from the project director stress- ing the importance of the study, the respondent's anonymity and the importance of individual response were also included. The utilization of the campus mail system enabled the respondents to return the surveys in the self-addressed envelopes without postage. Four days after the initial mailing, a follow-up postcard was sent to the selected students. Because the survey was anonymous, the follow-up postcard served two purposes. First, it thanked those who had completed and returned the survey, and secondly, it encouraged those who did not yet return the survey to do so. The returns were received through bulk mail over a period of two weeks, which did not allow the effect of the follow-up postcard upon the return rate to be observed. After receiving the surveys, a codebook was devel0ped which listed all data items and their numerical codes. In the 1979 study the data items were coded on standard BO-column Fortran coding forms. The coding process was performed primarily by the project director in order to standardize any interpretation of the data and to reduce coding errors.4 The data was keypunched onto computer cards by a professional keypunching service provided at the MSU Computer Laboratory. This allowed the data to be transferred to a computer tape for analysis. The data coding for the 1980 and 1981 study was conducted some- what differently. With the use of a codebook, the data was coded onto data sheets or frequently called 'mark-sense' sheets. Data cards were generated by running the data sheets through an optical reader, a 79 service provided by the MSU Scoring Office. Finally, the data was transferred from cards to computer tape for ease of analysis. Analysis The first step in the analysis of data was to examine the level of fear. Data concerning how safe the respondents feel being out alone on campus during the day and at night will be presented. The wide mar- gin between the percentage of respondents who feel somewhat unsafe or very unsafe during the day, compared to at night, leaves the remaining analyses concerned only with the 'at night' question as the dependent variable. Several indirect measures concerning the seriousness of the fear of crime problem will also be presented in this section. The per- centage of respondents who felt somewhat unsafe or very unsafe in each of the three studies will be compared to results obtained in the National Crime Surveys. Respondents of the 1980 study were also asked what actions they have taken as a result of their perception of unsafety (combining somewhat unsafe and very unsafe). The responses to this question will be presented. The second section consists primarily of crosstabulating the per- sonal characteristics with the dependent variable. These analyses will determine which subgroups are associated with the fear of crime. The personal characteristics are sex, race, class level, residence complex and size of home town. In addition, a bivariate analysis will be con- ducted between the victimization status of the respondents and the dependent variable. In the 1981 study the respondents were asked how safe they felt being out alone at night in their home town. The reSponses to this question, which ranged from very safe to very unsafe, 80 were also crosstabulated with the dependent variable. The final section involves an analysis of the determinants. Fre- quency tables will be presented for male and female respondents in ref- erence to the most important reason why the respondents feel unsafe while walking alone at night on campus. In the 1980 study, several additional questions were asked to explore several of the determinants. Related to the direct victimization determinant was a question asking the respondents if they had ever been a victim of a personal crime. The responses to this question were 'yes' (coded as 1) and 'no' (coded as 0). This variable was crosstabulated with the dependent variable. In reference to the indirect victimization determinant, the res- pondents were asked if they had known anybody who was a victim of a personal crime. This question expanded the search for indirect or vicarious victims and the responses were crosstabulated with the depen- dent variable. The respondents were also asked how serious crime was, relative to what the media indicates. Respondents were able to indicate whether crime was less serious (coded as 1), about the same (coded as 2) or more serious (coded as 3) than what the media depicts. If the media is relied upon by the respondent or if the media has an impact on the res- pondent's perception of crime, then it would seem logical that the respondent would indicate that the seriousness of crime is about the same as the media indicates. If the reSpondent indicated that the crime level was less serious than what the media indicates, this would imply that the respondent did not perceive the media as providing an accurate picture or account of the crime level. Respondents could also perceive crime as being more serious than what is portrayed in the media. This 81 could imply that respondents gain knowledge of criminal events occurring in the environment, but are also aware of additional criminal events through other communication channels. The crosstabulation of this var- iable with the perception of safety will determine if respondents who perceive the level of crime as being about the same or more serious than what the media indicates are more fearful than respondents who perceive the level of crime as being less serious. In reference to the determinant concerning the environment, the major issue was the existence of poor lighting and excessive shrubbery and bushes which provide blind spots and hiding places. The respondents in the 1980 study were provided a list of ten locations on campus and were asked to indicate which areas or locations they would like to see either increased lighting or modification or removal of shrubbery. This question gave an overall indication that students preferred to see an increase in lighting instead of a modification of the campus landscape. In contrast to the concept of formal control is the concept of informal control. A social cohesion index was developed where respon- dents were classified as having low social cohesion, medium social cohesion or high social cohesion.5 In the 1980 study, respondents were asked the following three questions: "On some dormitory floors people do things together and help each other; in others people mostly go their own ways. In general, what kind of floor would you say you live on-- one where people help each other, or one where people go their own ways?" 1. Help each other 2. Go their own ways 82 "How many of those people that live near you (on your floor or otherwise close by) do you feel you know well enough to ask a favor of if you needed something--would you say most of them, some of them or almost none of them?" 1. Most 2. Some 3. Almost none "How long have you lived in this dorm?“ Respondents classified as having low social cohesion indicated the res- ponses 'go their own ways,’ 'almost none,‘ and lived in their residence hall for only one term. Any respondent who resided in their residence hall for two or three terms and indicated the response 'some' were assigned to the category of medium social cohesion. Finally, the high social cohesion category consisted of respondents who resided in their residence hall for four or more terms, and who indicated the responses 'help each other' and 'most.' This variable was crosstabulated with the perception of safety at night to determine if respondents who were classified as having high social cohesion had a lower fear of crime than respondents who were classified as having low social cohesion. To explore the determinant concerning the respondent's perception of crime on campus, respondents were asked what they perceived their chances of being sexually assaulted on campus were. This question was crosstabulated with the dependent variable to indicate that a large per- centage of female respondents who perceived their chances of being sexu- ally assaulted as very high or reasonably high also felt somewhat unsafe or very unsafe on campus at night. For each bivariate analysis conducted, the Significance levels of the tests will be presented. A significance level of less than .1 is regarded as representing a genuine or significant association between 83 the two variables. In other words, the possibility that the association resulted from sampling error is discounted if the significance level is less than .1. The conditions in which sex is used as a control variable should be noted. In situations where the first level analysis is nonsignifi- cant and the second level (controlling for sex) is significant, the bivariate analyses controlling for sex will be presented. For brevity, in situations where nonsignificant associations are present at the first and second levels of analysis, then the first level bivariate analysis will be simply presented. Finally, in the section dealing with the determinants of the fear of crime, only slight differences were found between males and females concerning their most important reason for feeling somewhat unsafe or very unsafe. The lack of major differences in this particular research question diminished the concern of control- ling for sex in the subsequent related research issues. Methodological Assumptions A major assumption in the study concerns the representativeness of the respondents versus the nonrespondents. In the area of criminal victimization, it is assumed that nonrespondents were not victims of crimes; particularly, personal crimes. James Walker Fox, in a criminal victimization and Opinion study among students at the University of Virginia, used a mailed questionnaire as a screening device to identify victims of six Specific crimes.6 Of the 2,002 students sampled, 1,345 usable responses were returned, for a 66 percent return rate. A follow- up telephone interview of nonreSpondents revealed some interesting find- ings. Of the nonreSpondents, 98.2 percent had not been victimized, and 84 the remaining 1.8 percent who were victimized considered the incident so minor that they would not classify it as a crime. Although transfer- ring this finding to this study is difficult, it does provide support for the idea that nonrespondents may not be representative of respon- dents in terms of actual victimization experience. It is also important to note that the victimization data is only used to examine the fear level between victims and nonvictims. The use of percentages in univariate and bivariate analysis will eliminate the concern of bias due to nonresponse, unlike that which would occur if inferential statistics were being applied to the victimization data. In terms of the dependent variable, the representativeness bet- ween respondents and nonrespondents is not a major concern. It does not seem likely that nonrespondents would single out the dependent var- iable as a deciding factor for not returning the survey. In other words, it would not be expected that individuals would first determine that they do not feel unsafe while walking alone on campus at night, and consequently, not return the survey. Limitations Although several advantages of excluding personal interviewing from the study were mentioned in the research approach, the benefits of using interviewing in addition to a survey are hard to ignore. There may always be some degree of nonresponse, but personal interviewing of the nonrespondents as to why they did not respond would be useful in clarifying the various assumptions concerning sample representativeness. Personal interviewing would also enable the researcher to collect more specific data from the respondents. \t-n . 1.1.11 .'-l. h...- 5 hr" 3 “I If? ‘ . . I I o ' . . . r u '1 45...... wum'mm *—— 85 Because a specific environment was chosen for this study, the most obvious limiting factor is the transferability of the findings. The Michigan State University campus is not representative of the sur- rounding areas or other college campuses. The research methodology may be used to perform similar studies in other environments, but the find- ings can not be justifiably used for decision making in other areas. Furthermore, this study only surveyed students that were living on cam- pus. Professors, off-campus students, other campus workers and visitors were excluded from the survey population, further limiting the findings. Finally, the personal knowledge, imagination and ability of the researcher did not permit the survey design, data coding and data anal- ysis to be free of error. It is recognized and welcomed that others may improve the research process and that the study of the fear of crime on the Michigan State University campus can be continued. Summar This study is primarily descriptive in nature and is aimed at describing the relationship between various controlling and independent variables and the fear of crime. The study actually consists of three separate studies or surveys that were conducted on the Michigan State University campus in January of 1979, 1980, and 1981. The population or sampling frame consists of the undergraduate students residing on the Michigan State University campus. The scope of this study was con- cerned with three major research questions: How prevalent is the fear of crime among the p0pulation studied?, Who is experiencing the fear of crime (in terms of personal characteristics)?, and What are the determinants of the fear of crime? A large sample size and the 86 methodology used in each survey were satisfactory for gathering data to provide adequate answers to these research questions. Chapter IV consists of a presentation of the survey data. Var- ious frequency tables and bivariate analyses of the survey data are presented. CHAPTER III FOOTNOTES 1Herbert H. Hyman, Interviewing in Social Research (Chicago: University of Chicago Press, 1954). See also Helen Metzner and Floyd Mann, "A Limited Comparison of Two Methods of Data Collections: The Fixed Alternative Questionnaire and the Open-Ended Interview," American Sociological Review 17 (August 1974): 486-491; Albert Ellis, "Ques- tionnaires vs. Interview Method in the Study of Human Lover Relation- ships," American Sociological Review 13 (January 1948): 61-65. 2Don A. Dillman, "Increasing Mail Questionnaire ReSponse in Large Samples of the General Public," Public Opinion Quarterly 36 (Summer 1972): 254-257. 3Part I crimes consist of murder and non-negligent manslaughter, forcible rape, robbery, aggravated assault, burglary, larceny/theft and motor vehicle theft. Data concerning Part I crimes were obtained from Crime in the United States: Uniform Crime Reports, 1980 (printed annually by the Federal Bureau of Investigation). 4As indicated by the survey questions presented in this chapter, little or no interpretation of the responses was needed. It was also felt that a more conscious effort would be put in the coding by the researcher because of the personal involvement or commitment in the studies. 5Adapted from Thomas A. Repetto, Residential Crime (Cambridge, MA: Ballinger Publishing Co., 1974). 6James Walker Fox, Criminal Victimization Among Students at a Selected University (Ph.D. Dissertation, University of Virginia, 1975). 87 CHAPTER IV ANALYSIS OF THE DATA In this chapter, an analysis of data obtained by surveying stu- dents residing on the Michigan State University campus is presented. The data collected is related to the fear of crime among the population being studied. The analysis of the data is presented within the frame- work of the research questions, as discussed in Chapters I and II. How Prevalent is the Fear of Crime? One of the primary purposes of this research is to determine the prevalency or extent of the fear of crime among the students residing on the MSU campus. The method for obtaining information concerning the fear of crime is very similar to the method that has been used in national surveys and public opinion polls. The respondent's perception of safety in relation to crime was collected for two reference periods: during the day and at night. In Table 4.1, the distribution of responses concerning the res- pondent's perception of safety during the day is presented. It can be seen that the vast majority of respondents feel very safe or reason- ably safe during the day.; The percentage of respondents feeling safe (combining very safe and reasonably safe) ranged from 98.0 percent in 1979 to 98.9 percent in 1981. In general, only 1.6 percent of the TOTAL number of respondents indicated that they feel any degree of unsafety (combining somewhat unsafe and very unsafe) during the day. 88 89 Table 4.1.--Respondent's Perception of Safety During the Day (1979, 1980, 1981) Survey Year Perception of Safety 1979 1980 1981 TOTAL Very Safe 64.3% 66.7% 62.8% 64.7% Reasonably Safe 33.7 31.7 36.1 33.7 Somewhat Unsafe 1.6 1.4 1.1 1.4 Very Unsafe .3 .2 .1 .2 TOTALS 100.0% 100.0% 100.1% 100.0% Number of Responses 1,836 1,890 1,516 5,242 As expected, the respondent's perception of safety shifts dramat- ically when the reference period is 'at night.’ Data portraying the respondent's perception of safety at night is presented in Table 4.2. The percentage of respondents who indicated any degree of unsafety at night (combining somewhat unsafe and very unsafe) ranged from 51.1 per- cent in 1980 to 58.2 percent in 1979. Table 4.2.--Respondent's Perception ofSSafety at Night (1979, 1980, 1981 Survey Year Perception of Safety 1979 1980 1981 TOTAL Very Safe 9.4% 10.6% 9.4% 9.8% Reasonably Safe 32.3 38.3 37.4 35.9 Somewhat Unsafe 33.8 31.5 35.8 33.6 Very Unsafe 24.4 19.6 17.4 20.7 TOTALS 100.0% 100.0% 100.0% 100.0% Number of ReSponses 1,837 1,890 1,518 5,245 90 The percentage of respondents indicating any degree of unsafety at night was compared to the results obtained in the eight impact cities during the National Crime Surveys.1 While different time periods and research methodology limit the conclusions that can be drawn from the data in Table 4.3, the information is presented to indicate that a sig- nificant proportion of the campus residents are associated with the fear of crime. Although the locations are ranked in descending order by the percentage of respondents expressing some degree of unsafety, it would be invalid to conclude from the data that campus residents 'feel unsafer' or that a greater percentage of campus residents 'feel unsafer' than residents of these selected cities. Nevertheless, for informational purposes, a large percentage of the campus residents harbor the fear of crime or feeling of unsafety while out alone at night on campus. In addition, the data supports the need for analysis at the neighborhood or community level. The fear of crime is not only found at the large city or urban level, but is also present within loca- tions consisting of smaller populations and varying environmental characteristics. The degree to which the reSpondents are affected by the fear of crime can be measured indirectly by asking the respondents what actions they have taken as a result of their perception of unsafety. A list of seven actions was presented and the respondents who indicated any degree of unsafety were asked to indicate which actions they have taken as a result of their perception of unsafety. As indicated in Table 4.4, one of the possible responses was 'nothing.’ Of all the responses, only 5.8 percent constituted this response. This relatively low percentage implies that the respondent's fear of crime or perception 91 Table 4.3.--Respondents Expressing Some Degree of Unsafety for MSU Campus and Eight Impact Cities (1979, 1980, 1981) Percent Responding Somewhat Location Unsafe or Very Unsafe MSU Campus (1979) . 58% Newark 58 MSU Campus (1981) 53 MSU Campus (1980) 51 Baltimore 51 Cleveland 49 St. Louis 47 Atlanta 46 Portland 38 Denver 37 Dallas 36 Table 4.4.--Actions Taken as a Result of Perceptions of Unsafety (Com- bining Somewhat Unsafe and Very Unsafe), 1980 Actions Taken Number of ReSponses Percent Nothing 90 5.8% Only Go Out with Friends 727 46.7 Use Dial-A-Ride 70 4.5 Use a Campus Escort Service 59 3.8 Never or Seldom Go Out at Night 457 29.4 Carry a Weapon 98 6.3 Take Self-Defense Classes 54 3.5 TOTALS 1,555 100.0% 92 of unsafety is serious enough to result in a Change in behavior. The most common behavioral change or action taken was the indica- tion that respondents 'only go out with friends.‘ 'Safety in numbers' appears to be a common expression, when 46.7 percent of the responses were attributed to this particular item. Two other responses which imply that the reSpondents only go out at night when escorted are use Dial-A-Ride' and 'use a campus escort service.‘ These two programs were established on campus to provide transportation or escorts in lieu of walking alone on campus. Combining these two responses, it can be seen in Table 4.4 that 8.3 percent of the responses are attributed to these two items. ' The second most common action consisted of the respondents' 'never or seldom going out at night.’ Approximately 29 percent of the respondents indicated this response. This particular reSponse, along with the response 'only go out with friends,‘ accounted for 76 percent of the total responses. These reSponses imply that the fear of crime has had a profound impact on the respondent's behavior concerning going out alone at night on campus. It is also important to note that a fairly large percentage of the respondents either 'carry a weapon' or 'participate in self-defense classes' for self-protection. These nonpassive actions accounted for 9.8 percent of the total responses. Who is Experiencing the Fear of Crime? The purpose of this section is to examine the fear of crime for each subgroup of the sample. The demographic or personal character- istics of sex, race, class level, residence complex, victimization 93 experience and size of home town will be crosstabulated with the percep- tion of safety. In addition, the reSpondent's perception of safety in their home town will be compared with their perception of safety on campus. The variable sex has often been the best predictor variable of the fear of crime. As it has been consistently reported in other studies, a higher percentage of females tend to feel somewhat unsafe or very unsafe while walking alone at night than males. The data in Table 4.5 supports this finding. In the 1979 study, 86.4 percent of the female respondents expressed some degree of fear or unsafety of being out alone at night, while only 22.3 percent of the male respon- dents expressed a comparable degree of fear. The findings for the 1980 and 1981 studies are similar. In 1980, the percentage of females expressing any degree of unsafety was 79.7 percent, while 78.0 percent was observed in 1981. The percentage of males responding somewhat unsafe or very unsafe were 14.2 percent and 15.0 percent in the 1980 and 1981 studies, respectively. The gamma statistics indicate that the variable sex has a very strong association with the perception of safety. The high chi-square statistics with three degrees of freedom yields a significance level of less than .001 for the three studies. It is interesting to note that for males, and particularly females, there was a reduction in the fear of crime from 1979 to 1981. Comparing the very unsafe responses for the females, it can be seen that 41.8 percent of the females felt very unsafe in 1979, compared to 27.8 percent in 1981. In the somewhat unsafe category, 44.6 percent of the females indicated this response in 1979, compared to 50.2 percent in 1981. Furthermore, a higher percentage of females felt reasonably 94 Table 4.5.--Perception of Safety at Night by Sex (1979, 1980, 1981) Perception of Safety Survey Year by Sex 1979 1980 1981 Very Safe Male 20.1% 23.5% 20.8% Female 1.1 .7 2.0 Reasonably Safe Male 57.6 62.2 64.2 Female 12.5 19.7 20.1 Somewhat Unsafe Male 20.1 11.9 13.7 Female 44.6 47.0 50.2 Very Unsafe Male 2.2 2.3 1.3 Female 41.8 32.7 27.8 TOTALS Male 100.0% 99.9% 100.0% Female 100.0% 100.1% 100.1% Number of Responses Male 786 821 597 Female 1,028 1,053 921 ' 1979: Chi-Square = 818.20 Gamma = -.89 Significance = (.001 1980: Chi-Square = 842.65 Gamma = -.89 Significance = <.OOl 1981: Chi-Square = 604.96 Gamma = -.88 Significance = (.001 95 safe in 1981 compared to 1979. It can be speculated that the higher level of fear of crime in 1979, compared to 1980 and 1981, was a result of a series of widely publicized crimes that occurred around the area prior to the 1979 survey. The association between race and perception of safety appears to be insignificant, as indicated by the data and statistics presented in Table 4.6. In the 1979 and 1980 studies, a higher percentage of nonwhites felt somewhat or very unsafe; however, the small number of nonwhite respondents makes the interpretation somewhat meaningless. In 1979, 69.4 percent of the 95 nonwhite respondents reported feeling somewhat or very unsafe while out alone at night on campus. This is compared to 58 percent of the 1,712 white respondents. Despite the percentage difference, the significance level for the two variables was only .128. The percentage difference between white and nonwhite reSpondents in the 1980 study was similar to the difference found in 1979. For the white respondents, 50.5 percent indicated some degree of unsafety, while 59.8 percent of the nonwhite respondents reported feeling somewhat or very unsafe. When observing the data for the 1981 study, the percentages for white and nonwhite respondents indicating any degree of unsafety are almost identical. Approximately 53 percent of the white and nonwhite respondents responded they felt somewhat or very unsafe while out alone at night on campus. The class level of the respondent was crosstabulated with the perception of safety to determine if freshman or first term students had a different perception of safety than sophomores, juniors or seniors. The significant levels greater than .1 indicates that a sta- tistically significant association is not present between the two 96 Table 4.6.-~Perception of Safety at Night by Race of Respondent (1979, 1980, 1981) Perception of Safety Survey Year by Race 1979 1980 1981 Very Safe White 9.5% 11.0% 9.7% Nonwhite 6.3 7.2 4.1 Reasonably Safe White 32.5 38.5 37.1 Nonwhite 24.2 33.0 43.3 Somewhat Unsafe White 33.8 31.4 35.7 Nonwhite 36.8 34.0 36.1 Very Unsafe White 24.2 19.1 17.5 Nonwhite 32.6 25.8 16.5 TOTALS White 100.0% 100.0% 100.0% Nonwhite 99.9% 100.0% 100.0% Number of Responses White 1,712 1,769 1,409 Nonwhite 95 97 97 1979: Chi-Square = 5.69 Gamma = .19 Significance = .128 1980: Chi-Square = 4.23 Gamma = .16 Significance = .238 1981: Chi-Square = 3.97 Gamma = .02 Significance = .265 97 variables. The percentage differences between the two subgroups for the three studies were very small. In the 1980 study, for example, 52.5 percent of the freshmen felt somewhat unsafe or very unsafe, while 50.0 percent of the sophomore-seniors felt unsafe (see Table 4.7). In addition to sex, race and class level, an analysis was con- ducted to determine if the respondent's residence location on campus was associated with the fear of crime or perception of safety. As stated in Chapter III, the residence halls were categorized into five complexes or communities based primarily upon geographical location. To support McPherson's view that crime and the fear of crime should be studied at the neighborhood level, the residence complexes were crosstabulated with the perception of safety.2 If a significant rela- tionship existed between the type of residence complex and the percep- tion of safety, then further research would have to be conducted to determine the specific determinants of fear that would be associated with each complex. At the time of these studies, several residence halls on campus were designated for only females, and one residence hall was designated for only males. To offset any bias that this may have on the analysis, the crosstabulation of residence hall complex and perception of safety was controlled by the variable sex. In Table 4.8, the male respondents' perception of safety by residence complex is presented. Following in Table 4.9 is the perception of safety by residence complex for females. The statistics at the bottom of Table 4.8 do not support any suspicions that the fear of crime varies significantly between complexes for the male respondents. Although the 1979 study yielded a significance level of .094, the gamma of .025 is extremely weak. 98 Table 4.7.--Perception of Safety at Night by Class Level of Respondent (1979, 1980, 1981) Survey Year Perception of Safety by Class Level 1979 1980 1981 Very Safe Freshmen 8.3% 9.5% 9.5% Sophomore-Senior 9.9 11.6 9.4 Reasonably Safe Freshmen 30.8 38.0 35.7 Sophomore-Senior 32.9 38.4 38.3 Somewhat Unsafe Freshmen 33.8 31.7 39.1 Sophomore-Senior 34.1 31.5 33.8 Very Unsafe Freshmen 27.1 20.8 15.7 Sophomore-Senior 23.1 18.5 18.5 TOTALS Freshmen 100.0% 100.0% 100.0% Sophomore-Senior 100.0% 100.0% 100.0% Number of Responses Freshmen 663 750 560 Sophomore-Senior 1,147 1,116 950 1979: Chi-Square = 4.356 Gamma = -.O75 Significance = .226 1980: Chi-Square = 3.284 Gamma = -.058 Significance = .349 1981: Chi-Square = 5.020 Gamma = -.000 Significance = .170 99 Table 4.8.--Perception of Safety at Night by Residence Complex for Males (1979, 1980, 1981) Perception of Residence Complex Safety for West Red Males Brody Circle Cedar East South Very Safe 1979 16.6% 22.0% 27.3% 23.1% 14.7% 1980 26.9 30.4 20.8 25.3 17.5 1981 18.5 27.8 16.9 17.9 26.5 Reasonably Safe 1979 59.8 55.9 49.2 60.0 63.7 1980 60.2 58.9 64.8 61.9 63.7 1981 65.7 55.6 68.5 67.3 56.8 Somewhat Unsafe 1979 21.3 18.6 20.3 14.7 25.0 1980 11.7 10.7 13.0 10.5 14.0 1981 13.9 13.9 12.9 13.5 16.1 Very Unsafe 1979 2.4 3.4 3. 2. 1. 1980 1.2 0.0 2.3 2. 4. 1981 1. 2.8 1 1. TOTALS 1979 100.1% 99.9% 99.9% 100.0% 100.0% 1980 100.0% 100.0% 99.9% 99.9% 99.9% 1981 100.0% 100.1% 99.9% 100.0% 100.0% Number of Responses 1979 169 59 128 225 204 1980 171 56 162 257 171 1981 108 36 124 156 155 1979: Chi-Square = 1.877 Gamma = .025 Significance = .094 1980: Chi-Square = 13.840 Gamma = .099 Significance = .311 1981: Chi-Square = 9.370 Gamma = -.O48 Significance = .671 100 Table 4.9.--Perception of Safety at Night by Residence Complex for Females (1979, 1980, 1981) Residence Complex Perception of Safety for West Red Females Brody Circle Cedar East South Very Safe 1979 .7% 1.0% 1.5% .8% .5% 1980 0.0 1.0 .5 1.2 .4 1981 2.8 2.9 1.9 1.8 1.0 Reasonably Safe 1979 17.1 9.9 12.3 11.4 13.7 1980 19.3 18.5 20.8 21.3 17.8 1981 19.9 21.1 17.3 26.3 13.3 Somewhat Unsafe 1979 41.4 54.7 41.7 45.8 39.2 1980 44.4 49.3 42.1 49.0 48.5 1981 48.2 57.1 53.1 43.8 51.5 Very Unsafe 1979 40.8 34.4 44.6 42.0 46.7 1980 36.3 31.2 36.6 28.5 33.2 1981 29.1 18.9 27.8 28.1 34.2 TOTALS 1979 99.9% 100.0% 100.1% 100.0% 100.1% 1980 100.0% 100.0% 100.0% 100.0% 99.9% 1981 100.0% 100.0% 100.1% 100.0% 100.0% Number of Responses 1979 152 192 204 264 212 1980 171 205 183 249 241 1981 141 175 162 224 196 1979: Chi-Square = 16.27 Gamma = .065 Significance = .179 1980: Chi-Square = 8.51 Gamma = -.020 Significance = .744 = 23.84 Gamma = .080 Significance = .020 1981: Chi-Square 101 The general observations in Table 4.9 for the female respondents are similar to the male responses in Table 4.8 in that residence complex is not associated with the perception of safety. For the female respon- dents, the greatest differences were found in the 1981 study. The per- centage of female respondents indicating any degree of unsafety ranged from 71.9 percent in the East complex to 87.3 percent in the Brody complex. Data presented in Table 4.10 was obtained by asking the respon- dents if they had been victims of various crimes on campus. The objec- tive was to identify respondents who perceived themselves as personal crime victims, pr0perty crime victims or nonvictims. For informational purposes, the percentage of nonvictims in the three studies ranged from 84.8 percent in 1980 to 87.0 percent in 1979. As expected, the percent- age of personal crime victims was relatively small. The 1980 study had 1.9 percent of the respondents classified as personal crime victims, whereas 1.5 percent and 1.6 percent of the total respondents were per- sonal crime victims in the 1979 and 1981 studies, respectively. As with the nonvictim and personal crime victim categories, it can be seen in Table 4.10 that the percentage range for the property crime victims was also narrow. Approximately 10 percent of the 1981 return sample was classified as property crime victims with a slightly higher percent- age of 13.3 percent in 1980. In Table 4.11, the crosstabulation of the victimization experi- ence with the perception of safety is presented. The variables within the 1979 and 1980 studies are significantly associated with significance levels of .0107 and .0304, respectively. However, the importance of the associations is diminished due to the small number of personal crime 102 victims. Combining the three studies, there were only 70 personal crime victims out of the total 5,293 respondents. Table 4.10.--Respondent's Type of Criminal Victimization Experience (1979, 1980, 1981) Survey Year Victimization Experience 1979 1980 1981 No Victimization Experience (Nonvictim) 87.0% 84.8% 86.2% Personal Crime Victim 1.5 1.9 1.6 Property Crime Victim 11.5 13.3 10.1 TOTALS 100.0% 100.0% 99.9% Number of Responses 1,850 1,892 1,551 When comparing the pr0perty crime victims with the nonvictims, it can be seen in Table 4.11 that a higher percentage of nonvictims indicated some degree of unsafety than did actual property crime vic- tims. In the 1979 study, for example, 59.3 percent of the nonvictims responded somewhat unsafe or very unsafe, compared to 49.8 percent of the property crime victims. With the vast majority of the sample con- sisting of nonvictims and property crime victims, it is easy to see how the direct victimization determinant generally affects a small pro- portion of the p0pulation. In the 1981 study, respondents were asked to classify their home town as urban, suburban or rural. The purpose of collecting this infor- mation was to determine if the respondent's background or home town influences are associated with the perception of safety on campus. By controlling for sex of the respondent, a significant association was 103 Table 4.11.--Perception of Safety at Night by Victimization Experience (1979, 1980, 1981) Perception of Safety by Survey Year Victimization Experience 1979 1980 1981 Very Safe Personal Crime 17.4% 4.5% 20.0% Property Crime 11.8 16.5 10.3 Nonvictim 9.0 9. 9. Reasonably Safe Personal Crime 21.7 27.3 28.0 Pr0perty Crime 38.4 35.0 44.2 Nonvictim 31.7 38.9 36.8 Somewhat Unsafe Personal Crime 39.1 36.4 28.0 Property Crime 22.2 27.8 31.4 Nonvictim 35.2 32.0 36.5 Very Unsafe Personal Crime 21.7 31.8 24.0 Property Crime 27.6 20.7 14.1 Nonvictim 24.1 19.3 17.7 TOTALS Personal Crime 99.9% 100.0% 100.0% Property Crime 100.0% 100.0% 100.0% Nonvictim 100.0% 100.1% 100.1% Number of Responses Personal Crime 23 22 25 Property Crime 203 237 156 Nonvictim 1,611 1,631 1,337 1979: Chi-Square = 16.633 Gamma = .072 Significance = .0107 1980: Chi-Square = 13.932 Gamma = .027 Significance = .0304 1981: Chi-Square = 9.057 Gamma = .113 Significance = .1704 104 found for the male respondents. In Table 4.12, the responses for the female respondents are presented. A significance level of .2347 indi- cates that an association is not present between female respondents' home town size and their perception of safety. As can be seen in Table 4.12, 79.6 percent of the rural reSpondents felt somewhat or very unsafe compared to 73.3 percent of the urban respondents. Table 4.12.--Perception of Safety at Night by Size of Home Town for Females (1981) Size of Home Town Perception of Safety Urban Suburban Rural Very Safe 1.4% 2.0% 1.8% Reasonably Safe 25.4 19.0 18.7 Somewhat Unsafe 44.4 53.1 46.7 Very Unsafe 28.9 25.8 32.9 TOTALS 100.1% 99.9% 100.1% Number of ReSponses 142 546 225 Chi-Square = 8.046 Gamma = .074 Significance = .2347 Table 4.13 presents information concerning the male respondents' perception of safety, crosstabulated with their perceived size of their home town. The variables are statistically associated with a signifi- cance level of .0547. However, the strength of the association is very weak, with a gamma of -.054. Combining the somewhat unsafe and very unsafe reSponses, 21.2 percent of the urban male respondents indicated these responses, compared to 12.5 percent of the suburban male respon- dents and 17.3 percent of the rural male respondents. 105 Table 4.13.--Perception of Safety at Night by Size of Home Town For Males (1981) Size of Hbme Town Perception of Safety Urban Suburban Rural Very Safe 13.7% 22.2% 21.6% Reasonably Safe 65.0 65.2 61.1 Somewhat Unsafe 21.2 10.5 16.7 Very Unsafe 0.0 2.0 .6 TOTALS 99.9% 99.9% 100.0% Number of Responses 80 351 162 Chi-Square = 12.346 Gamma = -.054 Significance = .0547 The final analysis conducted in this section concerns the respon- dents' perception of safety at night while out alone in their home town, versus their perception of safety at night on the MSU campus. The res- pondents' home town perception of safety was crosstabulated with the perception of safety on campus, while controlling for the variable sex, to indicate that both male and female reSpondents who feel unsafe in their home town have a tendency to feel unsafe on campus. However, the data also indicates that a higher percentage of respondents feel more unsafe on campus than in their home town. Table 4.14 presents the res- ponses from the female reSpondents. The female respondents' perception of safety in their home townis.apparentlyrelated to their perception of safety on campus. The chi-square of 85.413 (df = 9) was found to be significant at less than a .0001 level, as indicated in Table 4.14. A gamma of .314 indicates a moderately strong association. In terms of 106 the percentages found in the table, it can be seen that a very high per- centage of female respondents who feel any degree of unsafety in their home town also feel unsafe on campus. Approximately 87 percent of the female respondents who felt somewhat unsafe or very unsafe at night in their home town also felt a comparable degree of unsafety while out alone at night on campus. Table 4.14.--Perception of Safety at Night on Campus by Perception of Safety at Night in their Home Town for Females (1981) Perception of Safety in Home wan Perception of Safety Very Reasonably Somewhat Very on Campus Safe Safe Unsafe Unsafe Very Safe 5.7% .4% 2.0% 1.6% Reasonably Safe 28.9 21.9 10.8 11.3 Somewhat Unsafe 43.8 55.7 50.5 27.4 Very Unsafe 21.6 21.9 36.8 59.7 TOTALS 100.0% 99.9% 100.1% 100.0% Number of ReSponses 194 447 204 62 Chi-Square = 85.413 Gamma = .314 Significance = <.0001 The most interesting finding is the large percentage of female respondents who felt very safe or reasonably safe in their home town, but felt somewhat unsafe or very unsafe on campus. Of the 194 female reSpondents who felt very safe in their home town, 127 or 65.4 percent felt somewhat unsafe or very unsafe on campus. In addition, 77.6 per- cent of the female respondents who felt reasonably safe in their home town indicated they felt somewhat or very unsafe on campus. this gives an indication that a higher percentage of the female Overall, 107 respondents have a feeling of unsafety on campus than in their home town. Although it is not shown in Table 4.14, 78.0 percent of the total female respondents felt somewhat or very unsafe on campus, whereas 29.3 percent of the total female respondents felt somewhat or very unsafe in their home town. Table 4.15 presents the data for the male respondents' perception of safety in their home town, versus on campus. The variables are asso- ciated with a significance level of less than .0001 and a strong associ- ation is indicated by a gamma of .437. Overall, 55 out of 586 (9.4 per- cent) of the total male reSpondents indicated any degree of unsafety while out alone at night in their home town. This is compared to 15 percent of the male respondents who felt somewhat or very unsafe on cam- pus at night. Table 4.15.--Perception of Safety at Night on Campus by Perception of Safety at Night in their Home Town for Males (1981) Perception of Safety in Home Town Perception of Safety Very Reasonably Somewhat Very on Campus Safe Safe Unsafe Unsafe Very Safe 33.7% 10.2% 4.0% 40.0% Reasonably Safe 54.7 73.7 66.0 40.0 Somewhat Unsafe 10.1 14.9 28.0 20.0 Very Unsafe 1.4 1.2 2.0 0.0 TOTALS 99.9% 100.0% 100.0% 100.0% Number of Responses 276 255 50 5 Chi-Square = 61.731 Gamma = .437 Significance = (.0001 108 What Are the Determinants of the Fear of Crime? This section consists of examining several known determinants of the fear of crime. From the review of literature, eight major deter- minants were identified, but the importance or relative strength of each determinant was not substantiated. The purpose of this section is not necessarily to identify the determinants, but to ascertain which determinants have the greatest impact on the fear of crime. As with the previous analyses, the perception of safety while out alone on the MSU campus at night was substituted for the term fear of crime. In all three studies, the respondents were asked what was their perception of safety while out alone on the MSU campus during the day and at night. The responses to these questions were provided in pre- vious sections of this chapter. The respondents were also asked two questions concerning the reasons why they felt unsafe. In the first question, a list of various reasons why an individual may feel unsafe was presented. Respondents were asked to indicate which reasons or determinants applied to their perception of unsafety. For analysis purposes, the respondents were asked to indicate what was the most important reason for feeling unsafe. While it is recognized that an individual may feel unsafe because of more than one reason, the identi- fication of the most important reasons provides a better understanding of which reasons or determinants have the greatest strength or impact on the dependent variable. In Table 4.16, a list of eight possible determinants or reasons why students feel unsafe on campus is pre- sented.3 The responses in this table are for female respondents, whereas the male responses are presented in Table 4.17. From Tables 4.16 and 4.17, it can be seen that the most often 109 Table 4.16.--Most Important Reason for Feeling Somewhat Unsafe or Very Unsafe While Out Alone on MSU Campus for Female Respondents (1979, 1980, 1981) Survey Year Determinants or Reasons for Feeling Unsafe 1979 1980 1981 Direct Victimization 1.6% 2.1% 2.8% Indirect Victimization 3.3 .5 5.2 Media Effect 16.7 19.9 20.5 Psychological 4.5 4.3 5.2 Vulnerability 9.6 11.7 9.3 Inadequate Formal Control 14.5 10.2 6.8 Environment 44.0 33.0 39.1 Perception of Crime 5.9 14.3 11.0 TOTALS 100.1% 100.0% 99.9% Number of Responses 940 931 782 Table 4.17.--Most Important Reason for Feeling Somewhat Unsafe or Very Unsafe While Out Alone on MSU Campus for Male Respondents (1979, 1980, 1981) Survey Year Determinants or Reasons for Feeling Unsafe 1979 1980 1981 Direct Victimization 2.6% 1.3% 2.5% Indirect Victimization 1.9 4.4 5.4 Media Effect 18.3 28.9 31.7 Psychological 2.6 2.8 2.0 Vulnerability 1.9 1.9 3.0 Inadequate Formal Control 25.9 20.4 14.4 Environment 38.9 30.2 34.7 Perception of Crime 7.9 10.1 6.4 TOTALS 100.0% 100.0% 99.9% Number of Responses 378 318 202 110 cited reasons for feeling unsafe on the MSU campus were determinants labeled as environment, inadequate formal control, media effect and perception of crime. For male and female respondents, environmental characteristics such as excessive shrubbery and bushes and limited lighting were considered the most important reason for feeling unsafe in each of the three studies. The media effect was also a common deter- minant among male and female reSpondents. As high as 31.7 percent of the male respondents and 20.5 percent of the female respondents indi- cated they felt unsafe as a result of reading or hearing about local crime in the news media. The perception of crime was concerned with the reSpondents' belief that 'a lot' of crime occurs on the MSU campus. In the 1980 study, 10.1 percent of the male respondents and 14.3 percent of the female respondents indicated this response. In addition, a sub- stantial percentage of respondents indicated they felt unsafe because of the lack of adequate law enforcement protection. This determinant was labeled as inadequate formal control. In the 1979 study, 14.5 per- cent of the female respondents and 25.9 percent of the male respondents indicated this response. While these four determinants were most often indicated by male and female reSpondents alike, the determinant labeled as vulnerability is considered to be influential among the female respondents. Female respondents indicated considerably more often than male respondents that they felt unsafe because of their inability to defend themselves against a criminal attack. Finally, the determinants labeled as direct victimization, indirect victimization and psychological are not con- sidered to be highly explanatory. Relatively few male and female res- pondents responded to these determinants as being the most important 111 reasons for feeling somewhat unsafe or very unsafe while out alone on the MSU campus. In the following paragraphs, further discussion of the findings in Tables 4.16 and 4.17 are presented. Findings to the additional research questions that were reviewed in Chapter III will also be presented. As can be seen in Table 4.16, female respondents cited most often the environment determinant as their most important reason for feeling unsafe. The statement briefly describing the environment determinant consisted of two factors: poor lighting and excessive shrubbery and bushes. The percentage of responses indicating the environment deter- minant ranged from 33.0 percent in 1980 to 44.0 percent in 1979. For the male respondents, data in Table 4.17 indicates that this particular determinant was also the most often cited. In the 1980 study, 30.2 percent of the male responses were attributed to the environment determinant, whereas in the 1979 study, the percentage increased to 38.9 percent. In the 1980 study, respondents who felt somewhat unsafe or very unsafe while out alone on campus were provided a list of ten areas or locations on campus. For each particular location, the respondents were asked to indicate if they would favor improvement in pedestrian lighting and removal or modification of the landscape. The ten loca- tions are listed in Table 4.18. For this analysis, respondents who indicated the environment as their most important reason for feeling unsafe were selected. The percentages presented in Table 4.18 repre- sent the percentage of respondents (n==407) who indicated an affirmative response to either improvement in lighting or modification of landscape 112 for each particular location or area. Several observations can be obtained from this data. Overall, it appears-that the respondents pre- fer an improvement in lighting instead of a modification of the campus landscape. For each location, a higher percentage of respondents indi— cated improvements in lighting than modification of landscape. The percentage of respondents who indicated improvements in lighting ranged from a low of 14.7 percent in service areas to a high of 74.2 percent along the river walks. In comparison, the percentage of respondents who indicated modification of campus landscape ranged from a low of 5.7 percent in service areas to a high of 25.3 percent along the river walks. Table 4.18.--Modification of Pedestria? Lighting and Campus Landscape 1980 Percentage of Respondents (n = 407) Favoring Improved Lighting and Modification of Landscape Improved Modification Location Lighting of Landscape River Walks 74.2% 25.3% Parking Lots 53.3 17.9 Residence Halls 49.1 18.9 Main Library 40.8 11.5 West Circle Central Park 39.8 10.6 Stadium 38.1 10.8 Doorways to All Buildings 31.7 17.7 Classroom Buildings 25.8 9.3 Auditorium or Fairchild 22.9 7.1 Service Areas 14.7 5.7 113 The improvement of pedestrian lighting is obviously the preferred recommendation by respondents who felt somewhat unsafe or very unsafe while out alone at night on campus because of factors within the campus environment. In addition to areas along the river walks, improvements in pedestrian lighting in areas around parking lots and residence halls were also highly favored. Approximately 53 percent of the selected respondents favored increased lighting in parking lots, whereas 49 per- cent favored such improvements around residence halls. The effect of the media upon the fear of crime was the second most often cited reason for the female respondents. For the 1980 and 1981 studies, approximately 20 percent of the female reSpondents cited this determinant as most important. This determinant implied that res- pondents felt unsafe while walking alone on campus because of criminal incidents they read about in newspapers or heard on radio and tele- vision. In other words, the respondents' perception of safety on campus was shaped or influenced by accounts of criminal incidents occurring on campus and reported in the media. A higher percentage of male res- pondents indicated this reason for feeling unsafe than female respon- dents. Although 18.3 percent of the male respondents gave this response in the 1979 study, the percentages increased to 28.9 percent and 31.7 percent in the 1980 and 1981 studies, respectively. One particular question from the 1980 study that deals directly with the media asked the respondents whether they thought crime on cam- pus was less (more, about as) serious than the radio, newspapers and TV indicate. In Table 4.19, responses to this question are crosstabu- lated with the perception of safety. Looking at the column percentages, the data Show that respondents who thought that crime was actually more 114 serious or about the same as portrayed in the media tend to feel some- what unsafe or very unsafe than do respondents who perceive crime as being less serious. Approximately 69 percent of the respondents who reported that crime was more serious than what the media portrays also reported feeling somewhat or very unsafe. A lower percentage of respon- dents who perceived crime as being less serious than what the media portrays also felt somewhat unsafe or very unsafe (24.3 percent). A fairly strong association is observed between the variables, with a gamma of .439. Table 4.19.--Perception of Safety at Night by Perceived Seriousness of Crime Relative to What the Media Say (1980) Seriousness of Crime Relative to What Media Say Less About More Perception of Safety Serious the Same Serious Very Safe 19.3% 9.8% 5.8% Reasonably Safe 56.4 40.9 24.6 Somewhat Unsafe 18.7 33.8 35.5 Very Unsafe 5.6 15.4 34.1 TOTALS 100.0% 99.9% 100.0% Number of Responses 321 928 602 1980: Chi-Square = 221.307 Gamma = .439 Significance = (.001 The determinant that has the greatest percentage difference bet- ween male and female respondents is vulnerability. As discussed in Chapters I and II, vulnerability is referred to as the inability to defend yourself against a criminal incident. The review of literature 115 purported that females tend to feel physically vulnerable to crime more often than males. The data obtained in these studies support this find- ing. Combining the three studies, approximately 10 percent of the female respondents indicated they felt unsafe on campus primarily because of their inability to defend themselves. The percentage of male respondents who indicated this response was 1.9 percent in the 1979 and 1980 studies, and 3.0 percent in the 1981 study. It is not surprising that the determinant concerning direct vic- timization is associated with very few respondents. Direct victimiza- tion implies that the respondent feels unsafe on campus because of past criminal incidents that have occurred to them. From the victimization data that was obtained in each study, it was shown in Table 4.10 that a small percentage of respondents were victims of personal crimes on campus during the prior fall terms of each study. This gave an initial indication that direct victimization experience would have little strength as a determinant of the fear of crime. For both male and female respondents, the percentage of respondents who felt unsafe on campus as a result of prior victimization experience was approximately 2 percent. In raw numbers, only 76 out of the total 3,551 respondents who responded to this question indicated direct victimization as the major reason for feeling unsafe on campus. The reference time period for the occurrence of the criminal vic- timizations was limited to the prior fall term for several reasons, as explained in Chapter III. Because of the relatively short reference period (approximately 3.5 months), respondents of the 1980 study were asked if they had gygg been a victim of a personal crime. This obvi- ously lengthened the reference time period, and as indicated in Table 116 4.20, the number of reSpondents who perceived themselves as personal crime victims was 373. This is approximately five times the number of respondents who were classified as personal crime victims in the 1979, 1980 and 1981 studies combined. When crosstabulating this question with the reSpondent's perception of safety, it is interesting to find that a higher percentage of nonvictims felt somewhat unsafe or very unsafe than personal crime victims. As indicated in Table 4.20, 52.8 percent of the nonvictims (no responses) felt somewhat or very unsafe, compared to 44.3 percent of the personal crime victims (yes reSponses). Overall, the data in the table indicate that direct victimization is not a major reason why individuals feel unsafe on campus. Table 4.20.--Respondent's Perception of Safety at Night by Personal Crime Victimization Status (1980) Personal Crime Victim Perception of Safety Yes No Very Safe 15.8% 9.2% Reasonably Safe 39.9 38.0 Somewhat Unsafe 27.1 32.8 Very Unsafe 17.2 20.0 TOTALS 100.0% 100.0% Number of Responses 373 1,502 Chi-Square = 17.111 Gamma = -.1507 Significance = .0007 In the review of literature, it was accepted that indirect vic- timization is generally more prevalent than direct victimization. The chances are greater that an individual is aware of more criminal 117 incidents that have occurred to friends or acquaintances than that have occurred to himself or herself. The indirect or vicarious victimization determinant implies that individuals feel unsafe while out alone on cam- pus as a result of knowing friends or acquaintances that have been vic- timized. As seen in Tables 4.16 and 4.17, relatively few respondents are associated with this determinant. Overall, the percentage of female respondents feeling unsafe on campus because of victimizations that have occurred to friends or acquaintances ranged from 3.3 percent in 1979 to 5.2 percent in 1981. The percentage of male respondents who indica- ted this response ranged from 1.9 percent in 1979 to 5.4 percent in 1981. In the 1980 study, respondents were asked if they personally knew anyone who has been a victim of a personal crime on campus. The res- ponses to this question were simply yes or no. As can be seen in Table 4.21, 607 respondents indicated in the affirmative to this item. This was approximately 32 percent of the total respondents. Crosstabulating this question with the respondents' perception of safety indicates that only a slightly higher percentage of respondents who personally knew someone who was a victim of a personal crime felt somewhat or very unsafe than respondents who did not know a personal crime victim. Approximately 54 percent of the reSpondents who knew a personal crime victim felt somewhat or very unsafe, whereas 49.7 percent of the reSpon- dents who did not know a personal crime victim felt somewhat unsafe or very unsafe. Overall, this question extended the reference time period, and there is little variation in the fear of crime among respondents who personally knew a personal crime victim and among respondents who did not have contact with other crime victims. 118 ‘Table 4.21.--Respondent's Perception of Safety at Night by Knowing a Personal Crime Victim (1980) Know a Personal Crime Victim Perception of Safety . Yes NO Very Safe 9.9% 11.0% Reasonably Safe 36.1 39.3 Somewhat Unsafe 29.7 32.6 Very Unsafe 24.4 17.1 TOTALS 100.1% 100.0 Number Of Responses 607 1,266 Chi-Square = 14.0558 Gamma = .1033 Significance = .0028 A higher percentage Of male respondents indicated the presence of an inadequate formal control or police department as the most impor- tant reason for feeling unsafe than female respondents. As discussed in the previous chapters, this determinant referred to two perceptions. One perception is that the police department or the Department Of Public Safety is inadequate in providing protection. As indicated in Chapter 111, this particular determinant also included the response that stu- dents feel unsafe because they do not see the police very often when walking on campus. This perception implies that the Department Of Public Safety does not provide protection to students, possibly because Of Operational practices that reduce the visibility of the public safety Officers. Referring back to Table 4.16, the percentage of female res- pondents who felt unsafe because of the lack Of formal control ranged from 6.8 percent in 1981 to 14.5 percent in 1979. Approximately twice 119 the percentage Of male respondents indicated this determinant. The per- centage range was a low of 14.4 percent in 1981 to a high of 25.9 per- cent in 1979. In the review Of literature, a concept referred to as informal control was discussed. This referred to the level of social cohesion that exists in a neighborhood. Theoretically, as social cohesion is increased, the level Of fear Of crime is decreased. In developing a social cohesion index, respondents were classified into three groups: low social cohesion, medium social cohesion and high social cohesion. Presented in Table 4.22 is a crosstabulation Of this variable with the reSpondent's perception of safety at night. The findings indicate that the variables are not statistically associated with each other. How- ever, it should be noted that this test is problematic in that the mea— sures of social cohesion were primarily oriented towards the respon- dent's social cohesion within their residence halls. Further tests should be conducted to determine the reSpondent's perceptions concerning the likelihood Of intervention by other individuals in the event Of a criminal attack on campus. The psychological determinant refers to respondents who feel unsafe because they are 'afraid Of being victimized anywhere.‘ This particular determinant was originally included as a 'catch-all' response to identify respondents who did not have a specific reason for feeling unsafe other than the fact that they feel unsafe in any place or situa- tion. While it would seem logical that these respondents would have additional underlying reasons for feeling unsafe, the scope Of these studies did not permit further exploration of various determinants on an individual basis. As with the determinants concerning indirect and 120 direct victimization, relatively few respondents indicated they felt unsafe because they were afraid of being victimized anywhere. On the average, approximately 4 percent Of the female respondents indicated this determinant as being the most important reason for feeling unsafe while alone on campus, compared to 2.5 percent Of the male respondents. Table 4.22.--Perception Of Safety at Night by Social Cohesion Index Social Cohesion Perception of Safety Low Medium High Very Safe 10.6% 9.7% 13.6% Reasonably Safe 39.1 36.9 41.6 Somewhat Unsafe 32.1 32.0 27.9 Very Unsafe 18.2 21.4 16.9 TOTALS 100.0% 100.1% 100.0% Number of Responses 642 953 308 Chi-Square = 9.651 Gamma = .020 Significance = .1401 The final determinant that is presented in Tables 4.16 and 4.17 is referred to as the 'perception of crime.‘ ReSpondents who indicated this item had the perception that 'a lot of crime occurs on the MSU cam- pus. As discussed in the review of literature, the perception or determination that a lot Of crime occurs on campus may be subjective and not necessarily based upon the Objective crime rate. In addition, this perception may be developed from a variety Of sources. Several of these sources may be related to other determinants such as personal crime victimization experience, awareness of incidents that have 121 occurred to friends or acquaintances or accounts of criminal incidents reported by the media. As Observed in Tables 4.16 and 4.17, a slightly higher percentage of female respondents indicated this was the most important reason for feeling unsafe on campus than male respondents. As high as 14.3 percent of the female respondents in the 1980 study indicated this response, whereas 10.1 percent of the male respondents felt unsafe because they perceived that a lot Of crime occurs on campus. According to Riger, Gordon and LeBailly, women have a further source of fear that men do not have, that of the threat of sexual assault.4 To measure this possible source of fear, female respondents in the 1980 study were asked what their perception of the Chances Of being sexually assaulted on campus were. The four given responses ranged from very low to very high. A crosstabulation of this question with the perception of safety at night is presented in Table 4.23. At the bottom of the table are the chi-square, gamma and significance level, which indicates that a fairly strong association exists between the variables. The association is negative, which implies that females who perceive their chances Of being sexually assaulted on campus as high tend to have an unsafe perception of safety. The percentages in the table should be read as column percentages. Approximately 99 percent Of the female respondents who felt their chances Of being sexually assaulted were very high also felt somewhat unsafe or very unsafe. Fur- thermore, 89.8 percent of the female respondents who responded reason- ably high also reSponded somewhat or very unsafe to the perception of safety. Overall, 59.8 percent Of the female reSpondents perceived their chances Of being sexually assaulted on campus as very high or reason- ably high. Table 4.23.--Perception Of Safety at Night by Perception Of Chances of Sexual Assault on Campus for Females (1980) Perception of Chances of Sexual Assault Very Reasonably Somewhat Very Perception Of Safety High High Low Low Very Safe 0.0% .2% .8% 5.5% Reasonably Safe 1.3 10.1 31.3 67.3 Somewhat Unsafe 24.0 48.6 51.9 23.6 Very Unsafe 74.7 41.2 16.0 3.6 TOTALS 100.0% 100.0% 100.0% 100.0% Number of Responses 75 554 368 55 Chi-Square = 261.76 Gamma = -.633 Significance = (.001 Summary A comparison of the perception Of safety while out alone on cam- pus during the day and at night showed a significant difference in the percentage of reSpondents who felt somewhat unsafe or very unsafe. Com- bining the three surveys, 1.6 percent Of the respondents felt somewhat unsafe or very unsafe while out alone on campus during the day, whereas 54.3 percent Of the total respondents felt a comparable degree of fear at night. Several additional analyses were presented which indirectly mea- sured the magnitude of the problem concerning the fear Of crime. A com- parison of the percentage Of respondents expressing some degree of unsafety for the three campus studies and the eight impact cities was presented. This comparison merely indicated that a sizeable proportion 123 Of the respondents are associated to the fear of crime. ReSpondents of the 1980 study were asked if they have taken any actions or precautions as a result of their perception of unsafety (com- bining somewhat unsafe and very unsafe) on campus. Only 5.8 percent Of the respondents indicated they have not taken any actions, despite their perception Of unsafety. A majority of the responses indicated that respondents who feel somewhat unsafe or very unsafe 'only go out with friends' or 'never or seldom go out at night.‘ Various demographic or personal characteristics were cross- tabulated with the respondents' perception of safety at night. In several instances, variables concerning the respondents' current resi- dence complex, victimization experience and size Of home town were asso- ciated to the dependent variable with significance levels of less than .1. However, these statistics are misleading because Of the positive effect that large sample size has on the chi-square statistic. In each of these instances, the low gammas were stated, which indicated extremely weak associations. The variable sex was crosstabulated with the dependent variable and as expected, a very strong association between the variables was found in each study. In the 1979 study, 86.4 percent of the females felt somewhat unsafe or very unsafe while out alone on campus at night, compared to 22.3 percent Of the males. A moderately strong association was also found between the respondents' perception of safety while on campus and the reSpondents' perception Of safety while in their home town. The associations remained while controlling for the sex Of the reSpondent. The final section involved an analysis Of the determinants. 124 Frequency tables presenting the most important reasons why male and female-respondents feel unsafe on campus indicated that factors within the campus environment were the primary concern. Improvements in pedes- trian lighting were preferred over the removal or modification Of the campus landscape. Of the respondents who felt unsafe on campus due to environment characteristics, over 40 percent favored improved lighting in such locations as the main library, residence halls, parking lots and along the river walks. Two other major reasons why respondents feel unsafe while out alone on campus were labeled as media effect and inadequate formal con- trol. As high as 20.5 percent of the females and 28.9 percent of the males indicated they feel unsafe because of being aware Of criminal incidents that are reported by the media. It was also shown that a lar- ger number Of respondents who perceived crime as being more serious (Or about the same as) than what the media reports also felt somewhat unsafe or very unsafe than did reSpondents who perceived crime as being less serious. Inadequate formal control refers to the respondents' perception that the local police department or Department Of Public Safety does not provide adequate protective services to the students. Respondents were able to indicate whether the police were 'inadequate in providing protection' or they 'did not see the police when walking on campus.‘ In the 1979 study, 14.5 percent of the females and 25.9 percent Of the males indicated this determinant as the most important reason for feel- ing unsafe on campus. Finally, a moderately strong association was found between the female respondents' perceived chances of being sexually assaulted on 125 campus and their perception Of safety on campus. A higher percentage of female respondents who perceived their chances of being sexually assaulted as reasonably high or very high were more likely to feel some- what unsafe or very unsafe on campus at night than the other female respondents. CHAPTER IV FOOTNOTES 1James Garofalo, Public Opinion About Crime: The Attitudes Of Victims and Nonvictims in Selected Cities (National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-I, 1971). 2Marlys McPherson, "Realities and Perceptions of Crime at the Neighborhood Level," Victimology: An International Journal, Vol. 3 1978 . 3As discussed in the Research Design section, Chapter III, the determinants were represented by statements which briefly described each determinant. For purposes Of brevity, the title of each determinant is presented in the tables instead Of the statements. 4Stephanie Riger, Margaret T. Gordon and Robert LeBailly, "Women's Fear of Crime: From Blaming to Restricting the Victim," Victimology: An International Journal, Vol. 3 (1978). 126 CHAPTER V SUMMARY. CONCLUSIONS AND RECOMMENDATIONS Summar This chapter provides a summary Of the three surveys or studies that were recently conducted on the Michigan State University campus. A review of the purpose, research methodology and major findings will be presented in this section. The final sections will consist Of con- clusions and recommendations for future research. Purpose The primary purpose of this research was to provide a descriptive analysis Of a problem concerning the fear Of crime. 'Because much of the research on the fear Of crime has focused on the fear levels and the demographic characteristics Of the fear victims, the determinants Of the fear Of crime have not received adequate attention. In order to start resolving the problem, the determinants must not only be identi- fied, but their relative strengths must be analyzed. This allows policy makers to focus their energies and resources on the determinants that have the greatest impact on resolving the problem. It is assumed that the relative strength Of the determinants may be different from one environment to another. In order to provide usable data for the resolution of the problem, the scope of the research had to be limited to an identifiable environment. Due primarily to accessibility, the Michigan State University campus was chosen as the 127 128 environment. Research Methodology The undergraduate student population residing on the MSU campus was considered to be the sampling frame. TO gather data concerning the fear of crime by MSU undergraduate students, three surveys were con- ducted during January in 1979, 1980, and 1981. The selection of the samples was based upon a diSproportionate stratified random sampling design and for each survey, a sample Of 3,000 undergraduate students was selected. Surveys were distributed to approximately 17 percent Of the sampling frame and the return rates ranged from 51.7 percent in 1981 to 62.3 percent in 1980. The 1979 survey received a return rate Of 61.7 percent. Each survey collected basic information to provide answers for the following research questions. 1. How prevalent is the fear of crime? 2. Who is experiencing the fear of crime? 3. What are the determinants of the fear of crime? The dependent variable consisted Of asking respondents how safe they felt being out alone at night on campus, with the responses being 'very safe,‘ 'reasonably safe,‘ 'somewhat unsafe,’ or 'very unsafe.’ Cross- tabulating the dependent variable with the respondents' demographic or personal characteristics provided information to identify which sub- groups are the most affected by the fear of crime. Finally, frequency tables for male and female respondents were presented which indicated the most important reasons why the respondents felt unsafe on campus. The frequency counts and percentages were simply used to determine the 129 relative strength Of each determinant. A discussion of each determi- nant, along with a few additional analyses, was presented in the last section Of Chapter IV. Major Findings 1. The percentage Of respondents who felt somewhat unsafe or very unsafe while out alone on campus during the day was significantly lower than the percentage of respondents who felt unsafe at night. Combining the three studies, approximately 1.6 percent of the respon- dents felt somewhat unsafe Or very unsafe during the day, compared to 54.3 percent Of the respondents who felt unsafe at night. 2. A strong association (gamma = -.633) was observed between the female respondents' perception of safety at night and their percep- tion of chances of being sexually assaulted. Approximately 91 percent, or 571 out of 629 female respondents, perceived their chances Of being sexually assaulted as very high or reasonably high and also felt some- what unsafe or very unsafe on campus at night. 3. The respondents' sex was associated with the perception of safety at night. The percentage Of females who felt somewhat unsafe or very unsafe ranged from 78.0 percent in 1981 to 86.4 percent in 1979. Relatively few male respondents felt somewhat unsafe or very unsafe. Out of a total 2,204 male respondents who responded to these variables, 382 or 17.3 percent felt any degree of unsafety on campus at night. 4. A moderately strong association was observed between the respondents' perception Of safety in their home town and their percep- tion Of safety on campus. A significance level Of .0000 and a positive gamma of .314 indicates that respondents who feel unsafe (or safe) in 130 their home town also feel unsafe (or safe) while out alone on campus. However, a higher percentage of respondents felt somewhat unsafe or very unsafe on campus than in their home town. For the female respon- dents, 77.9 percent felt somewhat unsafe or very unsafe on campus, com- pared to 29.3 percent who felt unsafe in their home town. 5. The respondents' perception of safety at night did not vary significantly between groups for the variables labeled as race (white, nonwhite), class level (freshman, sophomore-senior), victimization status (personal crime victim, property crime victim, nonvictim), size Of home town (urban, suburban, rural), and residence complex (Brody, West Circle, Red Cedar, East, South). 6. Of the eight determinants presented in Tables 4.16 and 4.17, the media effect, inadequate formal control, environment and perception Of crime determinants were most often indicated by respondents as being the most important reason for feeling unsafe on campus. Media reports of criminal incidents occurring on campus was a major reason why res- pondents felt unsafe. As high as 20.5 percent of the female respondents (1981 study) and 31.7 percent Of the male respondents (1981 study) indi- cated this response. Respondents also perceived the local police department as being inadequate in providing protection. The low visi- bility Of police Officers on campus at night was also included in this response. Factors within the environment, such as poor lighting and excessive shrubbery and bushes enhanced the perception Of unsafety on campus. An increase in pedestrian lighting was preferred by respondents instead of a modification of the campus landscape. Finally, the 1980 study revealed that 14.3 percent of the female respondents and 10.1 percent of the male respondents felt unsafe due to their perception 131 that a lot Of crime occurs on campus. Conclusions A substantial proportion of the undergraduate students residing on the Michigan State University campus expressed some degree of fear Of crime. In each survey, over 50 percent of the sample respondents indicated they felt somewhat unsafe or very unsafe while out alone on campus at night. While the decision to label the findings as a 'com- munity problem' is subjective, the Obviously large percentage of respon- dents who feel unsafe speaks for itself. The need for research at the community or neighborhood level is evident by the data presented. While the campus is not characterized as an urban area, the percentage of respondents indicating a fear of crime was similar to results Observed in the National Crime Surveys. In addition, it was indicated that a higher percentage of respondents felt somewhat unsafe or very unsafe while out alone on campus than in their home town. These findings support the research by McPherson, Conklin and others that the fear of crime is a neighborhood or community problem and, therefore, should be responded to at the neighborhood or community level. Prior research has conclusively shown that females are most likely to express a higher fear Of crime. The findings in this study support this contention. The elderly, lower-income groups, nonwhite racial groups and urban area residents have also been found to express higher levels of fear Of crime by prior research. Data pertaining to these groups was either lacking or not associated to the fear of crime in this study. If it is necessary to focus programs designed to reduce 132 the fear of crime towards a particular group, the female population could be categorically designated as a target population. Finally, there were several reasons explaining why individuals have a fear of crime. Environmental characteristics such as poor light- ing and excessive shrubbery and bushes were cited most Often as the primary reason for feeling unsafe while out alone on campus. In addi- tion, a number Of other determinants labeled as perception of crime, inadequate formal control, media effect and vulnerability were indicated often by the respondents. Although the 'most important' reasons for feeling unsafe were presented, respondents may Obviously have secondary reasons for their fear of crime. Consequently, several determinants may have to be addressed before a significant reduction in the fear Of crime level is evident. Recommendations for Future Research The primary research need with respect to the fear of crime is for in-depth studies concerning the major determinants. This study identified the relative strength or importance of each determinant. However, the causal relationship between each determinant and the fear Of crime levels was not clearly established. In terms Of environmental characteristics, it was evident that a large proportion Of respondents felt unsafe while out alone on campus, due to poor lighting conditions and excessive shrubbery and bushes. More respondents preferred an increase in lighting than a modification or removal of shrubbery and bushes. In addition, changes in the environment were preferred in such areas as the river walks, parking lots, around residence halls and the main library. These general Observations are important, but 133 effectiveness evaluations will have to be conducted to determine the lighting levels and landscape design that reduce the fear of crime. A final report by James Tien, et al. concerning a national evaluation program Of street lighting projects is a good example of this research need. Effectiveness evaluations of police Operational practices should also be conducted. A second major determinant of the fear of crime was the perception that the police were inadequate in providing protec- tion, or there was inadequate police visibility. The effectiveness in reducing crime and the fear of crime should be determined for a var- iety of operational practices such as preventive patrol, footpatrol, neighborhood watch programs, mO-ped patrol or the 'green light phone system.‘ The presumed effects of such practices can be misleading if they cannot be substantiated. The Kansas City preventive patrol experi- ment, San Diego's field interrogation project, and Hartford, Connecti- cut's neighborhood watch programs are examples Of evaluated programs measuring the effects on crime and the fear of crime. Finally, individuals who perceive that 'a lot' of crime is occur- ring in the environment is another concern, especially if the environ- ment has a relatively low crime rate. Additional research should be conducted to determine how individuals develop perceptions concerning the crime rate. Possibly related to this determinant is the effect that the news media has on the fear of crime. Sensationalism concerning criminal events occurring in the local environment may influence an individual's perception of the crime rate. The news media, aided by inaccurate record keeping by the police, may also misinterpret the reported crime rates. Without the use of evaluations, improved 134 Operational practices may appear to increase the reported crime rate while a decrease in the actual crime rate is occurring simultaneously. By merely relying on the reported crime rates and not on other victimi- zation sources such as victimization surveys, the media and the police can portray to the public that the crime rate is increasing when, in fact, the actual amount Of crime is decreasing or remaining the same. APPENDICES APPENDIX A 1979 SURVEY FORMAT CRIMINAL VICTIMIZATION SURVEY OF MSU CAMPUS The following questions refer only to incidents that occurred: 1. During Fall Term 1978 2. On MSU campus, i.e., in or around residence halls, academic building or administrative Offices, upon campus grounds, woodlots or parking lots Please do not include incidents that occurred "Off campus." PERSONAL CHARACTERISTICS 1. Sex 2. Race/Ethnicity 3. Age at last birthday .__M .__White __Indian __F __Black __Japanese __Latin American __All Other __Chinese 4. Class Level 5. Where are you currently 6. Where did you living (state name of live Fall Term __Freshman dorm)? 1978 (state name of dorm)? __SOphomore __Junior __Senior __Grad Student 135 136 In answering Questions I-VII, use the following key for Sections A-E. ANSWER KEY A. Where did the incident occur? (State specific location on campus or name of dorm, building, etc.) B. When did this incident occur, and if unknown, when did you discover the incident? 1. Occurred (please state the time to nearest hour using a.m. or p.m.) 2. Discovered (please state the time to nearest hour using a.m. or p.m. and please indicate with a 'D', e.g., 3 p.m. 'D') C. Did you know the Offender? O. NO I. Yes 0. Did you report this incident to Department of Public Safety? 0. NO 1. Yes E. If no, what are the reasons for not reporting? 1. Police already knew Of the incident 2. Nothing could be done--lack Of proof 3. Did not think it was important enough 4. Did not think police would be effective 5. Did not want to take the time 6. Did not want to get involved 7. Afraid of reprisal 8. Reported to someone else (Resident Hall Advisory Staff) 9. Afraid to deal with police 10. Other (specify) II. III. 137 DURING FALL TERM. DID ANYONE ACTUALLY TAKE SOMETHING DIRECTLY FROM YOU BY USING FORCE SUCH AS A STICKUP, MUGGING OR THREAT? (Check one box) 0. No--Skip to II 1. :ZZYes--How many times? ____ (Refer to key on page 2 for A-E) Incidents #1 #2 A. Where? B. When? C. Did you know Offender? 0. Did you report this to campus police? E. If not, why not? DURING FALL TERM, DID ANYONE TRY TO TAKE SOMETHING DIRECTLY FROM YOU BY USING FORCE OR THREATENING TO HARM YOU? (Check one box) 0. _ NO--Skip to III 1. __ Yes--How many times? ____ (Refer to key on page 2 for A-E) Incidents #1 #2 A. Where? B. When? C. Did you know Offender? 0. Did you report this to campus police? E. If not, why not? DURING FALL TERM, DID ANYONE PHYSICALLY ATTACK YOU, KNOCK YOU DOWN OR BEAT YOU UP? (Check one box) 0. __ No--Skip to IV 1. __ Yes--How many times? (Refer to key on page 2 for A-E) Incidents #1 #2 Where? When? Did you know Offender? Did you report this to campus police? E. If not, why not? DOW) III. IV. 138 (Continued) Incidents #1 #2 F. How many people attacked you? (Enter number of assailants for each incident, if known) G. Did the person(s) have a weapon? (Please state the weapon used if any) H. How did the person attack you? (Enter item number for each incident) 1. Raped (forced to perform or participate in a sexual act) Sexually molested Hit with Object held in hand, shot, knifed Hit by thrown Object Grabbed, held, tripped, jumped, pushed, etc. Other (specify) U145 (JUN 01 DURING FALL TERM, DID ANYONE THREATEN TO BEAT YOU UP OR THREATEN TO ATTACK OR HIT YOU? 0. __ NO--Skip to V 1. __ Yes--How many times? _____ (Refer to key on page 2 for A-E) Incidents #1 #2 A. Where? B. When? C. Did you know Offender? 0. Did you report this to campus police? If no, why not? How were you threatened? (Enter item number for each incident) 1. Verbal threat of rape or other sexual attack 2. Verbal threat Of attack 3. Weapon present or threatened with weapon 4 Attempted attack with weapon (for example, shot at) Object thrown at person Followed 0101 139 (Continued) Incidents #1 #2 7. Surrounded or threatened by group 8. Other (specify) Were you confronted in any way? (Enter item number(s) for each incident) 1. NO--Skip to V 2. Harass, unnecessarily argue or direct abusive language toward you 3. Forcibly enter or attempt to enter your room Forcibly enter or attempt to enter your car Threaten to damage or destroy your property Other (specify) GUI-h DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR DORM ROOM WHILE YOU WERE AWAY? (Check one box) 0. _ NO--Skip to v1 1. __ Yes--How many times? (Refer to key on page 2 for A-E) Incidents #1 #2 #3 A. Where? 8. When? C. Did you know offender? 0. Did you report this to campus police? E. If no, why not? F. What was stolen? (Specify for each incident) G. What was estimated value of stolen property? (Specify for each incident) H. From where in room was property taken? (Specify for each incident) Desk drawer Closet Bookshelf Top of dresser Dresser drawer Other (specify) Chm-bumb- 140 V. (Continued) Incidents #1 #2 #3 I. Was your corridor door: 1. Open 2. Closed J. If closed, was corridor door: 1. Locked 2. Unlocked VI. DURING FALL TERM, WAS ANYTHING (ELSE) STOLEN FROM YOU ON CAMPUS SUCH AS BOOKS, CLOTHING, WALLETS, ETC.? (Check one box) 0. _ No--Skip to VII 1. __ Yes--How many times? ____ (Refer to key on page 2 for A-E) Incidents #1 #2 #3 A. Where? B. When? C. Did you know Offender? 0. Did you report this to campus police? If no, why not? What was stolen? (Specify for each incident) G. What was estimated value Of stolen property? (Specify for each incident) VII. DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR VEHICLE WHILE IT WAS PARKED ON CAMPUS? (Check one box) 0. __ NO--Skip to VIII 1. __ Yes--How many times? (Refer to key on page 2 for A-E) Incidents #1 #2 #3 Where? When? Did you know Offender? Did you report this to campus police? DOW) If no, why not? What was stolen? (Specify for each incident) VII. VIII. IX. 141 (Continued) Incidents #1 #2 #3 G. What was estimated value Of stolen prOperty? (Specify for each incident) HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS DURING THE DAY? (Check one box) 1. __ Very safe 2. __ Reasonably safe 3. __ Somewhat unsafe 4. __ Very unsafe HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS AT NIGHT? (Check one box) 1. __ Very safe 2. __ Reasonably safe 3. __ Somewhat unsafe 4. __ Very unsafe IF You FEEL UNSAFE, WHAT ARE THE REASONS? (Check all that apply Past incident(s) that have occurred to me Because of incident(s) that have happened to friends From what I've read in newspapers, heard on radio, television I'm afraid of being victimized anywhere I feel inadequate to defend myself I feel the police are inadequate in providing protec- tion 7. I feel unsafe because I don't see police on campus very Often when I am walking from building to building 8. There are various places on campus that are poorly lit and have an excess of shrubbery, bushes, etc. that does not provide adequate visibility and a protective atmosphere 9. The general feeling and attitude among students is that alot of crime occurs on MSU campus camc- wNi—a WHAT WOULD YOU SAY IS THE MOST IMPORTANT? (Enter item number) WOULD YOU SAY IN GENERAL, THAT THE CAMPUS POLICE (DEPARTMENT OF PUBLIC SAFETY) IS DOING A: (Check one box) 1. __ Good job 2. __ Average job 3. __ Poor job 4. __ Don't know 142 IX. (Continued) b. COMMENTS: IN WHAT WAYS COULD THEY IMPROVE? (Check all that apply) . __ NO improvement needed . __ Need more police Officers . __ Patrol or investigate more . __ Be more prompt . __ Be more courteous, concerned . __ Don't discriminate . __ Need more traffic control . __ Need more police Officers in certain areas at certain times 9. __ Increase foot patrols 10. __ Increase bike or mo-ped patrols 11. __ Improved training, raise qualifications 12. __ Don't know 13. __ Other (specify) mummwai—a WHAT WOULD YOU SAY IS THE MOST IMPORTANT? (Enter item number) APPENDIX B 1980 SURVEY FORMAT CRIMINAL VICTIMIZATION SURVEY OF MSU CAMPUS The following questions refer only to incidents that occurred: 1. During Fall Term 1979 2. On MSU campus, i.e., in or around residence halls, academic building or administrative offices, upon campus grounds, woodlots or parking lots Please do not include incidents that occurred "Off campus." PERSONAL CHARACTERISTICS 1. Sex 2. Race/Ethnicity 3. Age at last birthday __M __White __Indian __F __Black __Japanese __Latin American __All Other __Chinese 4. Class Level 5. Where are you currently 6. Where did you live living (state name of Fall Term 1979 __Freshman dorm)? (state name Of dorm)? __Sophomore __Junior __Senior __Grad Student 7. Home Town: City County State 143 144 CAMPUS VICTIMIZATION In answering Questions 1-8, please use the following key for Section B and Section E. Answer Key B. At what time did the incident occur? If unknown, when did you discover the incident? (Please indicate with a '0' if discovered, e.g., 3 p.m. '0') During the day (6 a.m. to 6 p.m.) 6 p.m. to midnight Midnight to 6 a.m. Don't know booms—t E. If no, what are the reasons for not reporting? 1. Police already knew of the incident 2. Nothing could be done--lack of proof 3. Did not think it was important enough 4. Did not think police would be effective 5. Did not want to take the time 6. Did not want to get involved 7. Fear Of unsympathetic treatment by police 8. Did not want others to know about incident 9. Afraid Of reprisal 10. Reported to someone else (Resident Hall Advisory Staff) 11. Other (specify) 145 DURING FALL TERM, DID ANYONE ACTUALLY TAKE SOMETHING DIRECTLY FROM YOU BY USING FORCE SUCH AS A STICKUP. MUGGING OR THREAT? (Check one box) 0. __ No--Skip to 2 1. __ Yes--How many times? Incidents #1 #2 Location Of incident? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) DURING FALL TERM, DID ANYONE TRY TO TAKE SOMETHING DIRECTLY FROM YOU BY USING FORCE OR THREATENING TO HARM YOU? (Check one box) 0. __ NO--Skip to 3 1. ;_ Yes--How many times? - Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) DURING FALL TERM. DID ANYONE PHYSICALLY ATTACK YOU, KNOCK YOU DOWN 0R BEAT YOU UP? (Check one box) 0. __ NO--Skip to 4 1. __ Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) 146 3. (Continued) Incidents #1 #2 C. Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) 4. DURING FALL TERM, WERE YOU A VICTIM OF A SEXUAL ASSAULT? (Check one box) 0. __ NO--Skip to 5 1. __ Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) C. Did you know this person? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was the nature of this assault? (Enter item number for each incident) 1. Rape (forced to perform or participate in a sexual act) 2. Sexually molested 3. Other (specify) G. Who did you tell about this incident? (Enter item number for each incident) 1. NO one 2. Friends 3. Medical personnel 4. Parents 5. Crisis line counselor 6. Residence hall advisor 5. DURING FALL TERM, DID ANYONE THREATEN TO BEAT YOU UP OR THREATEN TO ATTACK OR HIT YOU? (Check one box) 0. __ No--Skip to 6 1. __ Yes--How many times? 147 5. (Continued) Incidents #1 #2 Location Of incident? 8. At what time did the incident occur? (See answer key on page 2) C. Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. How were you threatened? (Enter item number for each incident) 1. Verbal threat of sexual attack 2. Verbal threat of attack 3. Threatened with weapon 4. Other (specify) 6. DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR DORM ROOM WHILE YOU WERE AWAY? (Check one box) 0. __ NO--Skip to 7 1. __ Yes--How many times? Incidents #1 #2 #3 Where did it occur? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was stolen? (Specify for each incident) G. What was estimated value of stolen prOperty? (Specify for each incident) H. Was your room door: 1. Open 2. Closed 148 6. (Continued) Incidents #1 #2 #3 I. If closed, was room door: 1. Locked 2. Unlocked 7. DURING FALL TERM, WAS ANYTHING (ELSE) STOLEN FROM YOU ON CAMPUS? (Check one box) 0. __ NO--Skip to 8 1. __ Yes--How many times? Incidents #1 #2 #3 Where did it occur? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was stolen? (Specify for each incident) G. What was estimated value of stolen property? (Specify for each incident) 8. DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR VEHICLE WHILE IT WAS PARKED ON CAMPUS? (Check one box) 0. __ NO--Skip to next section 1. __ Yes--How many times? Incidents #1 #2 #3 Location of incident? B. At what time did the incident occur? (See answer key on page 2) C. Did you know the Offender? 0. Did you report this to campus police? E. If not, why not? (See answer key on page 2) 149 8. (Continued) Incidents #1 #2 #3 F. What was stolen? (Specify for each incident) G. What was estimated value Of stolen property? (Specify for each incident) CAMPUS SAFETY 1. HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS AT NIGHT? 1. Very safe 2. :: Reasonably safe 3. __ Somewhat unsafe 4. __ Very unsafe 2. HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS DURING THE DAY? 1. _ Very safe 2. __ Reasonably safe 3. __ Somewhat unsafe 4. __ Very unsafe (A) H m YOU FEEL UNSAFE, WHAT ARE THE REASONS? (Check all that apply) 1. __ Past incident(s) that have occurred to me 2. __ Because of incident(s) that have happened to friends 3. __ From what I've read in newspapers, heard on radio, television 4. __ I'm afraid Of being victimized anywhere 5. __ I feel inadequate to defend myself 6. __ I feel the police are inadequate in providing protection 7. ___I feel unsafe because I don't see police on campus very Often when I am walking from building to building 8. There are various places on campus that are poorly lit and have an excess of shrubbery, bushes, etc. that do not provide adequate visibility and a protective atmosphere 9. The general feeling and attitude among students is that a lot of crime occurs on MSU campus WHAT WOULD YOU SAY IS THE MOST IMPORTANT REASON? (Enter item number) 4. DO YOU PERSONALLY KNOW ANYONE WHO HAS BEEN A VICTIM OF A PERSONAL CRIME (those referred to in Questions 1-4 of the previous section) ON CAMPUS? 1. __ NO 2. __ Yes 10. 11. 12. 150 HAVE YOU EVER BEEN A VICTIM OF A PERSONAL CRIME? 1. __ NO 2. __ Yes DO YOU FEEL THAT CRIME ON MSU CAMPUS IS MORE OR LESS SERIOUS THAN WHAT IS REPORTED IN THE NEWSPAPER, ON THE RADIO AND TELEVISION? 1. __ Less serious than presented in the media 2. __ About the same seriousness as presented in the media 3. __ More serious than presented in the media 00 YOU FEEL YOUR CHANCES OF BEING SEXUALLY ASSAULTED ON MSU CAMPUS ARE: 1. __ Very high 2. __ Reasonably high 3. __ Somewhat low 4. __ Very low WHAT ACTIONS HAVE YOU TAKEN AS A RESULT OF YOUR PERCEPTION OF CRIME ON CAMPUS? (Check all that apply) __ Nothing . __ Only go out with friends __ Use Dial-A-Ride .__ Never or seldom go out at night __ Carry a weapon .‘__ Take self-defense classes __ Use a campus escort service .___ Other (specify) mNOVU‘l-bOONI—I ON SOME DORMITORY FLOORS PEOPLE DO THINGS TOGETHER AND HELP EACH OTHER; IN OTHERS, PEOPLE MOSTLY GO THEIR OWN WAYS. IN GENERAL, WHAT KIND OF FLOOR WOULD YOU SAY YOU LIVE ON--ONE WHERE PEOPLE HELP EACH OTHER, OR ONE WHERE PEOPLE GO THEIR OWN WAYS? 1. __ Help each other 2. __ Go their own ways HOW MANY OF THOSE PEOPLE THAT LIVE NEAR YOU (on your floor or otherwise close by) DO YOU FEEL YOU KNOW WELL ENOUGH TO ASK A FAVOR OF IF YOU NEEDED SOMETHING--WOULD YOU SAY MOST OF THEM, SOME OF THEM OR ALMOST NONE OF THEM? 1. __ Most 2. __ Some 3. __ Almost none HOW LONG HAVE YOU LIVED IN THIS DORM? a. WOULD YOU SAY IN GENERAL, THAT THE CAMPUS POLICE ARE DOING A: (Check one box) 1. __ Good job 2. __ Average job 3. __ Poor job 4. __ Don't know 12. (Continued) b. 151 IN WHAT WAYS COULD THEY IMPROVE? (Check all that apply) mummwai—I _— Other (specify) No improvement needed Need more police officers Patrol or investigate more Be more prompt Be more courteous, concerned Don't discriminate Need more traffic control Need more police officers in certain areas at certain times Increase foot patrols Increase bike or mo-ped patrols Improved training, raise qualifications Don't know WHAT WOULD YOU SAY IS THE MOST IMPORTANT? (Enter item number) 13. IF YOU FEEL UNSAFE AT NIGHT, WOULD YOU IMPROVE PEDESTRIAN LIGHTING IN THE VICINITY OF: (Check all that apply) mmummprt—a 10. .___ Library . __ Residence halls . __ Classroom buildings __ River walks . __ Stadium . __ West Circle Central Park area . __ Doorways to all buildings . __ Service areas __ Parking areas __ Auditorium to Fairchild 14. IF YOU FEEL UNSAFE AT NIGHT BECAUSE OF CAMPUS PLANTINGS, WOULD YOU FAVOR REMOVAL OR MODIFICATIONS TO THE LANDSCAPE AT: (Check all that apply) . __ Library __ Residence halls . Classroom buildings River walks Stadium West Circle Central Park area . __ Doorways to all buildings __ Service areas . __ Parking areas . __ Auditorium or Fairchild C‘OQVOIU‘I-wav—I ._a THANK YOU FOR YOUR COOPERATION AND PARTICIPATION IN THIS STUDY! COMMENTS: APPENDIX C 1981 SURVEY FORMAT CRIMINAL VICTIMIZATION SURVEY OF MSU CAMPUS The following questions refer only to incidents that occurred: 1. During Fall Term 1980 2. 0n MSU campus, i.e., in or around residence halls, academic building or administrative Offices, upon campus grounds, woodlots or parking lots Please do not include incidents that occurred "Off campus." PERSONAL CHARACTERISTICS 1. Sex 2. Race/Ethnicity 3. Age at last . birthday __M __White __Indian __F __Black __Japanese __Chinese __All Other __Latin American 4. Class Level 5. Where are you currently 6. Where did you live living (state name Of Fall Term 1980 __Freshman dorm)? (state name of dorm)? __Sophomore .__Junior __Senior __Grad Student 7. Home Town: City State County 8. DO you consider your home town to be: __Urban ___Suburban __Rural 152 153 CAMPUS VICTIMIZATION In answering Questions 1-7, please use the following key for Section B and Section E. Answer Key B. At what time did the incident occur? If unknown, when did you discover the incident? (Please indicate with a '0' if discovered, e.g., 3 p.m. '0') During the day (6 a.m. to 6 p.m.) 6 p.m. to midnight Midnight to 6 a.m. Don't know wai-s E. If no, what are the reasons for not reporting? 1. Police already know Of the incident Nothing could be done--lack of proof Did not think it was important enough Did not think police would be effective Did not want to take the time Did not want to get involved Fear Of unsympathetic treatment by police Did not want others to know about incident Afraid Of reprisal Reported to someone else (Resident Hall Advisor) wombwm oooo 154 DURING FALL TERM, DID ANYONE ACTUALLY TAKE SOMETHING FROM YOU BY USING FORCE SUCH AS A STICKUP. MUGGING OR THREAT? (Check no or yes) __No--Skip to 2 .__Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) DURING FALL TERM, DID ANYONE TRY TO TAKE SOMETHING DIRECTLY FROM YOU BY USING FORCE OR THREATENING TO HARM YOU? (Check no or yes) __NO--Skip to 3 __Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) DURING FALL TERM, DID ANYONE PHYSICALLY ATTACK YOU. KNOCK YOU DOWN. OR BEAT YOU UP? (Check no or yes) __NO--Skip to 4 __Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) C. Did you know the Offender? 155 3. (Continued) Incidents #1 #2 0. Did you report this to campus police? E. If not, why not? (See answer key on page 2) 4. DURING FALL TERM. WERE YOU A VICTIM OF A SEXUAL ASSAULT? (Check no or yes) __No--Skip to 5 .__Yes--How many times? Incidents #1 #2 Location of incident? B. At what time did the incident occur? (See answer key on page 2) Did you know this person? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was the nature of this assault? 1. Rape (forced to perform or participate in a sexual act) 2. Sexually molested 3. Other (specify) G. Who did you tell about this incident? 1. NO one 2. Friends 3. Medical personnel 4. Parents 5. Crisis line counselor 5. DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR VEHICLE WHILE IT WAS PARKED ON CAMPUS? (Check no or yes) __NO--Skip to 6 __Yes--How many times? Incidents #1 #2 #3 A. Where did it occur? 156 5. (Continued) Incidents #1 #2 #3 B. At what time did the incident occur? (See answer key on page 2) Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was stolen? (Specify for each incident) G. What was estimated value Of stolen prOperty? (Specify for each incident) 6. DURING FALL TERM, WAS ANYTHING STOLEN FROM YOUR DORM ROOM WHILE YOU WERE AWAY? (Check no or yes) __No--Skip to 7 __Yes--How many times? Incidents #1 #2 #3 Where did it occur? B. At what time did the incident occur? (See answer key on page 2) _ Did you know the Offender? Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was stolen? (Specify for each incident) G. What was estimated value of stolen prOperty? (Specify for each incident) H. Was your room door: 1. Open 2. Closed I. If closed. was room door: 1. Locked 2. Unlocked 157 7. DURING FALL TERM. WAS ANYTHING (ELSE) STOLEN FROM YOU ON CAMPUS? (Check no or yes) __No--Skip to next section .__Yes--How many times? Incidents #1 #2 #3 Where did it occur? B. At what time did the incident occur? (See answer key on page 2) C. Did you know the offender? 0. Did you report this to campus police? E. If not, why not? (See answer key on page 2) F. What was stolen? (Specify for each incident) G. What was estimated value Of stolen prOperty? (Specify . for each incident) PERCEPTION OF SAFETY 1. HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS AT NIGHT? (Check one) 1. __ Very safe 2. __ Reasonably safe 3. __ Somewhat unsafe 4. ___Very unsafe 2. HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE ON THE MSU CAMPUS DURING THE DAY? (Check one) 1. __ Very safe 2. __ Reasonably safe 3. __ Somewhat unsafe 4. __ Very unsafe 3. IF YOU FEEL UNSAFE, WHAT ARE THE REASONS? (Check all that apply) 1. __ Past incident(s) that have occurred to me 2. __ Because of incident(s) that have happened to friends 3. __ From what I've read in newspapers, heard on radio, television 4. __ I'm afraid of being victimized anywhere 5. __ I feel inadequate to defend myself 6. I feel the police are inadequate in providing protection 158 (Continued) 7. __ I feel unsafe because I don't see police on campus very Often when I am walking from building to building 8. __ There are various places on campus that are poorly lit and have an excess of shrubbery, bushes, etc. that do not provide adequate visibility and a protective atmosphere 9. __ The general feeling and attitude among students is that a lot Of crime occurs on MSU campus WHAT WOULD YOU SAY IS THE MOST IMPORTANT REASON? (Enter item number) WHEN YOU ARE IN YOUR HOMETOWN (Neighborhood), HOW SAFE DO YOU FEEL IT IS TO BE OUT ALONE AT NIGHT? (Check one) . __ Very safe . __ Reasonably safe . __ Somewhat unsafe . __ Very unsafe wai—I DO YOU FEEL THAT YOU ARE REALLY A PART OF THE GROUP OR 'FLOOR' IN WHICH YOU LIVE? (Check one) 1. __ Really a part of my living group 2 __ Included in most ways 3. __ Included in some ways, but not in others 4. __ Don't feel I really belong in this group IF YOU HAD A CHANCE TO LIVE IN A SIMILAR BUILDING WITH ANOTHER GROUP OR FLOOR, HOW WOULD YOU FEEL ABOUT MOVING? (Check one) __ Would want to very much .'__ Would rather move than stay where I am . __ Would rather stay where I am than move . __ Would want very much to stay where I am pump- HOW LONG HAVE YOU LIVED IN THIS DORM? PERCEPTION OF CAMPUS POLICE WOULD YOU SAY IN GENERAL, THAT THE CAMPUS POLICE ARE DOING A: (Check one) 1. __ Good job 2. __ Average job 3. __ Poor job 4. __ Don't know 159 WHICH OF THE FOLLOWING BEST DESCRIBES YOUR IMPRESSION OF THE CAMPUS POLICE? (Check one) THEIR PRIMARY PURPOSE SHOULD BE TO PROTECT STUDENTS AND STUDENT PROPERTY. 1_Strongly agree 2 :Moderately agree 3. _Slightly agree 4. _Slightly disagree 5 __ _Moderately disagree 6 __ Strongly disagree THEY ARE GENERALLY HELPFUL AND COURTEOUS TO STUDENTS. .__ Strongly agree _Moderately agree _Slightly agree _Slightly disagree __ Moderately disagree __ Strongly disagree mm~bWNH THEY ARE ABLE AND WILLING TO ASSIST STUDENTS WHO HAVE BEEN VICTIM- IZED IN SOME CRIME. 1. __ Strongly agree Moderately agree Slightly agree Slightly disagree __ Moderately disagree __ Strongly disagree 0101-th THEY 00 WHAT THEY CAN, BUT THEY CAN'T PROTECT A STUDENT FROM BEING VICTIMIZED. 1_Strongly agree 2 :Moderately agree 3 :Slightly agree 4. Slightly disagree 5 __ Moderately disagree 6 __ Strongly disagree THEY ARE MORE CONCERNED WITH HELPING STUDENTS THAN WITH GIVING OUT TRAFFIC TICKETS. _Strongly agree _Moderately agree :Slightly agree :Slightly disagree :Moderately disagree :Strongly disagree mm-thD-l 160 7. THEY LIKE STUDENTS, GENERALLY. . _ Strongly agree . _ Moderately agree _ Slightly agree . _ Slightly disagree . __ Moderately disagree . __ Strongly disagree mmtho—o COHENTS : BIBLIOGRAPHY BIBLIOGRAPHY Bickman, L; Green, S. K.; Edwards, J.; Shane-Dubow, 5.; Lavrakas, P. J.; North-Walker, N.; and Borkowski, S. "Towards Increasing Citizen Responsibility, Surveillance, and Reporting of Crime." Prepared for NILECJ, Law Enforcement Assistance Administration, Department of Justice. Boggs, Sarah L. "Formal and Informal Crime Control: An Exploratory Study of Urban, Suburban, and Rural Orientations.” The Socio- ‘lggical Quarterly 12 (Summer 1971): 319-327. Clemente, Frank, and Kleiman, Michael 8. "Fear of Crime in the United States: A Multivariate Analysis." Social Forces, Vol. 56 (September-June 1977/1978). Conklin, John E. The Impact of Crime. New York: Macmillan, 1975. Davis, James F. "Crime News in Colorado NewSpapers.“ American Journal of Sociology 57 (November 1952): 325-330. Dillman, Don A. "Increasing Mail Questionnaire Response in Large Samples of the General Public." Public Opinion Quarterly 36 (Summer 1972): 254-257. Dominick, Joseph R. "Crime and Law Enforcement on Prime Time Tele- vision.“ Public Opinion Quarterly 37 (Summer 1973): 241-250. Ellis, Albert. "Questionnaires vs. Interview Method in the Study in Human Lover Relationships.“ American Sociological Review 13 (January 1948): 61-65. Erskine, Hazel. "The Polls: Fear of Violence and Crime." Public Opinion Quarterly, Vol. 38 (Spring 1974). Fox, James Alan. Forecasting Crime Data: An Econometric Analysis. Lexington, MA: 0.0. Heath, 1978. Fox, James Walker. Criminal Victimization AmongStudents at a Selected University. Ph.D. Dissertation, University of Virginia, 1975. Frisbie, Douglas; Fishbine, Glenn; Hintz, Richard; Foelson, Mitchell; and Nutter, Julia Brown. Crime in Minneapolis: Proposals for Prevention. Minneapolis: Minnesota Crime Prevention Center, 1977. 161 162 Furstenberg, Frank F. "Public Reaction to Crime in the Streets." American Scholar, Vol. 40 (1970/1971). Gallup, George H. The Gallup Poll: Public Opinion 1978. Wilmington, DE: Scholarly Resources Inc., 1970. Gallup, George H., and Associates. "Fear of Crime, Victimization Now Common to Many Americans." The Gallup Opinion Index, Vol. 124 (October 1975). Gallup Organization. The Gallup Opinion Index. Princeton, NJ: Gallup Organization Inc., 1981. Garofalo, James. Public Opinion About Crime: The Attitudes of Victims and Nonvictims in Selected Cities. National Criminal Justice Information and Statistics Service Analytic Report SD-VAD-l, 1977. . "Victimization and the Fear of Crime.“ Journal of Research in Crime and Delinquency 16 (1979): 80. Gordon, Margaret T., and Heath, Linda. "The News Business, Crime, and Fear." Reactions to Crime. Edited by Dan A. Lewis. Beverly Hills: Sage Publications, 1981. Henig, Jeffrey, and Maxfield, Michael G. "Reducing Fear of Crime: Strategies for Intervention." Victimology: An International Journal, Vol. 3 (1978). Hindelang, Michael J.; Gottfredson, Michael R.; and Garofalo, James. Victims of Personal Crime. Cambridge, MA: Ballinger, 1978. Hyman, Herbert H. Interviewingpin Social Research. Chicago: Univer- sity of Chicago Press, 1954. Isaac, Stephen, and Michael, William B. ‘flandbook in Research and Evaluation. San Deigo: EDITS, 1971. Jacobs, Jane. .Ihe Death and Life of Great American Cities. New York: Vintage Books, 1961. Jaycox, Victoria H. "The Elderly's Fear of Crime: Rational or Irrational?" Victimology: An International Journal, Vol. 3 1978 . Kim, Young Ja. "The Social Correlate of Perceptions of Neighborhood Crime Problems and Fear of Victimization." Working Paper, Reactions to Crime Project, Center for Urban Affairs, Northwestern University, 1976. Lewis, Dan A., and Maxfield, Michael G. "Fear in the Neighborhoods: An Investigation of the Impact of Crime." Journal of Research in Crime and Delinquency, Vol. 17 (1980). 163 McPherson, Marlys. "Realities and Perceptions of Crime at the Neighbor- hood Level.“ Victimology: An International Journal, Vol. 3 1978 . Metzner, Helen, and Mann, Floyd. "A Limited Comparison of Two Methods of Data Collections: The Fixed Alternative Questionnaire and the Open-Ended Interview.“ American Sociological Review, Vol. 17 (August 1974). . Molumby, Thomas. "Patterns of Crime in a University Housing Project." American Behavioral Scientist, Vol. 20. Beverly Hills: Sage Publications, 1976. Newman, Oscar. Defensible Space: Crime Prevention Through Urban Design. New York: Macmillan, 1972. . Architectural Design for Crime Prevention. New York: New York University, 1973. The President's Commission on Law Enforcement and Administration of Justice. The Challengeof Crime in a Free Society. Washington, DC: United States Government Printing Office, 1967. Rainwater, Lee. "Fear and the House-as-Haven in the Lower Class." Journal of the American Institute of Planners, Vol. 32 (January 1966). Repetto, Thomas A. Residential Crime. Cambridge, MA: Ballinger Publishing Co., 1974. Riger, Stephanie. "On Women." Reactions to Crime. Edited by Dan A. Lewis. Beverly Hills: Sage Publications, 1981. Riger, Stephanie; Gordon, Margaret T.; and LeBailly, Robert. "Women's Fear of Crime: From Blaming to Restricting the Victim." Victimology: An International Journal, Vol. 3 (1978). Roshier, Bob. "The Selection of Crime News by the Press." The Manu- facturer of News. Edited by S. Cohen and J. Young. Beverly Hills: SagePublications, 1973. Rubenstein, Herb; Murray, Charles; Motoyama, Tetsuro; and Rouse, W. V. The Link Between Crime and the Built Environment, Vol. 1. National Institute of Justice: United States Government Printing Office, 1980. Skogan, Wesley G. "Public Policy and the Fear of Crime in Large American Cities." Public Law and Public Policy. Edited by John A. Gardiner. New York: Praeger Publishers, 1977. . "0n Attitudes and Behaviors." Reactions to Crime. Edited by Dan A. Lewis. Beverly Hills: Sage Publications, 1981. 164 Skogan, Wesley G., and Maxfield, Michael G. COping With Crime. Beverly Hills: Sage Publications, 1981. Tien, James M.; O'Donnell, Vincent; Barnett, Arnold; and Mirchandani, Pitu. Street LightingProjects: National Evaluation Program, Phase 1, Final Report. Cambridge, MA: Public Systems Evalua- tion, Inc. 1977. Toseland, Ronald W. “Fear of Crime: Who is Most Vulnerable?" Journal of Criminal Justice, Vol. 10 (1982). IIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIII2IIIIIEIIIIIIIIIIIIIIIIIIIIIIIIIIIIII1IIIIIIIIIIIIIIIII